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,
216 JH/45 Use a (new) separate store pool-pair for DKIM verify working data.
217 Previously the permanent pool was used, so the sore could not be freed.
218 This meant a connection with many messages would use continually-growing
221 JH/46 Use an exponentially-increasing block size when malloc'ing store. Do it
222 per-pool so as not to waste too much space. Previously a constant size
223 was used which resulted in O(n^2) behaviour; now we get O(n log n) making
224 DOS attacks harder. The cost is wasted memory use in the larger blocks.
226 JH/47 Use explicit alloc/free for DNS lookup workspace. This permits using the
227 same space repeatedly, and a smaller process footprint.
229 JH/48 Use a less bogus-looking filename for a temporary used for DH-parameters
230 for GnuTLS. Previously the name started "%s" which, while not a bug,
231 looked as if if might be one.
233 JH/49 Bug 2710: when using SOCKS for additional messages after the first (a
234 "continued connection") make the $proxy_* variables available. Previously
235 the information was not passed across the exec() call for subsequent
236 transport executions. This also mean that the log lines for the
237 messages can show the proxy information.
239 JH/50 Bug 2672: QT elements in log lines, unless disabled, now exclude the
240 receive time. With modern systems the difference is significant.
241 The historical behaviour can be restored by disabling (a new) log_selector
242 "queue_time_exclusive".
244 JH/51 Taint-check ACL line. Previously, only filenames (for out-of-line ACL
245 content) were specifically tested for. Now, also cover epxansions
246 rerulting in acl names and inline ACL content.
248 JH/52 Fix ${ip6norm:} operator. Previously, any trailing line text was dropped,
249 making it unusable in complex expressions.
251 JH/53 Bug 2743: fix immediate-delivery via named queue. Previously this would
252 fail with a taint-check on the spoolfile name, and leave the message
255 HS/01 Enforce absolute PID file path name.
257 HS/02 Handle SIGINT as we handle SIGTERM: terminate the Exim process.
259 PP/01 Add a too-many-bad-recipients guard to the default config's RCPT ACL.
261 PP/02 Bug 2643: Correct TLS DH constants.
262 A missing NUL termination in our code-generation tool had led to some
263 incorrect Diffie-Hellman constants in the Exim source.
264 Reported by kylon94, code-gen tool fix by Simon Arlott.
266 PP/03 Impose security length checks on various command-line options.
267 Fixes CVE-2020-SPRSS reported by Qualys.
269 PP/04 Fix Linux security issue CVE-2020-SLCWD and guard against PATH_MAX
270 better. Reported by Qualys.
272 PP/05 Fix security issue CVE-2020-PFPSN and guard against cmdline invoker
273 providing a particularly obnoxious sender full name.
276 PP/06 Fix CVE-2020-28016 (PFPZA): Heap out-of-bounds write in parse_fix_phrase()
278 PP/07 Refuse to allocate too little memory, block negative/zero allocations.
281 PP/08 Change default for recipients_max from unlimited to 50,000.
283 PP/09 Fix security issue with too many recipients on a message (to remove a
284 known security problem if someone does set recipients_max to unlimited,
285 or if local additions add to the recipient list).
286 Fixes CVE-2020-RCPTL reported by Qualys.
288 PP/10 Fix security issue in SMTP verb option parsing
289 Fixes CVE-2020-EXOPT reported by Qualys.
291 PP/11 Fix security issue in BDAT state confusion.
292 Ensure we reset known-good where we know we need to not be reading BDAT
293 data, as a general case fix, and move the places where we switch to BDAT
294 mode until after various protocol state checks.
295 Fixes CVE-2020-BDATA reported by Qualys.
297 HS/03 Die on "/../" in msglog file names
299 QS/01 Creation of (database) files in $spool_dir: only uid=0 or the uid of
300 the Exim runtime user are allowed to create files.
302 QS/02 PID file creation/deletion: only possible if uid=0 or uid is the Exim
305 QS/03 When reading the output from interpreted forward files we do not
306 pass the pipe between the parent and the interpreting process to
307 executed child processes (if any).
309 QS/04 Always die if requested from internal logging, even is logging is
312 JH/54 DMARC: recent versions of the OpenDMARC library appear to have broken
313 the API; compilation noo longer completes with DMARC support included.
314 This affects 1.4.1-1 on Fedora 33 (1.3.2-3 is functional); and has
315 been reported on other platforms.
317 JH/55 TLS: as server, reject connections with ALPN indicating non-smtp use.
323 JH/01 Avoid costly startup code when not strictly needed. This reduces time
324 for some exim process initialisations. It does mean that the logging
325 of TLS configuration problems is only done for the daemon startup.
327 JH/02 Early-pipelining support code is now included unless disabled in Makefile.
329 JH/03 DKIM verification defaults no long accept sha1 hashes, to conform to
330 RFC 8301. They can still be enabled, using the dkim_verify_hashes main
333 JH/04 Support CHUNKING from an smtp transport using a transport_filter, when
334 DKIM signing is being done. Previously a transport_filter would always
335 disable CHUNKING, falling back to traditional DATA.
337 JH/05 Regard command-line recipients as tainted.
339 JH/06 Bug 340: Remove the daemon pid file on exit, when due to SIGTERM.
341 JH/07 Bug 2489: Fix crash in the "pam" expansion condition. It seems that the
342 PAM library frees one of the arguments given to it, despite the
343 documentation. Therefore a plain malloc must be used.
345 JH/08 Bug 2491: Use tainted buffers for the transport smtp context. Previously
346 on-stack buffers were used, resulting in a taint trap when DSN information
347 copied from a received message was written into the buffer.
349 JH/09 Bug 2493: Harden ARC verify against Outlook, whick has been seen to mix
350 the ordering of its ARC headers. This caused a crash.
352 JH/10 Bug 2492: Use tainted memory for retry record when needed. Previously when
353 a new record was being constructed with information from the peer, a trap
356 JH/11 Bug 2494: Unset the default for dmarc_tld_file. Previously a naiive
357 installation would get error messages from DMARC verify, when it hit the
358 nonexistent file indicated by the default. Distros wanting DMARC enabled
359 should both provide the file and set the option.
360 Also enforce no DMARC verification for command-line sourced messages.
362 JH/12 Fix an uninitialised flag in early-pipelining. Previously connections
363 could, depending on the platform, hang at the STARTTLS response.
365 JH/13 Bug 2498: Reset a counter used for ARC verify before handling another
366 message on a connection. Previously if one message had ARC headers and
367 the following one did not, a crash could result when adding an
368 Authentication-Results: header.
370 JH/14 Bug 2500: Rewind some of the common-coding in string handling between the
371 Exim main code and Exim-related utities. The introduction of taint
372 tracking also did many adjustments to string handling. Since then, eximon
373 frequently terminated with an assert failure.
375 JH/15 When PIPELINING, synch after every hundred or so RCPT commands sent and
376 check for 452 responses. This slightly helps the inefficieny of doing
377 a large alias-expansion into a recipient-limited target. The max_rcpt
378 transport option still applies (and at the current default, will override
379 the new feature). The check is done for either cause of synch, and forces
380 a fast-retry of all 452'd recipients using a new MAIL FROM on the same
381 connection. The new facility is not tunable at this time.
383 JH/16 Fix the variables set by the gsasl authenticator. Previously a pointer to
384 library live data was being used, so the results became garbage. Make
385 copies while it is still usable.
387 JH/17 Logging: when the deliver_time selector ise set, include the DT= field
388 on delivery deferred (==) and failed (**) lines (if a delivery was
389 attemtped). Previously it was only on completion (=>) lines.
391 JH/18 Authentication: the gsasl driver not provides the $authN variables in time
392 for the expansion of the server_scram_iter and server_scram_salt options.
394 WB/01 SPF: DNS lookups for the obsolete SPF RR type done by the libspf2 library
395 are now specifically given a NO_DATA response without hitting the system
396 resolver. The library goes on to do the now-standard TXT lookup.
397 Use of dnsdb lookups is not affected.
399 JH/19 Bug 2507: Modules: on handling a dynamic-module (lookups) open failure,
400 only retrieve the errormessage once. Previously two calls to dlerror()
401 were used, and the second one (for mainlog/paniclog) retrieved null
404 JH/20 Taint checking: disallow use of tainted data for
405 - the appendfile transport file and directory options
406 - the pipe transport command
407 - the autoreply transport file, log and once options
408 - file names used by the redirect router (including filter files)
410 - paths used by single-key lookups
411 Previously this was permitted.
413 JH/21 Bug 2501: Fix init call in the heimdal authenticator. Previously it
414 adjusted the size of a major service buffer; this failed because the
415 buffer was in use at the time. Change to a compile-time increase in the
416 buffer size, when this authenticator is compiled into exim.
418 JH/22 Taint-checking: move to safe-mode taint checking on all platforms. The
419 previous fast-mode was untenable in the face of glibs using mmap to
420 support larger malloc requests.
422 PP/01 Update the openssl_options possible values through OpenSSL 1.1.1c.
423 New values supported, if defined on system where compiled:
424 allow_no_dhe_kex, cryptopro_tlsext_bug, enable_middlebox_compat,
425 no_anti_replay, no_encrypt_then_mac, prioritize_chacha, tlsext_padding
427 JH/23 Performance improvement in the initial phase of a two-pass queue run. By
428 running a limited number of proceses in parallel, a benefit is gained. The
429 amount varies with the platform hardware and load. The use of the option
430 queue_run_in_order means we cannot do this, as ordering becomes
433 JH/24 Bug 2524: fix the cyrus_sasl auth driver gssapi usage. A previous fix
434 had introduced a string-copy (for ensuring NUL-termination) which was not
435 appropriate for that case, which can include embedded NUL bytes in the
436 block of data. Investigation showed the copy to actually be needless, the
437 data being length-specified.
439 JH/25 Fix use of concurrent TLS connections under GnuTLS. When a callout was
440 done during a receiving connection, and both used TLS, global info was
441 used rather than per-connection info for tracking the state of data
442 queued for transmission. This could result in a connection hang.
444 JH/26 Fix use of the SIZE parameter on MAIL commands, on continued connections.
445 Previously, when delivering serveral messages down a single connection
446 only the first would provide a SIZE. This was due to the size information
447 not being properly tracked.
449 JH/27 Bug 2530: When operating in a timezone with sub-minute offset, such as
450 TAI (at 37 seconds currently), pretend to be in UTC for time-related
451 expansion and logging. Previously, spurious values such as a future
452 minute could be seen.
454 JH/28 Bug 2533: Fix expansion of ${tr } item. When called in some situations
455 it could crash from a null-deref. This could also affect the
456 ${addresses: } operator and ${readsock } item.
458 JH/29 Bug 2537: Fix $mime_part_count. When a single connection had a non-mime
459 message following a mime one, the variable was not reset.
461 JH/30 When an pipelined-connect fails at the first response, assume incorrect
462 cached capability (perhaps the peer reneged?) and immediately retry in
465 JH/31 Fix spurious detection of timeout while writing to transport filter.
467 JH/32 Bug 2541: Fix segfault on bad cmdline -f (sender) argument. Previously
468 an attempt to copy the string was made before checking it.
470 JH/33 Fix the dsearch lookup to return an untainted result. Previously the
471 taint of the lookup key was maintained; we now regard the presence in the
472 filesystem as sufficient validation.
474 JH/34 Fix the readsocket expansion to not segfault when an empty "options"
475 argument is supplied.
477 JH/35 The dsearch lookup now requires that the directory is an absolute path.
478 Previously this was not checked, and nonempty relative paths made an
479 access under Exim's current working directory.
481 JH/36 Bug 2554: Fix msg:defer event for the hosts_max_try_hardlimit case.
482 Previously no event was raised.
484 JH/37 Bug 2552: Fix the check on spool space during reception to use the SIZE
485 parameter supplied by the sender MAIL FROM command. Previously it was
486 ignored, and only the check_spool_space option value for the required
489 JH/38 Fix $dkim_key_length. This should, after a DKIM verification, present
490 the size of the signing public-key. Previously it was instead giving
491 the size of the signature hash.
493 JH/39 DKIM verification: the RFC 8301 restriction on sizes of RSA keys is now
494 the default. See the (new) dkim_verify_min_keysizes option.
496 JH/40 Fix a memory-handling bug: when a connection carried multiple messages
497 and an ACL use a lookup for checking either the local_part or domain,
498 stale data could be accessed. Ensure that variable references are
499 dropped between messages.
501 JH/41 Bug 2571: Fix SPA authenticator. Running as a server, an offset supplied
502 by the client was not checked as pointing within response data before
503 being used. A malicious client could thus cause an out-of-bounds read and
504 possibly gain authentication. Fix by adding the check.
506 JH/42 Internationalisation: change the default for downconversion in the smtp
507 transport to be "if needed". Previously it was "as previously set" for
508 the message, which usually meant "if needed" for message-submission but
509 "no" for everything else. However, MTAs have been seen using SMTPUTF8
510 even when the envelope addresses did not need it, resulting in forwarding
511 failures to non-supporting MTAs. A downconvert in such cases will be
512 a no-op on the addresses, merely dropping the use of SMTPUTF8 by the
513 transport. The change does mean that addresses needing conversion will
514 be converted when previously a delivery failure would occur.
516 JH/43 Fix possible long line in DSN. Previously when a very long SMTP error
517 response was received it would be used unchecked in a fail-DSN, violating
518 standards on line-length limits. Truncate if needed.
520 HS/01 Remove parameters of the link to www.open-spf.org. The linked form
521 doesn't work. (Additionally add a new main config option to configure the
528 JH/01 OpenSSL: With debug enabled output keying information sufficient, server
529 side, to decode a TLS 1.3 packet capture.
531 JH/02 OpenSSL: Suppress the sending of (stateful) TLS1.3 session tickets.
532 Previously the default library behaviour applied, sending two, each in
535 JH/03 Debug output for ACL now gives the config file name and line number for
538 JH/04 The default received_header_text now uses the RFC 8314 tls cipher clause.
540 JH/05 DKIM: ensure that dkim_domain elements are lowercased before use.
542 JH/06 Fix buggy handling of autoreply bounce_return_size_limit, and a possible
543 buffer overrun for (non-chunking) other transports.
545 JH/07 GnuTLS: Our use of late (post-handshake) certificate verification, under
546 TLS1.3, means that a server rejecting a client certificate is not visible
547 to the client until the first read of encrypted data (typically the
548 response to EHLO). Add detection for that case and treat it as a failed
549 TLS connection attempt, so that the normal retry-in-clear can work (if
550 suitably configured).
552 JB/01 Bug 2375: fix expansions of 822 addresses having comments in local-part
553 and/or domain. Found and fixed by Jason Betts.
555 JH/08 Add hardening against SRV & TLSA lookups the hit CNAMEs (a nonvalid
556 configuration). If a CNAME target was not a wellformed name pattern, a
559 JH/09 Logging: Fix initial listening-on line for multiple ports for an IP when
560 the OS reports them interleaved with other addresses.
562 JH/10 OpenSSL: Fix aggregation of messages. Previously, when PIPELINING was
563 used both for input and for a verify callout, both encrypted, SMTP
564 responses being sent by the server could be lost. This resulted in
565 dropped connections and sometimes bounces generated by a peer sending
568 JH/11 Harden plaintext authenticator against a badly misconfigured client-send
569 string. Previously it was possible to cause undefined behaviour in a
570 library routine (usually a crash). Found by "zerons".
572 JH/12 Bug 2384: fix "-bP smtp_receive_timeout". Previously it returned no
575 JH/13 Bug 2386: Fix builds with Dane under LibreSSL 2.9.0 onward. Some old
576 API was removed, so update to use the newer ones.
578 JH/14 Bug 1891: Close the log file if receiving a non-smtp message, without
579 any timeout set, is taking a long time. Previously we would hang on to a
580 rotated logfile "forever" if the input was arriving with long gaps
581 (a previous attempt to fix addressed lack, for a long time, of initial
584 HS/01 Bug 2390: Use message_id for tempfile creation to avoid races in a
585 shared (NFS) environment. The length of the tempfile name is now
586 4 + 16 ("hdr.$message_exim_id") which might break on file
587 systems which restrict the file name length to lower values.
590 HS/02 Bug 2390: Use message_id for tempfile creation to avoid races in a
591 shared (NFS) environment.
593 HS/03 Bug 2392: exigrep does case sensitive *option* processing (as it
594 did for all versions <4.90). Notably -M, -m, --invert, -I may be
597 JH/15 Use unsigned when creating bitmasks in macros, to avoid build errors
598 on some platforms for bit 31.
600 JH/16 GnuTLS: rework ciphersuite strings under recent library versions. Thanks
601 to changes apparently associated with TLS1.3 handling some of the APIs
602 previously used were either nonfunctional or inappropriate. Strings
603 like TLS1.3:ECDHE_SECP256R1__RSA_PSS_RSAE_SHA256__AES_256_GCM__AEAD:256
604 and TLS1.2:ECDHE_SECP256R1__RSA_SHA256__AES_128_CBC__SHA256:128 replace
605 the previous TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256 .
606 This affects log line X= elements, the $tls_{in,out}_cipher variables,
607 and the use of specific cipher names in the encrypted= ACL condition.
609 JH/17 OpenSSL: the default openssl_options now disables ssl_v3.
611 JH/18 GnuTLS: fix $tls_out_ocsp under hosts_request_ocsp. Previously the
612 verification result was not updated unless hosts_require_ocsp applied.
614 JH/19 Bug 2398: fix listing of a named-queue. Previously, even with the option
615 queue_list_requires_admin set to false, non-admin users were denied the
618 JH/20 Bug 2389: fix server advertising of usable certificates, under GnuTLS in
619 directory-of-certs mode. Previously they were advertised despite the
622 JH/21 The smtp transport option "hosts_noproxy_tls" is now unset by default.
623 A single TCP connection by a client will now hold a TLS connection open
624 for multiple message deliveries, by default. Previously the default was to
627 JH/22 The smtp transport option "hosts_try_dane" now enables all hosts by
628 default. If built with the facility, DANE will be used. The facility
629 SUPPORT_DANE is now enabled in the prototype build Makefile "EDITME".
631 JH/23 The build default is now for TLS to be included; the SUPPORT_TLS define
632 is replaced with DISABLE_TLS. Either USE_GNUTLS or (the new) USE_OPENSSL
633 must be defined and you must still, unless you define DISABLE_TLS, manage
634 the the include-dir and library-file requirements that go with that
635 choice. Non-TLS builds are still supported.
637 JH/24 Fix duplicated logging of peer name/address, on a transport connection-
640 JH/25 The smtp transport option "hosts_try_fastopen" now enables all hosts by
641 default. If the platform supports and has the facility enabled, it will
642 be requested on all coneections.
644 JH/26 The PIPE_CONNECT facility is promoted from experimental status and is now
645 controlled by the build-time option SUPPORT_PIPE_CONNECT.
647 PP/01 Unbreak heimdal_gssapi, broken in 4.92.
649 JH/27 Bug 2404: Use the main-section configuration option "dsn_from" for
650 success-DSN messages. Previously the From: header was always the default
651 one for these; the option was ignored.
653 JH/28 Fix the timeout on smtp response to apply to the whole response.
654 Previously it was reset for every read, so a teergrubing peer sending
655 single bytes within the time limit could extend the connection for a
656 long time. Credit to Qualsys Security Advisory Team for the discovery.
658 JH/29 Fix DSN Final-Recipient: field. Previously it was the post-routing
659 delivery address, which leaked information of the results of local
660 forwarding. Change to the original envelope recipient address, per
663 JH/30 Bug 2411: Fix DSN generation when RFC 3461 failure notification is
664 requested. Previously not bounce was generated and a log entry of
665 error ignored was made.
667 JH/31 Avoid re-expansion in ${sort } expansion. (CVE-2019-13917)
669 JH/32 Introduce a general tainting mechanism for values read from the input
670 channel, and values derived from them. Refuse to expand any tainted
671 values, to catch one form of exploit.
673 JH/33 Bug 2413: Fix dkim_strict option. Previously the expansion result
674 was unused and the unexpanded text used for the test. Found and
675 fixed by Ruben Jenster.
677 JH/34 Fix crash after TLS shutdown. When the TCP/SMTP channel was left open,
678 an attempt to use a TLS library read routine dereffed a nul pointer,
681 JH/35 Bug 2409: filter out-of-spec chars from callout response before using
682 them in our smtp response.
684 JH/36 Have the general router option retry_use_local_part default to true when
685 any of the restrictive preconditions are set (to anything). Previously it
686 was only for check_local user. The change removes one item of manual
687 configuration which is required for proper retries when a remote router
688 handles a subset of addresses for a domain.
690 JH/37 Appendfile: when evaluating quota use (non-quota_size_regex) take the file
691 link count into consideration.
693 HS/04 Fix handling of very log lines in -H files. If a -<key> <value> line
694 caused the extension of big_buffer, the following lines were ignored.
696 JH/38 Bug 1395: Teach the DNS negative-cache about TTL value from the SOA in
697 accordance with RFC 2308. Previously there was no expiry, so a longlived
698 receive process (eg. due to ACL delays) versus a short SOA value could
701 HS/05 Handle trailing backslash gracefully. (CVE-2019-15846)
703 JH/39 Promote DMARC support to mainline.
705 JH/40 Bug 2452: Add a References: header to DSNs.
707 JH/41 With GnuTLS 3.6.0 (and later) do not attempt to manage Diffie-Hellman
708 parameters. The relevant library call is documented as "Deprecated: This
709 function is unnecessary and discouraged on GnuTLS 3.6.0 or later. Since
710 3.6.0, DH parameters are negotiated following RFC7919."
712 HS/06 Change the default of dnssec_request_domains to "*"
714 JH/42 Bug 2545: Fix CHUNKING for all RCPT commands rejected. Previously we
715 carried on and emitted a BDAT command, even when PIPELINING was not
718 JH/43 Bug 2465: Fix taint-handling in dsearch lookup. Previously a nontainted
719 buffer was used for the filename, resulting in a trap when tainted
720 arguments (eg. $domain) were used.
722 JH/44 With OpenSSL 1.1.1 (onwards) disable renegotiation for TLS1.2 and below;
723 recommended to avoid a possible server-load attack. The feature can be
724 re-enabled via the openssl_options main cofiguration option.
726 JH/45 local_scan API: documented the current smtp_printf() call. This changed
727 for version 4.90 - adding a "more data" boolean to the arguments.
728 Bumped the ABI version number also, this having been missed previously;
729 release versions 4.90 to 4.92.3 inclusive were effectively broken in
730 respect of usage of smtp_printf() by either local_scan code or libraries
731 accessed via the ${dlfunc } expansion item. Both will need coding
732 adjustment for any calls to smtp_printf() to match the new function
733 signature; a FALSE value for the new argument is always safe.
735 JH/46 FreeBSD: fix use of the sendfile() syscall. The shim was not updating
736 the file-offset (which the Linux syscall does, and exim expects); this
737 resulted in an indefinite loop.
739 JH/47 ARC: fix crash in signing, triggered when a configuration error failed
740 to do ARC verification. The Authentication-Results: header line added
741 by the configuration then had no ARC item.
747 JH/01 Remove code calling the customisable local_scan function, unless a new
748 definition "HAVE_LOCAL_SCAN=yes" is present in the Local/Makefile.
750 JH/02 Bug 1007: Avoid doing logging from signal-handlers, as that can result in
751 non-signal-safe functions being used.
753 JH/03 Bug 2269: When presented with a received message having a stupidly large
754 number of DKIM-Signature headers, disable DKIM verification to avoid
755 a resource-consumption attack. The limit is set at twenty.
757 JH/04 Add variables $arc_domains, $arc_oldest_pass for ARC verify. Fix the
758 report of oldest_pass in ${authres } in consequence, and separate out
759 some descriptions of reasons for verification fail.
761 JH/05 Bug 2273: Cutthrough delivery left a window where the received messsage
762 files in the spool were present and unlocked. A queue-runner could spot
763 them, resulting in a duplicate delivery. Fix that by doing the unlock
764 after the unlink. Investigation by Tim Stewart. Take the opportunity to
765 add more error-checking on spoolfile handling while that code is being
768 PP/01 Refuse to open a spool data file (*-D) if it's a symlink.
769 No known attacks, no CVE, this is defensive hardening.
771 JH/06 Bug 2275: The MIME ACL unlocked the received message files early, and
772 a queue-runner could start a delivery while other operations were ongoing.
773 Cutthrough delivery was a common victim, resulting in duplicate delivery.
774 Found and investigated by Tim Stewart. Fix by using the open message data
775 file handle rather than opening another, and not locally closing it (which
776 releases a lock) for that case, while creating the temporary .eml format
777 file for the MIME ACL. Also applies to "regex" and "spam" ACL conditions.
779 JH/07 Bug 177: Make a random-recipient callout success visible in ACL, by setting
780 $sender_verify_failure/$recipient_verify_failure to "random".
782 JH/08 When generating a selfsigned cert, use serial number 1 since zero is not
785 JH/09 Bug 2274: Fix logging of cmdline args when starting in an unlinked cwd.
786 Previously this would segfault.
788 JH/10 Fix ARC signing for case when DKIM signing failed. Previously this would
791 JH/11 Bug 2264: Exim now only follows CNAME chains one step by default. We'd
792 like zero, since the resolver should be doing this for us, But we need one
793 as a CNAME but no MX presence gets the CNAME returned; we need to check
794 that doesn't point to an MX to declare it "no MX returned" rather than
795 "error, loop". A new main option is added so the older capability of
796 following some limited number of chain links is maintained.
798 JH/12 Add client-ip info to non-pass iprev ${authres } lines.
800 JH/13 For receent Openssl versions (1.1 onward) use modern generic protocol
801 methods. These should support TLS 1.3; they arrived with TLS 1.3 and the
802 now-deprecated earlier definitions used only specified the range up to TLS
803 1.2 (in the older-version library docs).
805 JH/14 Bug 2284: Fix DKIM signing for body lines starting with a pair of dots.
807 JH/15 Rework TLS client-side context management. Stop using a global, and
808 explicitly pass a context around. This enables future use of TLS for
809 connections to service-daemons (eg. malware scanning) while a client smtp
810 connection is using TLS; with cutthrough connections this is quite likely.
812 JH/16 Fix ARC verification to do AS checks in reverse order.
814 JH/17 Support a "tls" option on the ${readsocket } expansion item.
816 JH/18 Bug 2287: Fix the protocol name (eg utf8esmtp) for multiple messages
817 using the SMTPUTF8 option on their MAIL FROM commands, in one connection.
818 Previously the "utf8" would be re-prepended for every additional message.
820 JH/19 Reject MAIL FROM commands with SMTPUTF8 when the facility was not advertised.
821 Previously thery were accepted, resulting in issues when attempting to
822 forward messages to a non-supporting MTA.
824 PP/02 Let -n work with printing macros too, not just options.
826 JH/20 Bug 2296: Fix cutthrough for >1 address redirection. Previously only
827 one parent address was copied, and bogus data was used at delivery-logging
828 time. Either a crash (after delivery) or bogus log data could result.
829 Discovery and analysis by Tim Stewart.
831 PP/03 Make ${utf8clean:} expansion operator detect incomplete final character.
832 Previously if the string ended mid-character, we did not insert the
833 promised '?' replacement.
835 PP/04 Documentation: current string operators work on bytes, not codepoints.
837 JH/21 Change as many as possible of the global flags into one-bit bitfields; these
838 should pack well giving a smaller memory footprint so better caching and
839 therefore performance. Group the declarations where this can't be done so
840 that the byte-sized flag variables are not interspersed among pointer
841 variables, giving a better chance of good packing by the compiler.
843 JH/22 Bug 1896: Fix the envelope from for DMARC forensic reports to be possibly
844 non-null, to avoid issues with sites running BATV. Previously reports were
845 sent with an empty envelope sender so looked like bounces.
847 JH/23 Bug 2318: Fix the noerror command within filters. It wasn't working.
848 The ignore_error flag wasn't being returned from the filter subprocess so
849 was not set for later routers. Investigation and fix by Matthias Kurz.
851 JH/24 Bug 2310: Raise a msg:fail:internal event for each undelivered recipient,
852 and a msg:complete for the whole, when a message is manually removed using
853 -Mrm. Developement by Matthias Kurz, hacked on by JH.
855 JH/25 Avoid fixed-size buffers for pathnames in DB access. This required using
856 a "Gnu special" function, asprintf() in the DB utility binary builds; I
857 hope that is portable enough.
859 JH/26 Bug 2311: Fix DANE-TA verification under GnuTLS. Previously it was also
860 requiring a known-CA anchor certificate; make it now rely entirely on the
861 TLSA as an anchor. Checking the name on the leaf cert against the name
862 on the A-record for the host is still done for TA (but not for EE mode).
864 JH/27 Fix logging of proxy address. Previously, a pointless "PRX=[]:0" would be
865 included in delivery lines for non-proxied connections, when compiled with
866 SUPPORT_SOCKS and running with proxy logging enabled.
868 JH/28 Bug 2314: Fire msg:fail:delivery event even when error is being ignored.
869 Developement by Matthias Kurz, tweaked by JH. While in that bit of code,
870 move the existing event to fire before the normal logging of message
871 failure so that custom logging is bracketed by normal logging.
873 JH/29 Bug 2322: A "fail" command in a non-system filter (file) now fires the
874 msg:fail:internal event. Developement by Matthias Kurz.
876 JH/30 Bug 2329: Increase buffer size used for dns lookup from 2k, which was
877 far too small for todays use of crypto signatures stored there. Go all
878 the way to the max DNS message size of 64kB, even though this might be
879 overmuch for IOT constrained device use.
881 JH/31 Fix a bad use of a copy function, which could be used to pointlessly
882 copy a string over itself. The library routine is documented as not
883 supporting overlapping copies, and on MacOS it actually raised a SIGABRT.
885 JH/32 For main options check_spool_space and check_inode_space, where the
886 platform supports 64b integers, support more than the previous 2^31 kB
887 (i.e. more than 2 TB). Accept E, P and T multipliers in addition to
888 the previous G, M, k.
890 JH/33 Bug 2338: Fix the cyrus-sasl authenticator to fill in the
891 $authenticated_fail_id variable on authentication failure. Previously
894 JH/34 Increase RSA keysize of autogen selfsign cert from 1024 to 2048. RHEL 8.0
895 OpenSSL didn't want to use such a weak key. Do for GnuTLS also, and for
896 more-modern GnuTLS move from GNUTLS_SEC_PARAM_LOW to
897 GNUTLS_SEC_PARAM_MEDIUM.
899 JH/35 OpenSSL: fail the handshake when SNI processing hits a problem, server
900 side. Previously we would continue as if no SNI had been received.
902 JH/36 Harden the handling of string-lists. When a list consisted of a sole
903 "<" character, which should be a list-separator specification, we walked
904 off past the nul-terimation.
906 JH/37 Bug 2341: Send "message delayed" warning MDNs (restricted to external
907 causes) even when the retry time is not yet met. Previously they were
908 not, meaning that when (say) an account was over-quota and temp-rejecting,
909 and multiple senders' messages were queued, only one sender would get
910 notified on each configured delay_warning cycle.
912 JH/38 Bug 2351: Log failures to extract envelope addresses from message headers.
914 JH/39 OpenSSL: clear the error stack after an SSL_accept(). With anon-auth
915 cipher-suites, an error can be left on the stack even for a succeeding
916 accept; this results in impossible error messages when a later operation
919 AM/01 Bug 2359: GnuTLS: repeat lowlevel read and write operations while they
920 return error codes indicating retry. Under TLS1.3 this becomes required.
922 JH/40 Fix the feature-cache refresh for EXPERIMENTAL_PIPE_CONNECT. Previously
923 it only wrote the new authenticators, resulting in a lack of tracking of
924 peer changes of ESMTP extensions until the next cache flush.
926 JH/41 Fix the loop reading a message header line to check for integer overflow,
927 and more-often against header_maxsize. Previously a crafted message could
928 induce a crash of the recive process; now the message is cleanly rejected.
930 JH/42 Bug 2366: Fix the behaviour of the dkim_verify_signers option. It had
931 been totally disabled for all of 4.91. Discovery and fix by "Mad Alex".
937 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
938 When redis_servers is set to a list of > 1 element, and the Redis servers
939 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
940 case of MOVED into a DEFER case instead, thus moving the query onto the
941 next server in the list. For a cluster of N elements, all N servers must
942 be defined in redis_servers.
944 GF/02 Catch and remove uninitialized value warning in exiqsumm
945 Check for existence of @ARGV before looking at $ARGV[0]
947 JH/01 Replace the store_release() internal interface with store_newblock(),
948 which internalises the check required to safely use the old one, plus
949 the allocate and data copy operations duplicated in both (!) of the
950 extant use locations.
952 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
953 modifier. This matches the restriction on the commandline.
955 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
956 Previously only the last row was returned.
958 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
959 we assumed that tags in the header were well-formed, and parsed the
960 element content after inspecting only the first char of the tag.
961 Assumptions at that stage could crash the receive process on malformed
964 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
965 While running the DKIM ACL we operate on the Permanent memory pool so that
966 variables created with "set" persist to the DATA ACL. Also (at any time)
967 DNS lookups that fail create cache records using the Permanent pool. But
968 expansions release any allocations made on the current pool - so a dnsdb
969 lookup expansion done in the DKIM ACL releases the memory used for the
970 DNS negative-cache, and bad things result. Solution is to switch to the
971 Main pool for expansions.
972 While we're in that code, add checks on the DNS cache during store_reset,
973 active in the testsuite.
974 Problem spotted, and debugging aided, by Wolfgang Breyha.
976 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
977 When none of the hosts presented to a transport match an already-open
978 connection, close it and proceed with the list. Previously we would
979 queue the message. Spotted by Lena with Yahoo, probably involving
982 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
983 Previously a spurious "250 OK id=" response was appended to the proper
986 JH/08 The "support for" informational output now, which built with Content
987 Scanning support, has a line for the malware scanner interfaces compiled
988 in. Interface can be individually included or not at build time.
990 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
991 by the template makefile "src/EDITME". The "STREAM" support for an older
992 ClamAV interface method is removed.
994 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
995 rows affected is given instead).
997 JH/11 The runtime Berkeley DB library version is now additionally output by
998 "exim -d -bV". Previously only the compile-time version was shown.
1000 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
1001 SMTP connection. Previously, when one had more recipients than the
1002 first, an abortive onward connection was made. Move to full support for
1003 multiple onward connections in sequence, handling cutthrough connection
1004 for all multi-message initiating connections.
1006 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
1007 routers. Previously, a multi-recipient message would fail to match the
1008 onward-connection opened for the first recipient, and cause its closure.
1010 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
1011 a timeout on read on a GnuTLS initiating connection, resulting in the
1012 initiating connection being dropped. This mattered most when the callout
1013 was marked defer_ok. Fix to keep the two timeout-detection methods
1016 JH/15 Relax results from ACL control request to enable cutthrough, in
1017 unsupported situations, from error to silently (except under debug)
1018 ignoring. This covers use with PRDR, frozen messages, queue-only and
1021 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
1023 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
1024 metadata, resulting in a crash in free().
1026 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
1027 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
1028 Broken also in d185889f4, with init system revamp.
1030 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
1031 Previously we abruptly closed the connection after reading a malware-
1032 found indication; now we go on to read the "scan ok" response line,
1035 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
1036 ACL. Previously, a crash would result.
1038 JH/19 Speed up macro lookups during configuration file read, by skipping non-
1039 macro text after a replacement (previously it was only once per line) and
1040 by skipping builtin macros when searching for an uppercase lead character.
1042 JH/20 DANE support moved from Experimental to mainline. The Makefile control
1043 for the build is renamed.
1045 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
1046 was allocated for every new TLS startup, meaning one per message. Fix
1047 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
1049 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
1050 reported the original. Fix to report (as far as possible) the ACL
1051 result replacing the original.
1053 JH/23 Fix memory leak during multi-message connections using STARTTLS under
1054 OpenSSL. Certificate information is loaded for every new TLS startup,
1055 and the resources needed to be freed.
1057 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
1059 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
1062 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
1063 DATA response info to the (existing) per-recipient response info for
1064 the "C=" log element. It can have useful tracking info from the
1065 destination system. Patch from Simon Arlott.
1067 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
1068 length value. Previously this would segfault.
1070 HS/02 Support Avast multiline protoocol, this allows passing flags to
1071 newer versions of the scanner.
1073 JH/28 Ensure that variables possibly set during message acceptance are marked
1074 dead before release of memory in the daemon loop. This stops complaints
1075 about them when the debug_store option is enabled. Discovered specifically
1076 for sender_rate_period, but applies to a whole set of variables.
1077 Do the same for the queue-runner and queue-list loops, for variables set
1078 from spool message files. Do the same for the SMTP per-message loop, for
1079 certain variables indirectly set in ACL operations.
1081 JH/29 Bug 2250: Fix a longstanding bug in heavily-pipelined SMTP input (such
1082 as a multi-recipient message from a mailinglist manager). The coding had
1083 an arbitrary cutoff number of characters while checking for more input;
1084 enforced by writing a NUL into the buffer. This corrupted long / fast
1085 input. The problem was exposed more widely when more pipelineing of SMTP
1086 responses was introduced, and one Exim system was feeding another.
1087 The symptom is log complaints of SMTP syntax error (NUL chars) on the
1088 receiving system, and refused recipients seen by the sending system
1089 (propating to people being dropped from mailing lists).
1090 Discovered and pinpointed by David Carter.
1092 JH/30 The (EXPERIMENTAL_DMARC) variable $dmarc_ar_header is withdrawn, being
1093 replaced by the ${authresults } expansion.
1095 JH/31 Bug 2257: Fix pipe transport to not use a socket-only syscall.
1097 HS/03 Set a handler for SIGTERM and call exit(3) if running as PID 1. This
1098 allows proper process termination in container environments.
1100 JH/32 Bug 2258: Fix spool_wireformat in combination with LMTP transport.
1101 Previously the "final dot" had a newline after it; ensure it is CR,LF.
1103 JH/33 SPF: remove support for the "spf" ACL condition outcome values "err_temp"
1104 and "err_perm", deprecated since 4.83 when the RFC-defined words
1105 "temperror" and "permerror" were introduced.
1107 JH/34 Re-introduce enforcement of no cutthrough delivery on transports having
1108 transport-filters or DKIM-signing. The restriction was lost in the
1109 consolidation of verify-callout and delivery SMTP handling.
1110 Extend the restriction to also cover ARC-signing.
1112 JH/35 Cutthrough: for a final-dot response timeout (and nonunderstood responses)
1113 in defer=pass mode supply a 450 to the initiator. Previously the message
1116 PP/02 DANE: add dane_require_tls_ciphers SMTP Transport option; if unset,
1117 tls_require_ciphers is used as before.
1119 HS/03 Malware Avast: Better match the Avast multiline protocol. Add
1120 "pass_unscanned". Only tmpfails from the scanner are written to
1121 the paniclog, as they may require admin intervention (permission
1122 denied, license issues). Other scanner errors (like decompression
1123 bombs) do not cause a paniclog entry.
1125 JH/36 Fix reinitialisation of DKIM logging variable between messages.
1126 Previously it was possible to log spurious information in receive log
1129 JH/37 Bug 2255: Revert the disable of the OpenSSL session caching. This
1130 triggered odd behaviour from Outlook Express clients.
1132 PP/03 Add util/renew-opendmarc-tlds.sh script for safe renewal of public
1135 JH/38 DKIM: accept Ed25519 pubkeys in SubjectPublicKeyInfo-wrapped form,
1136 since the IETF WG has not yet settled on that versus the original
1137 "bare" representation.
1139 JH/39 Fix syslog logging for syslog_timestamp=no and log_selector +millisec.
1140 Previously the millisecond value corrupted the output.
1141 Fix also for syslog_pid=no and log_selector +pid, for which the pid
1142 corrupted the output.
1148 JH/01 Rework error string handling in TLS interface so that the caller in
1149 more cases is responsible for logging. This permits library-sourced
1150 string to be attached to addresses during delivery, and collapses
1151 pairs of long lines into single ones.
1153 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
1154 during configuration. Wildcards are allowed and expanded.
1156 JH/02 Rework error string handling in DKIM to pass more info back to callers.
1157 This permits better logging.
1159 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
1160 do not close it down and have the child transport start it up again on
1161 the passed-on TCP connection. Instead, proxy the child (and any
1162 subsequent ones) for TLS via a unix-domain socket channel. Logging is
1163 affected: the continued delivery log lines do not have any DNSSEC, TLS
1164 Certificate or OCSP information. TLS cipher information is still logged.
1166 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
1167 identical IP addresses on different listening ports. Will also affect
1170 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
1171 add noisy ifdef guards to special-case this sillyness.
1172 Patch from Bernd Kuhls.
1174 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
1175 than 255 are no longer allowed.
1177 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
1178 Disable the session-cache too, which might reduce our load. Since we
1179 currrectly use a new context for every connection, both as server and
1180 client, there is no benefit for these.
1181 GnuTLS appears to not support tickets server-side by default (we don't
1182 call gnutls_session_ticket_enable_server()) but client side is enabled
1183 by default on recent versions (3.1.3 +) unless the PFS priority string
1186 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
1187 <https://reproducible-builds.org/specs/source-date-epoch/>.
1189 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
1190 the check for any unsuccessful recipients did not notice the limit, and
1191 erroneously found still-pending ones.
1193 JH/08 Pipeline CHUNKING command and data together, on kernels that support
1194 MSG_MORE. Only in-clear (not on TLS connections).
1196 JH/09 Avoid using a temporary file during transport using dkim. Unless a
1197 transport-filter is involved we can buffer the headers in memory for
1198 creating the signature, and read the spool data file once for the
1199 signature and again for transmission.
1201 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
1202 4.77 as the kernel support then wasn't solid, having issues in 64bit
1203 mode. Now, it's been long enough. Add support for FreeBSD also.
1205 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
1206 case where the routing stage had gathered several addresses to send to
1207 a host before calling the transport for the first, we previously failed
1208 to close down TLS in the old transport process before passing the TCP
1209 connection to the new process. The new one sent a STARTTLS command
1210 which naturally failed, giving a failed delivery and bloating the retry
1211 database. Investigation and fix prototype from Wolfgang Breyha.
1213 JH/12 Fix check on SMTP command input synchronisation. Previously there were
1214 false-negatives in the check that the sender had not preempted a response
1215 or prompt from Exim (running as a server), due to that code's lack of
1216 awareness of the SMTP input buffering.
1218 PP/04 Add commandline_checks_require_admin option.
1219 Exim drops privileges sanely, various checks such as -be aren't a
1220 security problem, as long as you trust local users with access to their
1221 own account. When invoked by services which pass untrusted data to
1222 Exim, this might be an issue. Set this option in main configuration
1223 AND make fixes to the calling application, such as using `--` to stop
1226 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
1227 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
1228 responses to those, into a single TLS record each way (this usually means
1229 a single packet). As a side issue, smtp_enforce_sync now works on TLS
1232 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
1233 affects you only if you're dancing at the edge of the param size limits.
1234 If you are, and this message makes sense to you, then: raise the
1235 configured limit or use OpenSSL 1.1. Nothing we can do for older
1238 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
1239 cmdline element to get the documented default one. Previously it was
1242 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
1243 are unsuable for non-message-specific reasons - eg. connection timeout,
1244 banner-time rejection.
1246 JH/16 Fix logging of delivery remote port, when specified by router, under
1249 PP/06 Repair manualroute's ability to take options in any order, even if one
1250 is the name of a transport.
1253 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
1255 JH/17 Change the list-building routines interface to use the expanding-string
1256 triplet model, for better allocation and copying behaviour.
1258 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
1259 Previously it was constructed the first time a possibly-matching string
1260 was met in the configuration file input during startup; now it is done
1263 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
1264 compatible one, to avoid the (poorly documented) possibility of a config
1265 file in the working directory redirecting the DB files, possibly correpting
1266 some existing file. CVE-2017-10140 assigned for BDB.
1268 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
1269 cache-hot. Previously, although the result was properly cached, the
1270 initial verify call returned a defer.
1272 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
1273 the main verify for receipient in uncached-mode.
1275 JH/22 Retire historical build files to an "unsupported" subdir. These are
1276 defined as "ones for which we have no current evidence of testing".
1278 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
1279 if present. Previously it was ignored.
1281 JH/24 Start using specified-initialisers in C structure init coding. This is
1282 a C99 feature (it's 2017, so now considered safe).
1284 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
1285 if was a fixed-sized field and bitmask ops via macros; it is now more
1288 PP/07 GitHub PR 56: Apply MariaDB build fix.
1289 Patch provided by Jaroslav Škarvada.
1291 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
1292 during Coverity cleanups [4.87 JH/47]
1293 Diagnosis and fix provided by Michael Fischer v. Mollard.
1295 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
1296 the right size to place the terminating semicolon on its own folded
1297 line, the header hash was calculated to an incorrect value thanks to
1298 the (relaxed) space the fold became.
1300 HS/02 Fix Bug 2130: large writes from the transport subprocess were chunked
1301 and confused the parent.
1303 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
1304 which could crash as a result. This could lead to undeliverable messages.
1306 JH/28 Logging: "next input sent too soon" now shows where input was truncated
1309 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
1310 matters on fast-turnover and PID-randomising systems, which were getting
1311 out-of-order delivery.
1313 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
1314 a possibly-overlapping copy. The symptom was that "Remote host closed
1315 connection in response to HELO" was logged instead of the actual 4xx
1318 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
1319 Previously only that bufferd was discarded, resulting in SYMTP command
1322 JH/32 DKIM: when a message has multiple signatures matching an identity given
1323 in dkim_verify_signers, run the dkim acl once for each. Previously only
1324 one run was done. Bug 2189.
1326 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
1327 "panic the current process" to "deliberately defer". The panic log is
1328 still written with the problem list name; the mail and reject logs now
1329 get a temp-reject line for the message that was being handled, saying
1330 something like "domains check lookup or other defer". The SMTP 451
1331 message is still "Temporary local problem".
1333 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
1334 A crafted sequence of BDAT commands could result in in-use memory beeing
1335 freed. CVE-2017-16943.
1337 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
1338 from SMTP input. Previously it was always done; now only done for DATA
1339 and not BDAT commands. CVE-2017-16944.
1341 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
1342 to the message (such as an overlong header line). Previously this was
1343 not done and we did not exit BDAT mode. Followon from the previous item
1344 though a different problem.
1350 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
1351 than -2003 did; needs libidn2 in addition to libidn.
1353 JH/02 The path option on a pipe transport is now expanded before use.
1355 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
1356 Patch provided by "Björn", documentation fix added too.
1358 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
1359 missing a wire-to-host endian conversion.
1361 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
1362 close after a BDAT command line could be taken as a following command,
1363 giving a synch failure. Fix by only checking for synch immediately
1364 before acknowledging the chunk.
1366 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
1367 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
1368 macro. Patches provided by Josh Soref.
1370 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
1371 Previously we did not; the RFC seems ambiguous and VRFY is not listed
1372 by IANA as a service extension. However, John Klensin suggests that we
1375 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
1376 the dkim code may be unix-mode line endings rather than smtp wire-format
1377 CRLF, so prepend a CR to any bare LF.
1379 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
1380 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
1382 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
1383 the first were themselves being wrongly included in the feed into dkim
1384 processing; with most chunk sizes in use this resulted in an incorrect
1385 body hash calculated value.
1387 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
1388 DKIM signature block, for verification. Although advised against by
1389 standards it is specifically not ruled illegal.
1391 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
1393 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
1394 missing a body hash (the bh= tag).
1396 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
1397 It seems that HAProxy sends the Proxy Protocol information in clear and
1398 only then does a TLS startup, so do the same.
1400 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
1401 TCP connections (such as for Spamd) unless the daemon successfully set
1402 Fast Open mode on its listening sockets. This fixes breakage seen on
1403 too-old kernels or those not configured for Fast Open, at the cost of
1404 requiring both directions being enabled for TFO, and TFO never being used
1405 by non-daemon-related Exim processes.
1407 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
1408 endings, at least on the first header line. Try to canonify any that get
1409 past that check, despite the cost.
1411 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
1412 now limited to an arbitrary five deep, while parsing addresses with the
1413 strip_excess_angle_brackets option enabled.
1415 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
1416 instead leave the unprompted TLS handshake in socket buffer for the
1417 TLS library to consume.
1419 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
1421 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
1423 JH/16 Drop variables when they go out of scope. Memory management drops a whole
1424 region in one operation, for speed, and this leaves assigned pointers
1425 dangling. Add checks run only under the testsuite which checks all
1426 variables at a store-reset and panics on a dangling pointer; add code
1427 explicitly nulling out all the variables discovered. Fixes one known
1428 bug: a transport crash, where a dangling pointer for $sending_ip_address
1429 originally assigned in a verify callout, is re-used.
1431 PP/06 Drop '.' from @INC in various Perl scripts.
1433 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
1435 PP/08 Reduce a number of compilation warnings under clang; building with
1436 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
1437 should be warning-free.
1439 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
1441 HS/01 Fix portability problems introduced by PP/08 for platforms where
1442 realloc(NULL) is not equivalent to malloc() [SunOS et al].
1444 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
1445 chunk. This allows us to accept broken chunked messages. We need a more
1446 general solution here.
1448 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
1449 already-broken messages in the queue.
1451 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
1453 JH/19 Fix reference counting bug in routing-generated-address tracking.
1459 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
1460 supports it and a size is available (ie. the sending peer gave us one).
1462 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
1463 years of being deprecated). The replacements are the ACLs
1464 acl_smtp_mime and acl_not_smtp_mime.
1466 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
1467 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
1468 or even an in-clear connection were permitted. Now, if the host lookup
1469 was dnssec and dane was requested then the host is only used if the
1470 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
1471 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
1472 if one fails this test.
1473 This means that a poorly-configured remote DNS will make it incommunicado;
1474 but it protects against a DNS-interception attack on it.
1476 JH/04 Bug 1810: make continued-use of an open smtp transport connection
1477 non-noisy when a race steals the message being considered.
1479 JH/05 If main configuration option tls_certificate is unset, generate a
1480 self-signed certificate for inbound TLS connections.
1482 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
1483 in rewrites and routers.
1485 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
1486 and logged a warning sing 4.83; now they are a configuration file error.
1488 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
1489 (lacking @domain). Apply the same qualification processing as RCPT.
1491 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
1493 JH/10 Support ${sha256:} applied to a string (as well as the previous
1496 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
1497 a cutthrough deliver is pending, as we always want to make a connection.
1498 This also avoids re-routing the message when later placing the cutthrough
1499 connection after a verify cache hit.
1500 Do not update it with the verify result either.
1502 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
1503 when routing results in more than one destination address.
1505 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
1506 signing (which inhibits the cutthrough capability). Previously only
1507 the presence of an option was tested; now an expansion evaluating as
1508 empty is permissible (obviously it should depend only on data available
1509 when the cutthrough connection is made).
1511 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
1512 the relevant preceding SMTP command did not note the pipelining mode.
1514 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
1515 Previously they were not counted.
1517 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
1518 as one having no matching records. Previously we deferred the message
1519 that needed the lookup.
1521 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
1522 distinguished as "(=".
1524 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
1525 for missing MX records. Previously it only worked for missing A records.
1527 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
1529 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
1530 after the data-go-ahead and data-ack. Patch from Jason Betts.
1532 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
1533 even for a "none" policy. Patch from Tony Meyer.
1535 JH/22 Fix continued use of a connection for further deliveries. If a port was
1536 specified by a router, it must also match for the delivery to be
1539 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
1540 When one of the recipients of a message was unsuitable for the connection
1541 (has no matching addresses), we lost track of needing to mark it
1542 deferred. As a result mail would be lost.
1544 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
1546 JH/25 Decoding ACL controls is now done using a binary search; the source code
1547 takes up less space and should be simpler to maintain. Merge the ACL
1548 condition decode tables also, with similar effect.
1550 JH/26 Fix problem with one_time used on a redirect router which returned the
1551 parent address unchanged. A retry would see the parent address marked as
1552 delivered, so not attempt the (identical) child. As a result mail would
1555 JH/27 Fix a possible security hole, wherein a process operating with the Exim
1556 UID can gain a root shell. Credit to http://www.halfdog.net/ for
1557 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
1560 JH/28 Enable {spool,log} filesystem space and inode checks as default.
1561 Main config options check_{log,spool}_{inodes,space} are now
1562 100 inodes, 10MB unless set otherwise in the configuration.
1564 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
1565 Previously it only applied to the main-section connection policy
1568 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
1570 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
1571 by me. Added RFC7919 DH primes as an alternative.
1573 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
1574 are not in the system include path.
1576 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
1577 GnuTLS, when a session startup failed (eg because the client disconnected)
1578 Exim did stdio operations after fclose. This was exposed by a recent
1579 change which nulled out the file handle after the fclose.
1581 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
1582 signed directly by the cert-signing cert, rather than an intermediate
1583 OCSP-signing cert. This is the model used by LetsEncrypt.
1585 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
1587 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
1588 an incoming connection.
1590 HS/02 Bug 1802: Do not half-close the connection after sending a request
1593 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
1594 fallback to "prime256v1".
1596 JH/34 SECURITY: Use proper copy of DATA command in error message.
1597 Could leak key material. Remotely exploitable. CVE-2016-9963.
1603 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
1604 and 3.4.4 - once the server is enabled to respond to an OCSP request
1605 it does even when not requested, resulting in a stapling non-aware
1606 client dropping the TLS connection.
1608 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
1609 support variable-length bit vectors. No functional change.
1611 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
1612 The I= interface field on outgoing lines is now after the H= remote
1613 host field, same as incoming lines. There is a separate
1614 outgoing_interface log selector which allows you to disable the
1617 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
1618 If not running log_selector +smtp_connection the mainlog would be held
1619 open indefinitely after a "too many connections" event, including to a
1620 deleted file after a log rotate. Leave the per net connection logging
1621 leaving it open for efficiency as that will be quickly detected by the
1622 check on the next write.
1624 HS/01 Bug 1671: Fix post transport crash.
1625 Processing the wait-<transport> messages could crash the delivery
1626 process if the message IDs didn't exist for some reason. When
1627 using 'split_spool_directory=yes' the construction of the spool
1628 file name failed already, exposing the same netto behaviour.
1630 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
1631 mime_regex ACL conditions.
1633 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
1634 to DSN fail messages (bounces): remote IP, remote greeting, remote response
1635 to HELO, local diagnostic string.
1637 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
1638 log line to debug. Even when configured with a tls authenticator many
1639 client connections are expected to not authenticate in this way, so
1640 an authenticate fail is not an error.
1642 HS/02 Add the Exim version string to the process info. This way exiwhat
1643 gives some more detail about the running daemon.
1645 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
1646 matter for fast-change records such as DNSBLs.
1648 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
1649 retry record, even if constant. There may be multiple transports with
1650 different interface settings and the retry behaviour needs to be kept
1653 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
1655 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
1657 JH/10 Bug 840: fix log_defer_output option of pipe transport
1659 JH/11 Bug 830: use same host for all RCPTS of a message, even under
1660 hosts_randomize. This matters a lot when combined with mua_wrapper.
1662 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
1663 ${quote_pgsql:<string>} operator.
1665 JH/13 Bug 1708: avoid misaligned access in cached lookup.
1667 JH/14 Change header file name for freeradius-client. Relevant if compiling
1668 with Radius support; from the Gentoo tree and checked under Fedora.
1670 JH/15 Bug 1712: Introduce $prdr_requested flag variable
1672 JH/16 Bug 1714: Permit an empty string as expansion result for transport
1673 option transport_filter, meaning no filtering.
1675 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
1677 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
1678 defaults to "*" (all hosts). The variable is now available when not built
1679 with TLS, default unset, mainly to enable keeping the testsuite sane.
1680 If a server certificate is not supplied (via tls_certificate) an error is
1681 logged, and clients will find TLS connections fail on startup. Presumably
1682 they will retry in-clear.
1683 Packagers of Exim are strongly encouraged to create a server certificate
1684 at installation time.
1686 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
1687 with the $config_file variable.
1689 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
1690 in transport context, after the attempt, and per-recipient. The latter type
1691 is per host attempted. The event data is the error message, and the errno
1692 information encodes the lookup type (A vs. MX) used for the (first) host,
1693 and the trailing two digits of the smtp 4xx response.
1695 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
1696 to write to mainlog (or rejectlog, paniclog) in the window between file
1697 creation and permissions/ownership being changed. Particularly affects
1698 installations where exicyclog is run as root, rather than exim user;
1699 result is that the running daemon panics and dies.
1701 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
1703 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
1704 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
1705 "pri" and "weight". Note that the previous implicit priority given by the
1706 list order is no longer honoured.
1708 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
1709 for DKIM processing.
1711 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
1712 by defining SUPPORT_SOCKS.
1714 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
1715 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
1716 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
1717 variables are renamed to proxy_{local,external}_{address,port}.
1719 JH/27 Move Internationalisation support from Experimental to mainline, enabled
1720 for a build by defining SUPPORT_I18N
1722 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
1723 of the query string, and make ${quote_redis:} do that quoting.
1725 JH/29 Move Events support from Experimental to mainline, enabled by default
1726 and removable for a build by defining DISABLE_EVENT.
1728 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
1730 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
1731 cached by the daemon.
1733 JH/32 Move Redis support from Experimental to mainline, enabled for a build
1734 by defining LOOKUP_REDIS. The libhiredis library is required.
1736 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
1737 keys are given for lookup.
1739 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
1740 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
1741 only supported when built with TLS support. The PolarSSL SHA routines
1742 are still used when the TLS library is too old for convenient support.
1744 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
1745 openssl_options), for security. OpenSSL forces this from version 1.1.0
1746 server-side so match that on older versions.
1748 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
1749 allocation for $value could be released as the expansion processing
1750 concluded, but leaving the global pointer active for it.
1752 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
1753 and to use the domains and local_parts ACL conditions.
1755 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
1756 incorrectly not doubled on cutthrough transmission, hence seen as a
1757 body-termination at the receiving system - resulting in truncated mails.
1758 Commonly the sender saw a TCP-level error, and retransmitted the message
1759 via the normal store-and-forward channel. This could result in duplicates
1760 received - but deduplicating mailstores were liable to retain only the
1761 initial truncated version.
1763 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
1765 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
1767 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
1768 we're in there, support oversigning also; bug 1309.
1770 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
1772 HS/04 Add support for keep_environment and add_environment options.
1774 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
1775 either intentional arithmetic overflow during PRNG, or testing config-
1778 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
1779 delivery resulted in actual delivery. Cancel cutthrough before DATA
1782 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
1783 rejects a recipient: pass the reject to the originator.
1785 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
1786 Many were false-positives and ignorable, but it's worth fixing the
1789 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
1790 for the new environment-manipulation done at startup. Move the routines
1791 from being local to tls.c to being global via the os.c file.
1793 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
1794 an extract embedded as result-arg for a map, the first arg for extract
1795 is unavailable so we cannot tell if this is a numbered or keyed
1796 extraction. Accept either.
1802 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
1805 JH/02 The smtp transport option "multi_domain" is now expanded.
1807 JH/03 The smtp transport now requests PRDR by default, if the server offers
1810 JH/04 Certificate name checking on server certificates, when exim is a client,
1811 is now done by default. The transport option tls_verify_cert_hostnames
1812 can be used to disable this per-host. The build option
1813 EXPERIMENTAL_CERTNAMES is withdrawn.
1815 JH/05 The value of the tls_verify_certificates smtp transport and main options
1816 default to the word "system" to access the system default CA bundle.
1817 For GnuTLS, only version 3.0.20 or later.
1819 JH/06 Verification of the server certificate for a TLS connection is now tried
1820 (but not required) by default. The verification status is now logged by
1821 default, for both outbound TLS and client-certificate supplying inbound
1824 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
1827 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
1828 Status Notification (bounce) messages are now MIME format per RFC 3464.
1829 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
1830 under the control of the dsn_advertise_hosts option, and routers may
1831 have a dsn_lasthop option.
1833 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
1834 default, modifiable by a malware= option. The list separator for
1835 the options can now be changed in the usual way. Bug 68.
1837 JH/10 The smtp_receive_timeout main option is now expanded before use.
1839 JH/11 The incoming_interface log option now also enables logging of the
1840 local interface on delivery outgoing connections.
1842 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
1843 if the interface and destination host and port all match.
1845 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
1848 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
1849 Patch from Andrew Lewis.
1851 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
1852 now supports optional time-restrictions, weighting, and priority
1853 modifiers per server. Patch originally by <rommer@active.by>.
1855 JH/16 The spamd_address main option now supports a mixed list of local
1856 and remote servers. Remote servers can be IPv6 addresses, and
1857 specify a port-range.
1859 JH/17 Bug 68: The spamd_address main option now supports an optional
1860 timeout value per server.
1862 JH/18 Bug 1581: Router and transport options headers_add/remove can
1863 now have the list separator specified.
1865 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
1868 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
1871 JH/21 Support for the A6 type of dns record is withdrawn.
1873 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
1874 rather than the verbs used.
1876 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
1877 from 255 to 1024 chars.
1879 JH/24 Verification callouts now attempt to use TLS by default.
1881 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
1882 are generic router options now. The defaults didn't change.
1884 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
1885 Original patch from Alexander Shikoff, worked over by JH.
1887 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
1888 files if ZCAT_COMMAND is not executable.
1890 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
1892 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
1894 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
1895 Normally benign, it bites when the pair was led to by a CNAME;
1896 modern usage is to not canonicalize the domain to a CNAME target
1897 (and we were inconsistent anyway for A-only vs AAAA+A).
1899 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
1901 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
1902 when evaluating $sender_host_dnssec.
1904 JH/31 Check the HELO verification lookup for DNSSEC, adding new
1905 $sender_helo_dnssec variable.
1907 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
1909 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
1911 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
1913 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
1914 documented as working, but never had. Support all but $spam_report.
1916 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
1917 added for tls authenticator.
1919 HS/03 Add perl_taintmode main config option
1925 TL/01 When running the test suite, the README says that variables such as
1926 no_msglog_check are global and can be placed anywhere in a specific
1927 test's script, however it was observed that placement needed to be near
1928 the beginning for it to behave that way. Changed the runtest perl
1929 script to read through the entire script once to detect and set these
1930 variables, reset to the beginning of the script, and then run through
1931 the script parsing/test process like normal.
1933 TL/02 The BSD's have an arc4random API. One of the functions to induce
1934 adding randomness was arc4random_stir(), but it has been removed in
1935 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
1936 function when detected.
1938 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
1939 cause callback expansion.
1941 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
1942 syntax errors in an expansion can be treated as a string instead of
1943 logging or causing an error, due to the internal use of bool_lax
1944 instead of bool when processing it.
1946 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
1947 server certificates when making smtp deliveries.
1949 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
1951 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
1953 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
1955 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
1956 Merged patch from Sebastian Wiedenroth.
1958 JH/05 Fix results-pipe from transport process. Several recipients, combined
1959 with certificate use, exposed issues where response data items split
1960 over buffer boundaries were not parsed properly. This eventually
1961 resulted in duplicates being sent. This issue only became common enough
1962 to notice due to the introduction of connection certificate information,
1963 the item size being so much larger. Found and fixed by Wolfgang Breyha.
1965 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
1966 size buffer was used, resulting in syntax errors when an expansion
1969 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1970 version 3.3.6 or later.
1972 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1973 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1974 both become "event_action", the variables become $event_name, $event_data
1975 and $event_defer_errno. There is a new variable $verify_mode, usable in
1976 routers, transports and related events. The tls:cert event is now also
1977 raised for inbound connections, if the main configuration event_action
1980 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1981 early OCSP support, but no stapling (appears to be less than 1.0.0).
1983 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1984 server certificate names available under the smtp transport option
1985 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1988 JH/10 Time-related extraction expansions from certificates now use the main
1989 option "timezone" setting for output formatting, and are consistent
1990 between OpenSSL and GnuTLS compilations. Bug 1541.
1992 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1993 encoded parameter in the incoming message. Bug 1558.
1995 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1996 include certificate info, eximon was claiming there were spoolfile
1999 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
2001 JH/14 Log delivery-related information more consistently, using the sequence
2002 "H=<name> [<ip>]" wherever possible.
2004 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
2005 are problematic for Debian distribution, omit them from the release
2008 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
2010 JH/16 Fix string representation of time values on 64bit time_t architectures.
2013 JH/17 Fix a null-indirection in certextract expansions when a nondefault
2014 output list separator was used.
2019 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
2020 checkers that were complaining about end of non-void function with no
2023 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
2024 This was a regression introduced in 4.83 by another bugfix.
2026 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
2028 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
2029 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
2035 TF/01 Correctly close the server side of TLS when forking for delivery.
2037 When a message was received over SMTP with TLS, Exim failed to clear up
2038 the incoming connection properly after forking off the child process to
2039 deliver the message. In some situations the subsequent outgoing
2040 delivery connection happened to have the same fd number as the incoming
2041 connection previously had. Exim would try to use TLS and fail, logging
2042 a "Bad file descriptor" error.
2044 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
2045 utilities have not been installed.
2047 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
2048 temporary space as the ACL may create new global variables.
2050 TL/01 LDAP support uses per connection or global context settings, depending
2051 upon the detected version of the libraries at build time.
2053 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
2054 to extract and use the src ip:port in logging and expansions as if it
2055 were a direct connection from the outside internet. PPv2 support was
2056 updated based on HAProxy spec change in May 2014.
2058 JH/02 Add ${listextract {number}{list}{success}{fail}}.
2060 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
2061 Properly escape header and check for NULL return.
2063 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
2066 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
2068 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
2069 characters in header names, implemented as a verify condition.
2070 Contributed by Michael Fischer v. Mollard.
2072 TL/05 Rename SPF condition results err_perm and err_temp to standardized
2073 results permerror and temperror. Previous values are deprecated but
2074 still accepted. In a future release, err_perm and err_temp will be
2075 completely removed, which will be a backward incompatibility if the
2076 ACL tests for either of these two old results. Patch contributed by
2077 user bes-internal on the mailing list.
2079 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
2081 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
2082 selectors, in both main and reject logs.
2084 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
2087 JH/07 Add malware type "sock" for talking to simple daemon.
2089 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
2091 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
2092 routers/transports under cutthrough routing.
2094 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
2095 numbers. Touch up "bool" conditional to keep the same definition.
2097 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
2099 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
2101 JH/12 Expand items in router/transport headers_add or headers_remove lists
2102 individually rather than the list as a whole. Bug 1452.
2104 Required for reasonable handling of multiple headers_ options when
2105 they may be empty; requires that headers_remove items with embedded
2106 colons must have them doubled (or the list-separator changed).
2108 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
2109 view the policy declared in the DMARC record. Currently, $dmarc_status
2110 is a combined value of both the record presence and the result of the
2113 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
2115 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
2116 dnslookup router and the smtp transport (applying to the forward
2119 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
2120 of ldap servers used for a specific lookup. Patch provided by Heiko
2123 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
2124 New variable $lookup_dnssec_authenticated for observability.
2126 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
2127 Patch submitted by Lars Timman.
2129 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
2131 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
2132 Requires trusted mode and valid format message id, aborts otherwise.
2133 Patch contributed by Heiko Schlichting.
2135 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
2136 certextract with support for various fields. Bug 1358.
2138 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
2139 is requested by default, modifiable by smtp transport option
2142 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
2143 operate on certificate variables to give certificate fingerprints
2144 Also new ${sha256:cert_variable}.
2146 JH/23 The PRDR feature is moved from being Experimental into the mainline.
2148 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
2149 Christian Aistleitner.
2151 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
2153 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
2154 file. Patch from Wolfgang Breyha.
2156 JH/25 Expand the coverage of the delivery $host and $host_address to
2157 client authenticators run in verify callout. Bug 1476.
2159 JH/26 Port service names are now accepted for tls_on_connect_ports, to
2160 align with daemon_smtp_ports. Bug 72.
2162 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
2163 support and error reporting did not work properly.
2165 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
2166 and is readable. Patch from Andrew Colin Kissa.
2168 TL/14 Enhance documentation of ${run expansion and how it parses the
2169 commandline after expansion, particularly in the case when an
2170 unquoted variable expansion results in an empty value.
2172 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
2174 PP/02 Fix internal collision of T_APL on systems which support RFC3123
2175 by renaming away from it. Addresses GH issue 15, reported by
2178 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
2180 TL/15 SECURITY: prevent double expansion in math comparison functions
2181 (can expand unsanitized data). Not remotely exploitable.
2188 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
2190 PP/02 Make -n do something, by making it not do something.
2191 When combined with -bP, the name of an option is not output.
2193 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
2196 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
2197 $sender_host_name and config options to manage this, and basic check
2200 PP/05 DSCP support for outbound connections and control modifier for inbound.
2202 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
2203 (Only plugin which currently uses this is kerberos4, which nobody should
2204 be using, but we should make it available and other future plugins might
2205 conceivably use it, even though it would break NAT; stuff *should* be
2206 using channel bindings instead).
2208 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
2209 name; added for Sendmail compatibility; requires admin caller.
2210 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
2211 just ignore it); requires trusted caller.
2212 Also parse but ignore: -Ac -Am -X<logfile>
2215 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
2217 TL/02 Add +smtp_confirmation as a default logging option.
2219 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
2220 Patch by Magnus Holmgren from 2007-02-20.
2222 TL/04 Bugzilla 1281 - Spec typo.
2223 Bugzilla 1283 - Spec typo.
2224 Bugzilla 1290 - Spec grammar fixes.
2226 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
2228 TL/06 Add Experimental DMARC support using libopendmarc libraries.
2230 TL/07 Fix an out of order global option causing a segfault. Reported to dev
2231 mailing list by by Dmitry Isaikin.
2233 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
2235 JH/02 Support "G" suffix to numbers in ${if comparisons.
2237 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
2239 NM/01 Bugzilla 1197 - Spec typo
2240 Bugzilla 1196 - Spec examples corrections
2242 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
2244 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
2245 gnutls_enable_pkcs11, but renamed to more accurately indicate its
2248 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
2249 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
2251 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
2252 "acl {{name}{arg}...}", and optional args on acl condition
2255 JH/05 Permit multiple router/transport headers_add/remove lines.
2257 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
2259 JH/07 Avoid using a waiting database for a single-message-only transport.
2260 Performance patch from Paul Fisher. Bugzilla 1262.
2262 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
2265 JH/09 Add $headers_added variable, with content from use of ACL modifier
2266 add_header (but not yet added to the message). Bugzilla 199.
2268 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
2269 Pulled from Bugzilla 817 by Wolfgang Breyha.
2271 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
2273 (nb: this is the same fix as in Exim 4.80.1)
2275 JH/11 Add A= logging on delivery lines, and a client_set_id option on
2278 JH/12 Add optional authenticated_sender logging to A= and a log_selector
2281 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
2283 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
2284 advertise SMTP AUTH mechanism to us, instead of a generic
2285 protocol violation error. Also, make Exim more robust to bad
2286 data from the Dovecot auth socket.
2288 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
2290 When a queue runner is handling a message, Exim first routes the
2291 recipient addresses, during which it prunes them based on the retry
2292 hints database. After that it attempts to deliver the message to
2293 any remaining recipients. It then updates the hints database using
2296 So if a recipient address works intermittently, it can get repeatedly
2297 deferred at routing time. The retry hints record remains fresh so the
2298 address never reaches the final cutoff time.
2300 This is a fairly common occurrence when a user is bumping up against
2301 their storage quota. Exim had some logic in its local delivery code
2302 to deal with this. However it did not apply to per-recipient defers
2303 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
2305 This change adds a proper retry rule check during routing so that the
2306 final cutoff time is checked against the message's age. We only do
2307 this check if there is an address retry record and there is not a
2308 domain retry record; this implies that previous attempts to handle
2309 the address had the retry_use_local_parts option turned on. We use
2310 this as an approximation for the destination being like a local
2311 delivery, as in LMTP.
2313 I suspect this new check makes the old local delivery cutoff check
2314 redundant, but I have not verified this so I left the code in place.
2316 TF/02 Correct gecos expansion when From: is a prefix of the username.
2318 Test 0254 submits a message to Exim with the header
2322 When I ran the test suite under the user fanf2, Exim expanded
2323 the header to contain my full name, whereas it should have added
2324 a Resent-Sender: header. It erroneously treats any prefix of the
2325 username as equal to the username.
2327 This change corrects that bug.
2329 GF/01 DCC debug and logging tidyup
2330 Error conditions log to paniclog rather than rejectlog.
2331 Debug lines prefixed by "DCC: " to remove any ambiguity.
2333 TF/03 Avoid unnecessary rebuilds of lookup-related code.
2335 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
2336 Bug spotted by Jeremy Harris; was flawed since initial commit.
2337 Would have resulted in OCSP responses post-SNI triggering an Exim
2338 NULL dereference and crash.
2340 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
2342 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
2343 Bug detection, analysis and fix by Samuel Thibault.
2344 Bugzilla 1331, Debian bug #698092.
2346 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
2348 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
2349 Server implementation by Todd Lyons, client by JH.
2350 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
2351 config variable "prdr_enable" controls whether the server
2352 advertises the facility. If the client requests PRDR a new
2353 acl_data_smtp_prdr ACL is called once for each recipient, after
2354 the body content is received and before the acl_smtp_data ACL.
2355 The client is controlled by both of: a hosts_try_prdr option
2356 on the smtp transport, and the server advertisement.
2357 Default client logging of deliveries and rejections involving
2358 PRDR are flagged with the string "PRDR".
2360 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
2361 fclose(). Diagnosis by Todd Lyons.
2363 PP/17 Update configure.default to handle IPv6 localhost better.
2364 Patch by Alain Williams (plus minor tweaks).
2367 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
2368 This is now consistent with GnuTLS, and is now documented: the
2369 previous undocumented portable approach to treating the option as
2370 unset was to force an expansion failure. That still works, and
2371 an empty string is now equivalent.
2373 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
2374 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
2375 not performing validation itself.
2377 PP/20 Added force_command boolean option to pipe transport.
2378 Patch from Nick Koston, of cPanel Inc.
2380 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
2383 TF/04 Added udpsend ACL modifier and hexquote expansion operator
2385 PP/21 Fix eximon continuous updating with timestamped log-files.
2386 Broken in a format-string cleanup in 4.80, missed when I repaired the
2387 other false fix of the same issue.
2388 Report and fix from Heiko Schlichting.
2391 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
2392 Report from Prashanth Katuri.
2394 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
2395 It's SecureTransport, so affects any MacOS clients which use the
2396 system-integrated TLS libraries, including email clients.
2398 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
2399 using a MIME ACL for non-SMTP local injection.
2400 Report and assistance in diagnosis by Warren Baker.
2402 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
2404 JH/16 Fix comparisons for 64b. Bugzilla 1385.
2406 TL/09 Add expansion variable $authenticated_fail_id to keep track of
2407 last id that failed so it may be referenced in subsequent ACL's.
2409 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
2412 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
2413 ldap library initialization, allowing self-signed CA's to be
2414 used. Also properly sets require_cert option later in code by
2415 using NULL (global ldap config) instead of ldap handle (per
2416 session). Bug diagnosis and testing by alxgomz.
2418 TL/12 Enhanced documentation in the ratelimit.pl script provided in
2419 the src/util/ subdirectory.
2421 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
2422 renamed to Transport Post Delivery Action by Jeremy Harris, as
2425 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
2426 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
2427 redis_servers = needs to be configured which will be used by the redis
2428 lookup. Patch from Warren Baker, of The Packet Hub.
2430 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
2432 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
2433 hostname or reverse DNS when processing a host list. Used suggestions
2434 from multiple comments on this bug.
2436 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
2438 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
2439 Missed a few lines, added it to make the runtest require no keyboard
2442 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
2443 contains upper case chars. Make router use caseful_local_part.
2445 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
2446 support when GnuTLS has been built with p11-kit.
2452 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
2454 This, or similar/improved, will also be change PP/11 of 4.82.
2460 PP/01 Handle short writes when writing local log-files.
2461 In practice, only affects FreeBSD (8 onwards).
2462 Bugzilla 1053, with thanks to Dmitry Isaikin.
2464 NM/01 Bugzilla 949 - Documentation tweak
2466 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
2469 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
2471 PP/02 Implemented gsasl authenticator.
2473 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
2475 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
2476 `pkg-config foo` for cflags/libs.
2478 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
2479 with rest of GSASL and with heimdal_gssapi.
2481 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
2482 `pkg-config foo` for cflags/libs for the TLS implementation.
2484 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
2485 properties get this fed in as external SSF. A number of robustness
2486 and debugging improvements to the cyrus_sasl authenticator.
2488 PP/08 cyrus_sasl server now expands the server_realm option.
2490 PP/09 Bugzilla 1214 - Log authentication information in reject log.
2491 Patch by Jeremy Harris.
2493 PP/10 Added dbmjz lookup type.
2495 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
2497 PP/12 MAIL args handles TAB as well as SP, for better interop with
2498 non-compliant senders.
2499 Analysis and variant patch by Todd Lyons.
2501 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
2502 Bug report from Lars Müller <lars@samba.org> (via SUSE),
2503 Patch from Dirk Mueller <dmueller@suse.com>
2505 PP/13 tls_peerdn now print-escaped for spool files.
2506 Observed some $tls_peerdn in wild which contained \n, which resulted
2507 in spool file corruption.
2509 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
2510 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
2511 or write after TLS renegotiation, which otherwise led to messages
2514 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
2515 as a tracking header (ie: a signed header comes before the signature).
2516 Patch from Wolfgang Breyha.
2518 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
2519 comma-sep list; embedded commas doubled.
2521 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
2523 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
2525 Report and patch from Dmitry Banschikov.
2527 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
2528 Removed SSL_clear() after SSL_new() which led to protocol negotiation
2529 failures. We appear to now support TLS1.1+ with Exim.
2531 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
2532 lets Exim select keys and certificates based upon TLS SNI from client.
2533 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
2534 before an outbound SMTP session. New log_selector, +tls_sni.
2536 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
2537 NULL dereference. Report and patch from Alun Jones.
2539 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
2540 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
2541 Not seeing resolver debug output on NetBSD, but suspect this is a
2542 resolver implementation change.
2544 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
2545 Left warnings. Added "eximon gdb" invocation mode.
2547 PP/21 Defaulting "accept_8bitmime" to true, not false.
2549 PP/22 Added -bw for inetd wait mode support.
2551 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
2552 locate the relevant includes and libraries. Made this the default.
2554 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
2555 Bugzilla 1246, report and most of solution from Tomasz Kusy.
2557 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
2558 This may cause build issues on older platforms.
2560 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
2561 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
2562 gnutls_require_mac & gnutls_require_protocols (no longer supported).
2563 Added SNI support via GnuTLS too.
2564 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
2566 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
2568 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
2569 Applied second patch from Janne, implementing suggestion to default
2570 multiple-strings-in-record handling to match SPF spec.
2572 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
2574 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
2575 read-only, out of scope).
2576 Patch from Wolfgang Breyha, report from Stuart Northfield.
2578 PP/29 Fix three issues highlighted by clang analyser static analysis.
2579 Only crash-plausible issue would require the Cambridge-specific
2580 iplookup router and a misconfiguration.
2581 Report from Marcin Mirosław.
2583 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
2585 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
2586 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
2587 As part of this, removing so much warning spew let me fix some minor
2588 real issues in debug logging.
2590 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
2591 assignment on my part. Fixed.
2593 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
2594 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
2595 Janne Snabb (who went above and beyond: thank you).
2597 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
2598 string otherwise requires a connection and a bunch more work and it's
2599 relatively easy to get wrong. Should also expose TLS library linkage
2602 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
2603 64-bit ${eval} (JH/03).
2605 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
2606 GNU libc to support some of the 64-bit stuff, should not lead to
2607 conflicts. Defined before os.h is pulled in, so if a given platform
2608 needs to override this, it can.
2610 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
2611 protection layer was required, which is not implemented.
2612 Bugzilla 1254, patch from Wolfgang Breyha.
2614 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
2615 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
2616 tls_dhparam take prime identifiers. Also unbreak combination of
2617 OpenSSL+DH_params+TLSSNI.
2619 PP/39 Disable SSLv2 by default in OpenSSL support.
2625 PP/01 Solaris build fix for Oracle's LDAP libraries.
2626 Bugzilla 1109, patch from Stephen Usher.
2628 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
2630 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
2633 TF/02 Fix a couple more cases where we did not log the error message
2634 when unlink() failed. See also change 4.74-TF/03.
2636 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
2637 lock up or crash if it happened to be inside a call to libc when it
2638 got a SIGUSR1 from exiwhat.
2640 The SIGUSR1 handler appends the current process status to the process
2641 log which is later printed by exiwhat. It used to use the general
2642 purpose logging code to do this, but several functions it calls are
2643 not safe for signals.
2645 The new output code in the SIGUSR1 handler is specific to the process
2646 log, and simple enough that it's easy to inspect for signal safety.
2647 Removing some special cases also simplifies the general logging code.
2648 Removing the spurious timestamps from the process log simplifies
2651 TF/04 Improved ratelimit ACL condition.
2653 The /noupdate option has been deprecated in favour of /readonly which
2654 has clearer semantics. The /leaky, /strict, and /readonly update modes
2655 are mutually exclusive. The update mode is no longer included in the
2656 database key; it just determines when the database is updated. (This
2657 means that when you upgrade Exim will forget old rate measurements.)
2659 Exim now checks that the per_* options are used with an update mode that
2660 makes sense for the current ACL. For example, when Exim is processing a
2661 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
2662 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
2663 must specify per_mail/readonly. If you omit the update mode it defaults to
2664 /leaky where that makes sense (as before) or /readonly where required.
2666 The /noupdate option is now undocumented but still supported for
2667 backwards compatibility. It is equivalent to /readonly except that in
2668 ACLs where /readonly is required you may specify /leaky/noupdate or
2669 /strict/noupdate which are treated the same as /readonly.
2671 A useful new feature is the /count= option. This is a generalization
2672 of the per_byte option, so that you can measure the throughput of other
2673 aggregate values. For example, the per_byte option is now equivalent
2674 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
2676 The per_rcpt option has been generalized using the /count= mechanism
2677 (though it's more complicated than the per_byte equivalence). When it is
2678 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
2679 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
2680 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
2681 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
2682 Note that using per_rcpt with a non-readonly update mode in more than
2683 one ACL will cause the recipients to be double-counted. (The per_mail
2684 and per_byte options don't have this problem.)
2686 The handling of very low rates has changed slightly. If the computed rate
2687 is less than the event's count (usually one) then this event is the first
2688 after a long gap. In this case the rate is set to the same as this event's
2689 count, so that the first message of a spam run is counted properly.
2691 The major new feature is a mechanism for counting the rate of unique
2692 events. The new per_addr option counts the number of different
2693 recipients that someone has sent messages to in the last time period. It
2694 behaves like per_rcpt if all the recipient addresses are different, but
2695 duplicate recipient addresses do not increase the measured rate. Like
2696 the /count= option this is a general mechanism, so the per_addr option
2697 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
2698 example, measure the rate that a client uses different sender addresses
2699 with the options per_mail/unique=$sender_address. There are further
2700 details in the main documentation.
2702 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
2704 TF/06 Removed a few PCRE remnants.
2706 TF/07 Automatically extract Exim's version number from tags in the git
2707 repository when doing development or release builds.
2709 PP/02 Raise smtp_cmd_buffer_size to 16kB.
2710 Bugzilla 879. Patch from Paul Fisher.
2712 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
2713 Heavily based on revision 40f9a89a from Simon Arlott's tree.
2716 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
2718 PP/05 Variable $av_failed, true if the AV scanner deferred.
2719 Bugzilla 1078. Patch from John Horne.
2721 PP/06 Stop make process more reliably on build failure.
2722 Bugzilla 1087. Patch from Heiko Schlittermann.
2724 PP/07 Make maildir_use_size_file an _expandable_ boolean.
2725 Bugzilla 1089. Patch from Heiko Schlittermann.
2727 PP/08 Handle ${run} returning more data than OS pipe buffer size.
2728 Bugzilla 1131. Patch from Holger Weiß.
2730 PP/09 Handle IPv6 addresses with SPF.
2731 Bugzilla 860. Patch from Wolfgang Breyha.
2733 PP/10 GnuTLS: support TLS 1.2 & 1.1.
2735 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
2738 PP/11 match_* no longer expand right-hand-side by default.
2739 New compile-time build option, EXPAND_LISTMATCH_RHS.
2740 New expansion conditions, "inlist", "inlisti".
2742 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
2744 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
2746 PP/14 fix log_write() format string regression from TF/03.
2747 Bugzilla 1152. Patch from Dmitry Isaikin.
2753 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
2755 PP/02 Harmonised TLS library version reporting; only show if debugging.
2756 Layout now matches that introduced for other libraries in 4.74 PP/03.
2758 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
2760 PP/04 New "dns_use_edns0" global option.
2762 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
2765 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
2766 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
2768 TK/01 Updated PolarSSL code to 0.14.2.
2769 Bugzilla 1097. Patch from Andreas Metzler.
2771 PP/07 Catch divide-by-zero in ${eval:...}.
2772 Fixes bugzilla 1102.
2774 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
2777 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
2778 format-string attack -- SECURITY: remote arbitrary code execution.
2780 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
2781 time unintentionally subject to list matching rules, letting the header
2782 cause arbitrary Exim lookups (of items which can occur in lists, *not*
2783 arbitrary string expansion). This allowed for information disclosure.
2785 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
2786 INT_MIN/-1 -- value coerced to INT_MAX.
2792 NM/01 Workaround for PCRE version dependency in version reporting
2795 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
2796 This fixes portability to compilers other than gcc, notably
2797 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
2799 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
2800 makefiles for portability to HP-UX and POSIX correctness.
2802 PP/01 Permit LOOKUP_foo enabling on the make command-line.
2803 Also via indented variable definition in the Makefile.
2804 (Debugging by Oliver Heesakkers).
2806 PP/02 Restore caching of spamd results with expanded spamd_address.
2807 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
2809 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
2810 Improves build reliability. Fix from: Frank Elsner
2812 NM/02 Fix wide character breakage in the rfc2047 coding
2813 Fixes bug 1064. Patch from Andrey N. Oktyabrski
2815 NM/03 Allow underscore in dnslist lookups
2816 Fixes bug 1026. Patch from Graeme Fowler
2818 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
2819 Code patches from Adam Ciarcinski of NetBSD.
2821 NM/04 Fixed exiqgrep to cope with mailq missing size issue
2824 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
2825 is logged, to avoid truncation. Patch from John Horne.
2827 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
2828 Patch from Jakob Hirsch.
2830 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
2831 SQL string expansion failure details.
2832 Patch from Andrey Oktyabrski.
2834 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
2835 Patch from Simon Arlott.
2837 PP/09 New lookups functionality failed to compile on old gcc which rejects
2838 extern declarations in function scope.
2839 Patch from Oliver Fleischmann
2841 PP/10 Use sig_atomic_t for flags set from signal handlers.
2842 Check getgroups() return and improve debugging.
2843 Fixed developed for diagnosis in bug 927 (which turned out to be
2846 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
2847 Patch from Mark Zealey.
2849 PP/12 Bugzilla 1056: Improved spamd server selection.
2850 Patch from Mark Zealey.
2852 PP/13 Bugzilla 1086: Deal with maildir quota file races.
2853 Based on patch from Heiko Schlittermann.
2855 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
2856 Patch from Uwe Doering, sign-off by Michael Haardt.
2858 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
2859 variable declaration deep within a block. Bug and patch from
2862 PP/15 lookups-Makefile IRIX compatibility coercion.
2864 PP/16 Make DISABLE_DKIM build knob functional.
2866 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
2867 Patch by Simon Arlott
2869 TF/03 Fix valgrind.h portability to C89 compilers that do not support
2870 variable argument macros. Our copy now differs from upstream.
2876 TF/01 Failure to get a lock on a hints database can have serious
2877 consequences so log it to the panic log.
2879 TF/02 Log LMTP confirmation messages in the same way as SMTP,
2880 controlled using the smtp_confirmation log selector.
2882 TF/03 Include the error message when we fail to unlink a spool file.
2884 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
2885 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
2886 for maintaining out-of-tree patches for some time.
2888 PP/01 Bugzilla 139: Documentation and portability issues.
2889 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
2890 Handle per-OS dynamic-module compilation flags.
2892 PP/02 Let /dev/null have normal permissions.
2893 The 4.73 fixes were a little too stringent and complained about the
2894 permissions on /dev/null. Exempt it from some checks.
2895 Reported by Andreas M. Kirchwitz.
2897 PP/03 Report version information for many libraries, including
2898 Exim version information for dynamically loaded libraries. Created
2899 version.h, now support a version extension string for distributors
2900 who patch heavily. Dynamic module ABI change.
2902 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
2903 privilege escalation vulnerability whereby the Exim run-time user
2904 can cause root to append content of the attacker's choosing to
2907 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
2910 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
2911 If dropping privileges for untrusted macros, we disabled normal logging
2912 on the basis that it would fail; for the Exim run-time user, this is not
2913 the case, and it resulted in successful deliveries going unlogged.
2914 Fixed. Reported by Andreas Metzler.
2920 PP/01 Date: & Message-Id: revert to normally being appended to a message,
2921 only prepend for the Resent-* case. Fixes regression introduced in
2922 Exim 4.70 by NM/22 for Bugzilla 607.
2924 PP/02 Include check_rfc2047_length in configure.default because we're seeing
2925 increasing numbers of administrators be bitten by this.
2927 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
2929 PP/03 Bugzilla 994: added openssl_options main configuration option.
2931 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
2933 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
2935 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
2937 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
2938 without a peer certificate, leading to a segfault because of an
2939 assumption that peers always have certificates. Be a little more
2940 paranoid. Problem reported by Martin Tscholak.
2942 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
2943 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
2944 NB: ClamAV planning to remove STREAM in "middle of 2010".
2945 CL also introduces -bmalware, various -d+acl logging additions and
2946 more caution in buffer sizes.
2948 PP/09 Implemented reverse_ip expansion operator.
2950 PP/10 Bugzilla 937: provide a "debug" ACL control.
2952 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
2954 PP/12 Bugzilla 973: Implement --version.
2956 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
2958 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
2960 PP/15 Bugzilla 816: support multiple condition rules on Routers.
2962 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
2963 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
2964 ignore trailing whitespace.
2966 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2968 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2971 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2972 Notification from Dr Andrew Aitchison.
2974 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2975 ExtendedDetectionInfo response format.
2976 Notification from John Horne.
2978 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2981 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2982 XSL and documented dependency on system catalogs, with examples of how
2985 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2988 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2989 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2990 configuration file which is writeable by the Exim user or group.
2992 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2993 of configuration files to cover files specified with the -C option if
2994 they are going to be used with root privileges, not just the default
2997 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2998 option (effectively making it always true).
3000 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
3001 files to be used while preserving root privileges.
3003 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
3004 that rogue child processes cannot use them.
3006 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
3007 run-time user, instead of root.
3009 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
3010 Exim run-time user without dropping privileges.
3012 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
3013 result string, instead of calling string_vformat() twice with the same
3016 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
3017 for other users. Others should always drop root privileges if they use
3018 -C on the command line, even for a whitelisted configure file.
3020 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
3022 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
3028 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
3029 $data_path, and $header_path variables; fixed documentation bugs and
3032 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
3033 exipick to access non-standard spools, including the "frozen" queue
3036 NM/01 Bugzilla 965: Support mysql stored procedures.
3037 Patch from Alain Williams
3039 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
3041 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
3042 Patch from Andreas Metzler
3044 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
3045 Patch from Kirill Miazine
3047 NM/05 Bugzilla 671: Added umask to procmail example.
3049 JJ/03 installed exipick 20100323.0, fixing doc bug
3051 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
3052 directory. Notification and patch from Dan Rosenberg.
3054 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
3056 TK/02 Improve log output when DKIM signing operation fails.
3058 MH/01 Treat the transport option dkim_domain as a colon separated
3059 list, not as a single string, and sign the message with each element,
3060 omitting multiple occurences of the same signer.
3062 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
3063 Bugzilla 985, 986. Patch by Simon Arlott
3065 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
3066 Patch by Simon Arlott
3068 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
3069 MBX locking. Notification from Dan Rosenberg.
3075 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
3077 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
3079 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
3081 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
3083 NM/04 Bugzilla 915: Flag broken perl installation during build.
3089 TK/01 Added patch by Johannes Berg that expands the main option
3090 "spamd_address" if it starts with a dollar sign.
3092 TK/02 Write list of recipients to X-Envelope-Sender header when building
3093 the mbox-format spool file for content scanning (suggested by Jakob
3096 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
3097 (http://www.dcc-servers.net/) support via dccifd. Activated by
3098 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
3100 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
3101 by Mark Daniel Reidel <mr@df.eu>.
3103 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
3104 When building exim an external PCRE library is now needed -
3105 PCRE is a system library on the majority of modern systems.
3106 See entry on PCRE_LIBS in EDITME file.
3108 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
3109 conversation. Added nologin parameter to request.
3110 Patch contributed by Kirill Miazine.
3112 TF/01 Do not log submission mode rewrites if they do not change the address.
3114 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
3116 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
3117 log files in place. Contributed by Roberto Lima.
3119 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
3121 TF/03 Bugzilla 615: When checking the local_parts router precondition
3122 after a local_part_suffix or local_part_prefix option, Exim now
3123 does not use the address's named list lookup cache, since this
3124 contains cached lookups for the whole local part.
3126 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
3127 Robert Millan. Documentation is in experimental-spec.txt.
3129 TF/04 Bugzilla 668: Fix parallel build (make -j).
3131 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
3133 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
3134 Patch provided by Jan Srzednicki.
3136 TF/05 Leading white space used to be stripped from $spam_report which
3137 wrecked the formatting. Now it is preserved.
3139 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
3140 that they are available at delivery time.
3142 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
3144 TF/08 TLS error reporting now respects the incoming_interface and
3145 incoming_port log selectors.
3147 TF/09 Produce a more useful error message if an SMTP transport's hosts
3148 setting expands to an empty string.
3150 NM/06 Bugzilla 744: EXPN did not work under TLS.
3151 Patch provided by Phil Pennock.
3153 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
3154 Patch provided by Richard Godbee.
3156 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
3157 acl_smtp_notquit, added index entry.
3159 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
3160 Patch provided by Eugene Bujak.
3162 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
3163 accept(). Patch provided by Maxim Dounin.
3165 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
3166 Patch provided by Phil Pennock.
3168 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
3170 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
3171 Patch provided by Brad "anomie" Jorsch.
3173 NM/14 Bugzilla 622: Added timeout setting to transport filter.
3174 Patch provided by Dean Brooks.
3176 TK/05 Add native DKIM support (does not depend on external libraries).
3178 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
3179 Patch provided by Graeme Fowler.
3181 NM/16 Bugzilla 851: Documentation example syntax fix.
3183 NM/17 Changed NOTICE file to remove references to embedded PCRE.
3185 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
3188 NM/19 Bugzilla 745: TLS version reporting.
3189 Patch provided by Phil Pennock.
3191 NM/20 Bugzilla 167: bool: condition support.
3192 Patch provided by Phil Pennock.
3194 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
3195 clients. Patch provided by Phil Pennock.
3197 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
3198 Patch provided by Brad "anomie" Jorsch.
3200 NM/23 Bugzilla 687: Fix misparses in eximstats.
3201 Patch provided by Heiko Schlittermann.
3203 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
3204 Patch provided by Heiko Schlittermann.
3206 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
3207 plus update to original patch.
3209 NM/26 Bugzilla 799: Documentation correction for ratelimit.
3211 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
3212 Patch provided by David Brownlee.
3214 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
3216 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
3218 NM/30 Bugzilla 888: TLS documentation bugfixes.
3220 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
3222 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
3223 Unlike the original bugzilla I have changed all shell scripts in src tree.
3225 NM/33 Bugzilla 898: Transport filter timeout fix.
3226 Patch by Todd Rinaldo.
3228 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
3229 Patch by Serge Demonchaux.
3231 NM/35 Bugzilla 39: Base64 decode bug fixes.
3232 Patch by Jakob Hirsch.
3234 NM/36 Bugzilla 909: Correct connect() call in dcc code.
3236 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
3238 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
3240 NM/39 Bugzilla 911: Fixed MakeLinks build script.
3246 TK/01 Add preliminary DKIM support. Currently requires a forked version of
3247 ALT-N's libdkim that I have put here:
3248 http://duncanthrax.net/exim-experimental/
3250 Note to Michael Haardt: I had to rename some vars in sieve.c. They
3251 were called 'true' and it seems that C99 defines that as a reserved
3252 keyword to be used with 'bool' variable types. That means you could
3253 not include C99-style headers which use bools without triggering
3254 build errors in sieve.c.
3256 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
3257 as mailq or other aliases. Changed the --help handling significantly
3258 to do whats expected. exim_usage() emits usage/help information.
3260 SC/01 Added the -bylocaldomain option to eximstats.
3262 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
3264 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
3266 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
3272 PH/01 Another patch from the Sieve maintainer.
3274 PH/02 When an IPv6 address is converted to a string for single-key lookup
3275 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
3276 dots are used instead of colons so that keys in lsearch files need not
3277 contain colons. This was done some time before quoting was made available
3278 in lsearch files. However, iplsearch files do require colons in IPv6 keys
3279 (notated using the quote facility) so as to distinguish them from IPv4
3280 keys. This meant that lookups for IP addresses in host lists did not work
3281 for iplsearch lookups.
3283 This has been fixed by arranging for IPv6 addresses to be expressed with
3284 colons if the lookup type is iplsearch. This is not incompatible, because
3285 previously such lookups could never work.
3287 The situation is now rather anomalous, since one *can* have colons in
3288 ordinary lsearch keys. However, making the change in all cases is
3289 incompatible and would probably break a number of configurations.
3291 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
3294 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
3295 conversion specification without a maximum field width, thereby enabling
3296 a rogue spamd server to cause a buffer overflow. While nobody in their
3297 right mind would setup Exim to query an untrusted spamd server, an
3298 attacker that gains access to a server running spamd could potentially
3299 exploit this vulnerability to run arbitrary code as the Exim user.
3301 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
3302 $primary_hostname instead of what libspf2 thinks the hosts name is.
3304 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
3305 a directory entry by the name of the lookup key. Previously, if a
3306 symlink pointed to a non-existing file or a file in a directory that
3307 Exim lacked permissions to read, a lookup for a key matching that
3308 symlink would fail. Now it is enough that a matching directory entry
3309 exists, symlink or not. (Bugzilla 503.)
3311 PH/03 The body_linecount and body_zerocount variables are now exported in the
3314 PH/04 Added the $dnslist_matched variable.
3316 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
3317 This means they are set thereafter only if the connection becomes
3320 PH/06 Added the client_condition to authenticators so that some can be skipped
3321 by clients under certain conditions.
3323 PH/07 The error message for a badly-placed control=no_multiline_responses left
3324 "_responses" off the end of the name.
3326 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
3328 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
3329 (without spaces) instead of just copying the configuration text.
3331 PH/10 Added the /noupdate option to the ratelimit ACL condition.
3333 PH/11 Added $max_received_linelength.
3335 PH/12 Added +ignore_defer and +include_defer to host lists.
3337 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
3338 way in which PCRE > 7.0 is built.
3340 PH/14 Implemented queue_only_load_latch.
3342 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
3343 MAIL command. The effect was to mangle the value on 64-bit systems.
3345 PH/16 Another patch from the Sieve maintainer.
3347 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
3349 PH/18 If a system quota error occurred while trying to create the file for
3350 a maildir delivery, the message "Mailbox is full" was not appended to the
3351 bounce if the delivery eventually timed out. Change 4.67/27 below applied
3352 only to a quota excession during the actual writing of the file.
3354 PH/19 It seems that peer DN values may contain newlines (and other non-printing
3355 characters?) which causes problems in log lines. The DN values are now
3356 passed through string_printing() before being added to log lines.
3358 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
3359 and InterBase are left for another time.)
3361 PH/21 Added message_body_newlines option.
3363 PH/22 Guard against possible overflow in moan_check_errorcopy().
3365 PH/23 POSIX allows open() to be a macro; guard against that.
3367 PH/24 If the recipient of an error message contained an @ in the local part
3368 (suitably quoted, of course), incorrect values were put in $domain and
3369 $local_part during the evaluation of errors_copy.
3375 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
3376 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
3379 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
3380 issue a MAIL command.
3382 PH/02 In an ACL statement such as
3384 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
3386 if a client was not listed at all, or was listed with a value other than
3387 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
3388 the condition was not true (as it should be), so access was not denied.
3389 The bug was that the ! inversion was incorrectly passed on to the second
3390 item. This has been fixed.
3392 PH/03 Added additional dnslists conditions == and =& which are different from
3393 = and & when the dns lookup returns more than one IP address.
3395 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
3396 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
3398 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
3399 FSYNC, which compiles an option called disable_fsync that allows for
3400 bypassing fsync(). The documentation is heavily laced with warnings.
3402 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
3404 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
3405 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
3406 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
3407 including adding "make clean"; (3) Added -fPIC when compiling the test
3408 dynamically loaded module, to get rid of a warning.
3410 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
3411 message fails, move_frozen_messages = true and ignore_bounce_errors_after
3412 = 0s. The bug is otherwise harmless.
3414 PH/07 There was a bug in the dovecot authenticator such that the value of
3415 $auth1 could be overwritten, and so not correctly preserved, after a
3416 successful authentication. This usually meant that the value preserved by
3417 the server_setid option was incorrect.
3419 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
3421 PH/09 Installed PCRE release 7.0.
3423 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
3424 run for batched SMTP input. It is now run at the start of every message
3425 in the batch. While fixing this I discovered that the process information
3426 (output by running exiwhat) was not always getting set for -bs and -bS
3427 input. This is fixed, and it now also says "batched" for BSMTP.
3429 PH/11 Added control=no_pipelining.
3431 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
3432 patch, slightly modified), and move the expansion of helo_data till after
3433 the connection is made in the smtp transport (so it can use these
3436 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
3438 PH/14 Added log_selector = +pid.
3440 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
3442 PH/16 Add ${if forany and ${if forall.
3444 PH/17 Added dsn_from option to vary the From: line in DSNs.
3446 PH/18 Flush SMTP output before performing a callout, unless control =
3447 no_callout_flush is set.
3449 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
3450 was true (the default) a successful delivery failed to delete the retry
3451 item, thus causing premature timeout of the address. The bug is now
3454 PH/20 Added hosts_avoid_pipelining to the smtp transport.
3456 PH/21 Long custom messages for fakedefer and fakereject are now split up
3457 into multiline responses in the same way that messages for "deny" and
3458 other ACL rejections are.
3460 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
3461 with slight modification.
3463 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
3464 draft, changing the syntax and factoring some duplicate code".
3466 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
3467 for deliveries of the second and subsequent messages over the same SMTP
3470 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
3471 ${reduce, with only minor "tidies".
3473 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
3475 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
3476 expansion side effects.
3478 PH/27 When a message times out after an over-quota error from an Exim-imposed
3479 quota, the bounce message says "mailbox is full". This message was not
3480 being given when it was a system quota that was exceeded. It now should
3483 MH/03 Made $recipients available in local_scan(). local_scan() already has
3484 better access to the recipient list through recipients_list[], but
3485 $recipients can be useful in postmaster-provided expansion strings.
3487 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
3488 in the case of a MAIL command with additional options following the
3489 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
3490 were accidentally chopped off.
3492 PH/29 SMTP synchronization checks are implemented when a command is read -
3493 there is a check that no more input is waiting when there shouldn't be
3494 any. However, for some commands, a delay in an ACL can mean that it is
3495 some time before the response is written. In this time, more input might
3496 arrive, invalidly. So now there are extra checks after an ACL has run for
3497 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
3498 pipelining has not been advertised.
3500 PH/30 MH's patch to allow iscntrl() characters to be list separators.
3502 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
3503 returned in the SMTP response when smtp_return_error_details was false.
3504 This has been fixed.
3506 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
3507 instead of the C I/O that was originally supplied, because problems were
3508 reported on Solaris.
3510 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
3511 Exim which did not show up earlier: it was assuming that a call to
3512 SSL_CTX_set_info_callback() might give an error value. In fact, there is
3513 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
3514 was a macro that became an assignment, so it seemed to work. This has
3515 changed to a proper function call with a void return, hence the compile
3516 error. Exim's code has been fixed.
3518 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
3521 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
3523 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
3525 PH/37 If a message is not accepted after it has had an id assigned (e.g.
3526 because it turns out to be too big or there is a timeout) there is no
3527 "Completed" line in the log. When some messages of this type were
3528 selected by exigrep, they were listed as "not completed". Others were
3529 picked up by some special patterns. I have improved the selection
3530 criteria to be more general.
3532 PH/38 The host_find_failed option in the manualroute router can now be set
3533 to "ignore", to completely ignore a host whose IP address cannot be
3534 found. If all hosts are ignored, the behaviour is controlled by the new
3535 host_all_ignored option.
3537 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
3538 homing or because of multiple MX records with /mx) generated more than
3539 one IP address, and the following item turned out to be the local host,
3540 all the secondary addresses of the first item were incorrectly removed
3541 from the list, along with the local host and any following hosts (which
3542 is what is supposed to happen).
3544 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
3545 whoever called Exim into the -H file. In the case of the daemon it was
3546 behaving confusingly. When first started, it used values for whoever
3547 started the daemon, but after a SIGHUP it used the Exim user (because it
3548 calls itself on a restart). I have changed the code so that it now always
3551 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
3552 message are rejected with the same error (e.g. no authentication or bad
3553 sender address), and a DATA command is nevertheless sent (as can happen
3554 with PIPELINING or a stupid MUA), the error message that was given to the
3555 RCPT commands is included in the rejection of the DATA command. This is
3556 intended to be helpful for MUAs that show only the final error to their
3559 PH/42 Another patch from the Sieve maintainer.
3561 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
3562 Eximstats - Fixed some broken HTML links and added missing column headers
3564 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
3565 columns for Rejects, Temp Rejects, Ham, and Spam rows.
3567 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
3569 PH/43 Yet another patch from the Sieve maintainer.
3571 PH/44 I found a way to check for a TCP/IP connection going away before sending
3572 the response to the final '.' that terminates a message, but only in the
3573 case where the client has not sent further data following the '.'
3574 (unfortunately, this is allowed). However, in many cases there won't be
3575 any further data because there won't be any more messages to send. A call
3576 to select() can be used: if it shows that the input is "ready", there is
3577 either input waiting, or the socket has been closed. An attempt to read
3578 the next input character can distinguish the two cases. Previously, Exim
3579 would have sent an OK response which the client would never have see.
3580 This could lead to message repetition. This fix should cure that, at
3581 least in a lot of common cases.
3583 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
3584 advertised in response to EHLO.
3590 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
3591 fixed by 4.65/MH/01 (is this a record?) are fixed:
3593 (i) An empty string was always treated as zero by the numeric comparison
3594 operators. This behaviour has been restored.
3596 (ii) It is documented that the numeric comparison operators always treat
3597 their arguments as decimal numbers. This was broken in that numbers
3598 starting with 0 were being interpreted as octal.
3600 While fixing these problems I realized that there was another issue that
3601 hadn't been noticed. Values of message_size_limit (both the global option
3602 and the transport option) were treated as octal if they started with 0.
3603 The documentation was vague. These values are now always treated as
3604 decimal, and I will make that clear in the documentation.
3610 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
3611 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
3614 MH/01 Don't check that the operands of numeric comparison operators are
3615 integers when their expansion is in "skipping" mode (fixes bug
3616 introduced by 4.64-PH/07).
3618 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
3619 child addresses, Exim now panics and dies. Previously, because the count
3620 is held in a short int, deliveries were likely to be lost. As such a
3621 large number of recipients for a single message is ridiculous
3622 (performance will be very, very poor), I have chosen to impose a limit
3623 rather than extend the field.
3629 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
3630 leftover -K file (the existence of which was triggered by #402).
3631 While we were at it, introduced process PID as part of the -K
3632 filename. This should rule out race conditions when creating
3635 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
3636 processing considerably. Previous code took too long for large mails,
3637 triggering a timeout which in turn triggers #401.
3639 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
3640 in the DK code in transports.c. sendfile() is not really portable,
3641 hence the _LINUX specificness.
3643 TF/01 In the add_headers option to the mail command in an Exim filter,
3644 there was a bug that Exim would claim a syntax error in any
3645 header after the first one which had an odd number of characters
3648 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
3649 callout verification, Exim cached a "reject" for the entire domain. This
3650 is correct for most verifications, but it is not correct for a recipient
3651 verification with use_sender or use_postmaster set, because in that case
3652 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
3653 case of MAIL FROM:<> rejection from other early rejections (e.g.
3654 rejection of HELO). When verifying a recipient using a non-null MAIL
3655 address, the cache is ignored if it shows MAIL FROM:<> rejection.
3656 Whatever the result of the callout, the value of the domain cache is
3657 left unchanged (for any other kind of callout, getting as far as trying
3658 RCPT means that the domain itself is ok).
3660 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
3663 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
3664 manifest itself as EPIPE rather than ECONNECT. When tidying away a
3665 session, the daemon ignores ECONNECT errors and logs others; it now
3666 ignores EPIPE as well.
3668 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
3669 (quoted-printable decoding).
3671 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
3672 later the small subsequent patch to fix an introduced bug.
3674 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
3676 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
3678 PH/08 An error is now given if message_size_limit is specified negative.
3680 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
3681 to be given (somewhat) arbitrary names.
3683 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
3686 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
3687 miscellaneous code fixes
3689 PH/10 Added the log_reject_target ACL modifier to specify where to log
3692 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
3693 hostname. This is wrong, because it relates to the incoming message (and
3694 probably the interface on which it is arriving) and not to the outgoing
3695 callout (which could be using a different interface). This has been
3696 changed to use the value of the helo_data option from the smtp transport
3697 instead - this is what is used when a message is actually being sent. If
3698 there is no remote transport (possible with a router that sets up host
3699 addresses), $smtp_active_hostname is used.
3701 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
3702 tweaks were necessary in order to get it to work (see also 21 below):
3703 (a) The code assumed that strncpy() returns a negative number on buffer
3704 overflow, which isn't the case. Replaced with Exim's string_format()
3706 (b) There were several signed/unsigned issues. I just did the minimum
3707 hacking in of casts. There is scope for a larger refactoring.
3708 (c) The code used strcasecmp() which is not a standard C function.
3709 Replaced with Exim's strcmpic() function.
3710 (d) The code set only $1; it now sets $auth1 as well.
3711 (e) A simple test gave the error "authentication client didn't specify
3712 service in request". It would seem that Dovecot has changed its
3713 interface. Fortunately there's a specification; I followed it and
3714 changed what the client sends and it appears to be working now.
3716 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
3719 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
3720 address A is aliased to B and C, where B exists and C does not. Without
3721 -v the output is "A verified" because verification stops after a
3722 successful redirection if more than one address is generated. However,
3723 with -v the child addresses are also verified. Exim was outputting "A
3724 failed to verify" and then showing the successful verification for C,
3725 with its parentage. It now outputs "B failed to verify", showing B's
3726 parentage before showing the successful verification of C.
3728 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
3729 look up a TXT record in a specific list after matching in a combined
3732 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
3733 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
3734 they consult the DNS. I had assumed they would set it the way they
3735 wanted; and indeed my experiments on Linux seem to show that in some
3736 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
3737 To be on the safe side, however, I have now made the interface to
3738 host_find_byname() similar to host_find_bydns(), with an argument
3739 containing the DNS resolver options. The host_find_byname() function now
3740 sets these options at its start, just as host_find_bydns() does. The smtp
3741 transport options dns_qualify_single and dns_search_parents are passed to
3742 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
3743 of host_find_byname() use the default settings of RES_DEFNAMES
3744 (qualify_single) but not RES_DNSRCH (search_parents).
3746 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
3747 spool_read_header() do less string testing, by means of a preliminary
3748 switch on the second character of optional "-foo" lines. (This is
3749 overdue, caused by the large number of possibilities that now exist.
3750 Originally there were few.) While I was there, I also converted the
3751 str(n)cmp tests so they don't re-test the leading "-" and the first
3752 character, in the hope this might squeeze out yet more improvement.
3754 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
3755 flag allowing group syntax was set by the header_syntax check but not
3756 turned off, possible causing trouble later; (2) The flag was not being
3757 set at all for the header_verify test, causing "group"-style headers to
3758 be rejected. I have now set it in this case, and also caused header_
3759 verify to ignore an empty address taken from a group. While doing this, I
3760 came across some other cases where the code for allowing group syntax
3761 while scanning a header line wasn't quite right (mostly, not resetting
3762 the flag correctly in the right place). These bugs could have caused
3763 trouble for malformed header lines. I hope it is now all correct.
3765 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
3766 with the "reply" argument non-NULL. The code, however (which originally
3767 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
3768 but it didn't always do it. This confused somebody who was copying the
3769 code for some other use. I have removed all the tests.
3771 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
3772 feature that was used to support insecure browsers during the U.S. crypto
3773 embargo. It requires special client support, and Exim is probably the
3774 only MTA that supported it -- and would never use it because real RSA is
3775 always available. This code has been removed, because it had the bad
3776 effect of slowing Exim down by computing (never used) parameters for the
3777 RSA_EXPORT functionality.
3779 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
3780 authenticator to fail if there's a tab character in the incoming data
3781 (there should never be unless someone is messing about, as it's supposed
3782 to be base64-encoded). Also added, on Timo's advice, the "secured" option
3783 if the connection is using TLS or if the remote IP is the same as the
3784 local IP, and the "valid-client-cert option" if a client certificate has
3787 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
3788 authenticators. This can be used for authorization after authentication
3789 succeeds. (In the case of plaintext, it servers for both authentication
3792 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
3793 if any retry times were supplied.
3795 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
3796 connection, where there is no client IP address to check. In this
3797 situation, the verify now always succeeds.
3799 PH/25 Applied John Jetmore's -Mset patch.
3801 PH/26 Added -bem to be like -Mset, but loading a message from a file.
3803 PH/27 In a string expansion for a processed (not raw) header when multiple
3804 headers of the same name were present, leading whitespace was being
3805 removed from all of them, but trailing whitespace was being removed only
3806 from the last one. Now trailing whitespace is removed from each header
3807 before concatenation. Completely empty headers in a concatenation (as
3808 before) are ignored.
3810 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
3811 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
3813 PH/29 [Removed. This was a change that I later backed out, and forgot to
3814 correct the ChangeLog entry (that I had efficiently created) before
3815 committing the later change.]
3817 PH/30 Exim was sometimes attempting to deliver messages that had suffered
3818 address errors (4xx response to RCPT) over the same connection as other
3819 messages routed to the same hosts. Such deliveries are always "forced",
3820 so retry times are not inspected. This resulted in far too many retries
3821 for the affected addresses. The effect occurred only when there were more
3822 hosts than the hosts_max_try setting in the smtp transport when it had
3823 the 4xx errors. Those hosts that it had tried were not added to the list
3824 of hosts for which the message was waiting, so if all were tried, there
3825 was no problem. Two fixes have been applied:
3827 (i) If there are any address or message errors in an SMTP delivery, none
3828 of the hosts (tried or untried) are now added to the list of hosts
3829 for which the message is waiting, so the message should not be a
3830 candidate for sending over the same connection that was used for a
3831 successful delivery of some other message. This seems entirely
3832 reasonable: after all the message is NOT "waiting for some host".
3833 This is so "obvious" that I'm not sure why it wasn't done
3834 previously. Hope I haven't missed anything, but it can't do any
3835 harm, as the worst effect is to miss an optimization.
3837 (ii) If, despite (i), such a delivery is accidentally attempted, the
3838 routing retry time is respected, so at least it doesn't keep
3839 hammering the server.
3841 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
3842 in ${readsocket because some servers need this prod.
3844 PH/32 Added some extra debug output when updating a wait-xxx database.
3846 PH/33 The hint "could be header name not terminated by colon", which has been
3847 given for certain expansion errors for a long time, was not being given
3848 for the ${if def:h_colon_omitted{... case.
3850 PH/34 The spec says: "With one important exception, whenever a domain list is
3851 being scanned, $domain contains the subject domain." There was at least
3852 one case where this was not true.
3854 PH/35 The error "getsockname() failed: connection reset by peer" was being
3855 written to the panic log as well as the main log, but it isn't really
3856 panic-worthy as it just means the connection died rather early on. I have
3857 removed the panic log writing for the ECONNRESET error when getsockname()
3860 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
3861 runs only) independently of the message's sender address. This meant
3862 that, if the 4xx error was in fact related to the sender, a different
3863 message to the same recipient with a different sender could confuse
3864 things. In particular, this can happen when sending to a greylisting
3865 server, but other circumstances could also provoke similar problems.
3866 I have changed the default so that the retry time for these errors is now
3867 based a combination of the sender and recipient addresses. This change
3868 can be overridden by setting address_retry_include_sender=false in the
3871 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
3872 remote server are returned as part of bounce messages. This was not
3873 happening for LMTP over a pipe (the lmtp transport), but now it is the
3874 same for both kinds of LMTP.
3876 PH/38 Despite being documented as not happening, Exim was rewriting addresses
3877 in header lines that were in fact CNAMEs. This is no longer the case.
3879 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
3880 and queue runs started by the daemon processed all messages. This has
3881 been fixed so that -R and -S can now usefully be given with -q<time>.
3883 PH/40 Import PCRE release 6.7 (fixes some bugs).
3885 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
3887 PH/42 Give an error if -q is specified more than once.
3889 PH/43 Renamed the variables $interface_address and $interface_port as
3890 $received_ip_address and $received_port, to make it clear that these
3891 values apply to message reception, and not to the outgoing interface when
3892 a message is delivered. (The old names remain recognized, of course.)
3894 PH/44 There was no timeout on the connect() call when using a Unix domain
3895 socket in the ${readsocket expansion. There now is.
3897 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
3898 be meaningful with "accept".
3900 SC/01 Eximstats V1.43
3901 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
3903 SC/02 Eximstats V1.44
3904 Use a glob alias rather than an array ref in the generated
3905 parser. This improves both readability and performance.
3907 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
3908 Collect SpamAssassin and rejection statistics.
3909 Don't display local sender or destination tables unless
3910 there is data to show.
3911 Added average volumes into the top table text output.
3913 SC/04 Eximstats V1.46
3914 Collect data on the number of addresses (recipients)
3915 as well as the number of messages.
3917 SC/05 Eximstats V1.47
3918 Added 'Message too big' to the list of mail rejection
3919 reasons (thanks to Marco Gaiarin).
3921 SC/06 Eximstats V1.48
3922 Mainlog lines which have GMT offsets and are too short to
3923 have a flag are now skipped.
3925 SC/07 Eximstats V1.49 (Alain Williams)
3926 Added the -emptyok flag.
3928 SC/08 Eximstats V1.50
3929 Fixes for obtaining the IP address from reject messages.
3931 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
3932 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
3933 whitespace changes from 4.64-PH/27
3935 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
3938 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
3939 are found, allow negative numbers in numeric criteria)
3941 JJ/06 exipick.20061117.2, added new $message_body_missing variable
3943 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
3944 to match changes made in 4.64-PH/43
3946 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
3948 PH/47 Put in an explicit test for a DNS lookup of an address record where the
3949 "domain" is actually an IP address, and force a failure. This locks out
3950 those revolvers/nameservers that support "A-for-A" lookups, in
3951 contravention of the specifications.
3953 PH/48 When a host name was looked up from an IP address, and the subsequent
3954 forward lookup of the name timed out, the host name was left in
3955 $sender_host_name, contrary to the specification.
3957 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
3958 restricted to single-key lookups, Exim was not diagnosing an error if
3959 * or *@ was used with a query-style lookup.
3961 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
3963 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
3964 long ago, but noone interested enough thought of it. Let's just say that
3965 the "1.1" means that there are some new functions that weren't there at
3966 some point in the past.
3968 PH/51 Error processing for expansion failure of helo_data from an smtp
3969 transport during callout processing was broken.
3971 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3972 tested/used via the -bh/-bhc/-bs options.
3974 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3975 bug, fixed in subsequent PCRE releases).
3977 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3978 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3980 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3986 SC/01 Use a glob alias rather than an array ref in eximstats generated
3987 parser. This improves both readability and performance.
3989 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3990 Don't display local sender or destination tables in eximstats unless
3991 there is data to show.
3992 Added average volumes into the eximstats top table text output.
3994 SC/03 Collect data on the number of addresses (recipients) as well
3995 as the number of messages in eximstats.
3997 TF/01 Correct an error in the documentation for the redirect router. Exim
3998 does (usually) call initgroups() when daemonizing.
4000 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
4001 with consistent privilege compared to when running as a daemon.
4003 TF/03 Note in the spec that $authenticated_id is not set for local
4004 submissions from trusted users.
4006 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
4007 Thanks to Dean Brooks <dean@iglou.com> for the patch.
4009 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
4010 by adding some example configuration directives to the default
4011 configuration file. A little bit of work is required to uncomment the
4012 directives and define how usernames and passwords are checked, but
4013 there is now a framework to start from.
4015 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
4016 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
4017 without this. I don't know how relevant this is to other LDAP libraries.
4019 PH/02 Add the verb name to the "unknown ACL verb" error.
4021 PH/03 Magnus Holmgren's patch for filter_prepend_home.
4023 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
4025 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
4026 directory not expanded when it should be if an expanded home directory
4027 was set for the address (which is overridden by the transport).
4029 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
4032 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
4033 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
4034 because it is too late at that time, and has no effect.
4036 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
4037 security issue with \' (bugzilla #107). I could not use the
4038 PQescapeStringConn() function, because it needs a PGconn value as one of
4041 PH/08 When testing addresses using -bt, indicate those final addresses that
4042 are duplicates that would not cause an additional delivery. At least one
4043 person was confused, thinking that -bt output corresponded to deliveries.
4044 (Suppressing duplicates isn't a good idea as you lose the information
4045 about possibly different redirections that led to the duplicates.)
4047 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
4048 systems where poll() doesn't work, in particular OS X.
4050 PH/10 Added more information to debugging output for retry time not reached.
4052 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
4053 operations in malware.c.
4055 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
4058 PH/13 If write_rejectlog was set false when logging was sent to syslog with
4059 syslog_duplication set false, log lines that would normally be written
4060 both the the main log and to the reject log were not written to syslog at
4063 PH/14 In the default configuration, change the use of "message" in ACL warn
4064 statements to "add_header".
4066 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
4067 not followed by a command (e.g. "seen endif").
4069 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
4070 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
4073 PH/17 Added extra conditions to the default value of delay_warning_condition
4077 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
4078 { match{$h_precedence:}{(?i)bulk|list|junk} } \
4079 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
4082 The Auto-Submitted: and various List- headers are standardised, whereas I
4083 don't think Precedence: ever was.
4085 PH/18 Refactored debugging code in route_finduser() to show more information,
4086 in particular, the error code if getpwnam() issues one.
4088 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
4089 This is apparently needed in addition to the PH/07 change above to avoid
4090 any possible encoding problems.
4092 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
4093 but not after initializing Perl.
4095 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
4096 output them only if debugging. By default they are written stderr,
4097 apparently, which is not desirable.
4099 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
4102 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
4105 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
4107 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
4108 authenticated or an ident call has been made. Suppress the default
4109 values for $authenticated_id and $authenticated_sender (but permit -oMai
4110 and -oMas) when testing with -bh.
4112 PH/24 Re-jigged the order of the tests in the default configuration so that the
4113 tests for valid domains and recipients precede the DNS black list and CSA
4114 tests, on the grounds that those ones are more expensive.
4116 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
4117 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
4118 This bug exists in every version of Exim that I still have, right back to
4121 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
4122 However, an attempt to turn on case-sensitivity in a regex key by
4123 including (?-i) didn't work because the subject string was already
4124 lowercased, and the effects were non-intuitive. It turns out that a
4125 one-line patch can be used to allow (?-i) to work as expected.
4131 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
4132 other effects) broke the use of negated acl sub-conditions.
4134 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
4137 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
4138 "Deny" causes Exim to reject the incoming connection with a 554 error.
4139 Unfortunately, if there is a major crisis, such as a disk failure,
4140 tcp-wrappers gives "deny", whereas what one would like would be some
4141 kind of temporary error. A kludge has been added to help with this.
4142 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
4143 554 error is used if errno is still zero or contains ENOENT (which occurs
4144 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
4147 PH/03 Add -lutil to the default FreeBSD LIBS setting.
4149 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
4150 errors. Otherwise a message that provokes a temporary error (when other
4151 messages do not) can cause a whole host to time out.
4153 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
4154 addresses were routed directly to files or pipes from a redirect router.
4155 File deliveries just didn't batch; pipe deliveries might have suffered
4158 PH/06 A failure to get a lock for a hints database would erroneously always say
4159 "Failed to get write lock", even when it was really a read lock.
4161 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
4162 of 0600. This has been changed to use the value of the lockfile_mode
4163 option (which defaults to 0600).
4165 PH/08 Applied small patch from the Sieve maintainer.
4167 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
4168 folder from quota calculations, a direct delivery into this folder messed
4169 up the contents of the maildirsize file. This was because the regex was
4170 used only to exclude .Trash (or whatever) when the size of the mailbox
4171 was calculated. There was no check that a delivery was happening into an
4172 excluded directory. This bug has been fixed by ignoring all quota
4173 processing for deliveries into excluded directories.
4175 PH/10 Added the maildirfolder_create_regex option to appendfile.
4181 PH/01 The code for finding all the local interface addresses on a FreeBSD
4182 system running IPv6 was broken. This may well have applied to all BSD
4183 systems, as well as to others that have similar system calls. The broken
4184 code found IPv4 interfaces correctly, but gave incorrect values for the
4185 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
4186 that it would not match correctly against @[] and not recognize the IPv6
4189 PH/02 The ipliteral router was not recognizing addresses of the form user@
4190 [ipv6:....] because it didn't know about the "ipv6:" prefix.
4192 PH/03 Added disable_ipv6.
4194 PH/04 Changed $reply_address to use the raw form of the headers instead of the
4195 decoded form, because it is most often used to construct To: headers
4196 lines in autoreplies, and the decoded form may well be syntactically
4197 invalid. However, $reply_address has leading white space removed, and all
4198 newlines turned into spaces so that the autoreply transport does not
4201 PH/05 If group was specified without a user on a router, and no group or user
4202 was specified on a transport, the group from the router was ignored.
4204 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
4205 for visible compile-time settings that can be used to change these
4206 numbers, for those that want even more. Backwards compatibility with old
4207 spool files has been maintained. However, going back to a previous Exim
4208 release will lost any variables that are in spool files.
4210 PH/07 Two small changes when running in the test harness: increase delay when
4211 passing a TCP/IP connection to a new process, in case the original
4212 process has to generate a bounce, and remove special handling of
4213 127.0.0.2 (sic), which is no longer necessary.
4215 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
4216 be the same on different OS.
4218 PH/09 Moved a debug statement in filter processing to avoid a race problem when
4221 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
4222 whether --show-vars was specified or not
4224 JJ/02 exipick: Added support for new ACL variable spool format introduced
4227 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
4228 syntactically invalid From: or Reply-to: line, and a filter used this to
4229 generate an autoreply, and therefore failed to obtain an address for the
4230 autoreply, Exim could try to deliver to a non-existent relative file
4231 name, causing unrelated and misleading errors. What now happens is that
4232 it logs this as a hard delivery error, but does not attempt to create a
4235 PH/11 The exinext utility has a -C option for testing purposes, but although
4236 the given file was scanned by exinext itself; it wasn't being passed on
4237 when Exim was called.
4239 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
4240 an end-of-file indication when reading a command response.
4242 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
4243 compiled. In many other places in Exim, IPv6 addresses are always
4244 recognized, so I have changed this. It also means that IPv4 domain
4245 literals of the form [IPV4:n.n.n.n] are now always recognized.
4247 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
4248 used if the router is not running as root, for example, when verifying at
4249 ACL time, or when using -bh. The debugging output from this situation was
4250 non-existent - all you got was a failure to exec. I have made two
4253 (a) Failures to set uid/gid, the current directory, or a process leader
4254 in a subprocess such as that created by queryprogram now generate
4255 suitable debugging output when -d is set.
4257 (b) The queryprogram router detects when it is not running as root,
4258 outputs suitable debugging information if -d is set, and then runs
4259 the subprocess without attempting to change uid/gid.
4261 PH/15 Minor change to Makefile for building test_host (undocumented testing
4264 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
4265 additional section of a DNS packet that returns MX or SRV records.
4266 Instead, it always explicitly searches for A/AAAA records. This avoids
4267 major problems that occur when a DNS server includes only records of one
4268 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
4269 fixed another bug: if SRV records were looked up and the corresponding
4270 address records were *not* found in the additional section, the port
4271 values from the SRV records were lost.
4273 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
4274 using the correct key (the original address) when searching the retry
4275 rules in order to find which one to use for generating the retry hint.
4277 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
4278 adding the default one. Similarly, if it contains a Reply-To: header, the
4279 errors_reply_to option, if set, is not used.
4281 PH/19 When calculating a retry time, Exim used to measure the "time since
4282 failure" by looking at the "first failed" field in the retry record. Now
4283 it does not use this if it is later than than the arrival time of the
4284 message. Instead it uses the arrival time. This makes for better
4285 behaviour in cases where some deliveries succeed, thus re-setting the
4286 "first failed" field. An example is a quota failure for a huge message
4287 when small messages continue to be delivered. Without this change, the
4288 "time since failure" will always be short, possible causing more frequent
4289 delivery attempts for the huge message than are intended.
4290 [Note: This change was subsequently modified - see PH/04 for 4.62.]
4292 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
4293 $1, $2, $3) because the numerical variables can be reset during some
4294 expansion items (e.g. "match"), thereby losing the authentication data.
4296 PH/21 Make -bV show the size of off_t variables so that the test suite can
4297 decide whether to run tests for quotas > 2G.
4299 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
4300 mailbox_size, and mailbox_filecount in the appendfile transport. If a
4301 filecount value is greater than 2G or if a quota value is greater than 2G
4302 on a system where the size of off_t is not greater than 4, a panic error
4305 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
4306 never match. The debug and -bh output now contains an explicit error
4307 message indicating a malformed IPv4 address or mask.
4309 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
4310 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
4311 PH/23 above applies.
4313 PH/25 Do not write to syslog when running in the test harness. The only
4314 occasion when this arises is a failure to open the main or panic logs
4315 (for which there is an explicit test).
4317 PH/26 Added the /no_tell option to "control=freeze".
4319 PH/27 If a host name lookup failed very early in a connection, for example, if
4320 the IP address matched host_lookup and the reverse lookup yielded a name
4321 that did not have a forward lookup, an error message of the form "no IP
4322 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
4323 could be logged. Now it outputs the IP address instead of "NULL".
4325 PH/28 An enabling patch from MH: add new function child_open_exim2() which
4326 allows the sender and the authenticated sender to be set when
4327 submitting a message from within Exim. Since child_open_exim() is
4328 documented for local_scan(), the new function should be too.
4330 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
4331 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
4332 results in an empty string is now treated as unset.
4334 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
4336 PH/31 Added sender_verify_fail logging option.
4338 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
4339 needed by RFC 822 but not by RFC 2822 was commented out. I have now
4340 tidied the source and removed it altogether.
4342 PH/33 When a queue run was abandoned because the load average was too high, a
4343 log line was always written; now it is written only if the queue_run log
4344 selector is set. In addition, the log line for abandonment now contains
4345 information about the queue run such as the pid. This is always present
4346 in "start" and "stop" lines but was omitted from the "abandon" line.
4348 PH/34 Omit spaces between a header name and the colon in the error message that
4349 is given when verify = headers_syntax fails (if there are lots of them,
4350 the message gets confusing).
4352 PH/35 Change the default for dns_check_names_pattern to allow slashes within
4353 names, as there are now some PTR records that contain slashes. This check
4354 is only to protect against broken name servers that fall over on strange
4355 characters, so the fact that it applies to all lookups doesn't matter.
4357 PH/36 Now that the new test suite is complete, we can remove some of the
4358 special code in Exim that was needed for the old test suite. For example,
4359 sorting DNS records because real resolvers return them in an arbitrary
4360 order. The new test suite's fake resolver always returns records in the
4363 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
4364 bounces) except when queue_only is set, to avoid logging races between
4365 the different processes.
4367 PH/38 Panic-die if .include specifies a non-absolute path.
4369 PH/39 A tweak to the "H" retry rule from its user.
4371 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
4372 a label. They prevented compilation on older perls.
4374 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
4375 a warning to be raised on newish perls.
4377 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
4378 on queue. Changes to match documented behaviour of showing count of
4379 messages matching specified criteria.
4381 PH/40 Changed the default ident timeout from 30s to 5s.
4383 PH/41 Added support for the use of login_cap features, on those BSD systems
4384 that have them, for controlling the resources used by pipe deliveries.
4386 PH/42 The content-scanning code uses fopen() to create files in which to put
4387 message data. Previously it was not paying any attention to the mode of
4388 the files. Exim runs with umask(0) because the rest of the code creates
4389 files with open(), and sets the required mode explicitly. Thus, these
4390 files were ending up world-writeable. This was not a big issue, because,
4391 being within the spool directory, they were not world-accessible. I have
4392 created a function called modefopen, which takes an additional mode
4393 argument. It sets umask(777), creates the file, chmods it to the required
4394 mode, then resets the umask. All the relevant calls to fopen() in the
4395 content scanning code have been changed to use this function.
4397 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
4398 to 24 hours. This avoids potential overflow problems when processing G
4399 and H retry rules. I suspect nobody ever tinkers with this value.
4401 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
4403 PH/45 When the plaintext authenticator is running as a client, the server's
4404 challenges are checked to ensure they are valid base64 strings. By
4405 default, the authentication attempt is cancelled if an invalid string is
4406 received. Setting client_ignore_invalid_base64 true ignores these errors.
4407 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
4408 they are received. Thus, the responses can be made to depend on the
4409 challenges. If an invalid string is ignored, an empty string is placed in
4412 PH/46 Messages that are created by the autoreply transport now contains a
4413 References: header, in accordance with RFCs 2822 and 3834.
4415 PH/47 Added authenticated_sender_force to the smtp transport.
4417 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
4419 PH/49 Installed latest patch from the Sieve maintainer.
4421 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
4422 was also set, the appendfile transport was unnecessarily scanning a
4423 directory of message files (e.g. for maildir delivery) to find the count
4424 of files (along with the size), even though it did not need this
4425 information. It now does the scan only if it needs to find either the
4426 size of the count of files.
4428 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
4430 PH/52 Two bugs concerned with error handling when the smtp transport is
4433 (i) Exim was not creating retry information for temporary errors given
4434 for individual recipients after the DATA command when the smtp transport
4435 was used in LMTP mode. This meant that they could be retried too
4436 frequently, and not timed out correctly.
4438 (ii) Exim was setting the flag that allows error details to be returned
4439 for LMTP errors on RCPT commands, but not for LMTP errors for individual
4440 recipients that were returned after the DATA command.
4442 PH/53 This is related to PH/52, but is more general: for any failing address,
4443 when detailed error information was permitted to be returned to the
4444 sender, but the error was temporary, then after the final timeout, only
4445 "retry timeout exceeded" was returned. Now it returns the full error as
4446 well as "retry timeout exceeded".
4448 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
4449 do this, and (what is worse) MTAs that accept it.
4451 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
4452 will now be deprecated.
4454 PH/56 New os.c-cygwin from the Cygwin maintainer.
4456 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
4457 formats (previously only available in exim formats via -bpr, -bpru,
4458 and -bpra. Now also available in native and exiqgrep formats)
4460 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
4461 with very large, slow to parse queues
4463 JJ/08 exipick: added ! as generic prefix to negate any criteria format
4465 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
4467 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
4468 responses to authentication challenges, though it was showing the
4469 challenges; (ii) I've removed the CR characters from the debug output for
4472 PH/58 Allow for the insertion of a newline as well as a space when a string
4473 is turned into more than one encoded-word during RFC 2047 encoding. The
4474 Sieve code now uses this.
4476 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
4477 data_4xx, lost_connection, tls_required.
4479 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
4480 message was being sent as an SMTP response.
4482 PH/61 Add -l and -k options to exicyclog.
4484 PH/62 When verifying, if an address was redirected to one new address, so that
4485 verification continued, and the new address failed or deferred after
4486 having set something in $address_data, the value of $address_data was not
4487 passed back to the ACL. This was different to the case when no
4488 redirection occurred. The value is now passed back in both cases.
4490 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
4491 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
4492 use login_cap.h, so on its own it isn't the distinguishing feature. The
4493 new name refers directly to the setclassresources() function.
4495 PH/65 Added configuration files for NetBSD3.
4497 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
4499 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
4500 is preferred over IPv4.
4502 PH/68 The bounce_return_message and bounce_return_body options were not being
4503 honoured for bounces generated during the reception of non-SMTP messages.
4504 In particular, this applied to messages rejected by the ACL. This bug has
4505 been fixed. However, if bounce_return_message is true and bounce_return_
4506 body is false, the headers that are returned for a non-SMTP message
4507 include only those that have been read before the error was detected.
4508 (In the case of an ACL rejection, they have all been read.)
4510 PH/69 The HTML version of the specification is now built in a directory called
4511 spec_html instead of spec.html, because the latter looks like a path with
4512 a MIME-type, and this confuses some software.
4514 PH/70 Catch two compiler warnings in sieve.c.
4516 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
4517 function verify_get_ident() calls ip_connect() to connect a socket, but
4518 if the "connect()" function timed out, ip_connect() used to close the
4519 socket. However, verify_get_ident() also closes the socket later, and in
4520 between Exim writes to the log, which may get opened at this point. When
4521 the socket was closed in ip_connect(), the log could get the same file
4522 descriptor number as the socket. This naturally causes chaos. The fix is
4523 not to close the socket in ip_connect(); the socket should be closed by
4524 the function that creates it. There was only one place in the code where
4525 this was missing, in the iplookup router, which I don't think anybody now
4526 uses, but I've fixed it anyway.
4528 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
4529 well as to direct DNS lookups. Otherwise the handling of names in host
4530 lists is inconsistent and therefore confusing.
4536 PH/01 Two changes to the default runtime configuration:
4538 (1) Move the checks for relay_from_hosts and authenticated clients from
4539 after to before the (commented out) DNS black list checks.
4541 (2) Add control=submission to the relay_from_hosts and authenticated
4542 clients checks, on the grounds that messages accepted by these
4543 statements are most likely to be submissions.
4545 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
4547 (1) Generate an error if the third argument for the ${prvs expansion is
4550 (2) Treat a missing third argument of ${prvscheck as if it were an empty
4553 (3) Reset the variables that are obtained from the first argument of
4554 ${prvscheck and used in the second argument before leaving the code,
4555 because their memory is reclaimed, so using them afterwards may do
4558 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
4559 one (it's much easier than Tom thought :-).
4561 (5) Because of (4), we can now allow for the use of $prvscheck_result
4562 inside the third argument.
4564 PH/03 For some reason, the default setting of PATH when running a command from
4565 a pipe transport was just "/usr/bin". I have changed it to
4568 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
4569 anything to be listed in the output from -bV.
4571 PH/05 When a filter generated an autoreply, the entire To: header line was
4572 quoted in the delivery log line, like this:
4574 => >A.N.Other <ano@some.domain> <original@ddress> ...
4576 This has been changed so that it extracts the operative address. There
4577 may be more than one such address. If so, they are comma-separated, like
4580 => >ano@some.domain,ona@other.domain <original@ddress> ...
4582 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
4583 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
4584 not being looked up in the rDNS to get a host name, Exim was showing the
4585 IP address twice in Received: lines, even though the IP addresses were
4586 identical. For example:
4588 Received: from [1.2.3.4] (helo=[1.2.3.4])
4590 However, if the real host name was known, it was omitting the HELO data
4591 if it matched the actual IP address. This has been tidied up so that it
4592 doesn't show the same IP address twice.
4594 PH/07 When both +timestamp and +memory debugging was on, the value given by
4595 $tod_xxx expansions could be wrong, because the tod_stamp() function was
4596 called by the debug printing, thereby overwriting the timestamp buffer.
4597 Debugging no longer uses the tod_stamp() function when +timestamp is set.
4599 PH/08 When the original message was included in an autoreply transport, it
4600 always said "this is a copy of the message, including all the headers",
4601 even if body_only or headers_only was set. It now gives an appropriate
4604 PH/09 Applied a patch from the Sieve maintainer which:
4606 o fixes some comments
4607 o adds the (disabled) notify extension core
4608 o adds some debug output for the result of if/elsif tests
4609 o points to the current vacation draft in the documentation
4610 and documents the missing references header update
4614 o fixes a bug in processing the envelope test (when testing
4615 multiple envelope elements, the last element determined the
4618 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
4619 Electronic Mail") by including:
4621 Auto-submitted: auto-generated
4623 in the messages that it generates (bounce messages and others, such as
4624 warnings). In the case of bounce messages for non-SMTP messages, there was
4625 also a typo: it was using "Auto_submitted" (underscore instead of
4626 hyphen). Since every message generated by Exim is necessarily in response
4627 to another message, thes have all been changed to:
4629 Auto-Submitted: auto-replied
4631 in accordance with these statements in the RFC:
4633 The auto-replied keyword:
4635 - SHOULD be used on messages sent in direct response to another
4636 message by an automatic process,
4638 - MUST NOT be used on manually-generated messages,
4640 - MAY be used on Delivery Status Notifications (DSNs) and Message
4641 Disposition Notifications (MDNs),
4643 - MUST NOT be used on messages generated by automatic or periodic
4644 processes, except for messages which are automatic responses to
4647 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
4648 to the default Received: header definition.
4650 PH/12 Added log selector acl_warn_skipped (default on).
4652 PH/13 After a successful wildlsearch lookup, discard the values of numeric
4653 variables because (a) they are in the wrong storage pool and (b) even if
4654 they were copied, it wouldn't work properly because of the caching.
4656 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
4657 checking when decoding. Apparently there are clients that generate
4658 overlong encoded strings. Why am I not surprised?
4660 PH/15 If the first argument of "${if match_address" was not empty, but did not
4661 contain an "@" character, Exim crashed. Now it writes a panic log message
4662 and treats the condition as false.
4664 PH/16 In autoreply, treat an empty string for "once" the same as unset.
4666 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
4667 extension "envelope-auth". The code is finished and in agreement with
4668 other implementations, but there is no documentation so far and in fact,
4669 nobody wrote the draft yet. This extension is currently #undef'ed, thus
4670 not changing the active code.
4672 Print executed "if" and "elsif" statements when debugging is used. This
4673 helps a great deal to understand what a filter does.
4675 Document more things not specified clearly in RFC3028. I had all this
4676 sorted out, when out of a sudden new issues came to my mind. Oops."
4678 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
4681 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
4682 canonical form (as documented). However, after a host name lookup from
4683 the IP address, check_host() was doing a simple string comparison with
4684 addresses acquired from the DNS when checking that the found name did
4685 have the original IP as one of its addresses. Since any found IPv6
4686 addresses are likely to be in abbreviated form, the comparison could
4687 fail. Luckily, there already exists a function for doing the comparison
4688 by converting both addresses to binary, so now that is used instead of
4689 the text comparison.
4691 PH/20 There was another similar case to PH/19, when a complete host name was
4692 given in a host list; looking up its IP address could give an abbreviated
4693 form, whereas the current host's name might or might not be abbreviated.
4694 The same fix has been applied.
4700 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
4701 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
4704 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
4705 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
4707 PH/03 Typo: missing ".o" in src/pcre/Makefile.
4709 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
4710 header line, restrict the check to what is listed in RFCs 2369 and 2929.
4711 Also, for "Auto-Submitted", treat anything other than "no" as
4712 non-personal, in accordance with RFC 3834. (Previously it treated
4713 anything starting "auto-" as non-personal.)
4715 TF/01 The control=submission/name=... option had a problem with syntax
4716 errors if the name included a slash character. The /name= option
4717 now slurps the rest of the string, so it can include any characters
4718 but it must come last in the list of options (after /sender_retain
4721 PH/05 Some modifications to the interface to the fake nameserver for the new
4729 TK/01 Added the "success_on_redirect" address verification option. See
4730 NewStuff for rationale and an example.
4732 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
4734 PH/02 Patch to exigrep to allow it to work on syslog lines.
4736 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
4737 fread() to skip over the body file's header line, because in Cygwin the
4738 header line is locked and is inaccessible.
4740 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
4741 co-exist for some time) to make it clear that it is the Exim ID that is
4742 referenced, not the Message-ID: header line.
4744 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
4745 string_format() function, because snprintf() does not exist on all
4748 PH/06 The use of forbid_filter_existstest now also locks out the use of the
4749 ${stat: expansion item.
4751 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
4752 protocol synchronization error", to keep the pedants happy.
4754 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
4755 well as for IRIX systems, when gcc is being used. See the host.c source
4758 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
4760 PH/10 Named domain lists were not working if used in a queue_smtp_domains
4763 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
4764 transport and to the smtp transport in LMTP mode.
4766 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
4768 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
4769 run a filter in a subprocess. This could lead to confusion in subsequent
4770 lookups in the parent process. There should also be a search_tidyup() at
4771 the end of the subprocess.
4773 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
4774 only if the host matched helo_try_verify_hosts, which caused the
4775 verification to occur when the EHLO/HELO command was issued. The ACL just
4776 tested the remembered result. Now, if a previous verification attempt has
4777 not happened, "verify = helo" does it there and then.
4779 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
4781 TK/03 Fix log output including CR from clamd.
4783 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
4784 exist provoked a memory error which could cause a segfault.
4786 PH/15 Installed PCRE 6.2
4788 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
4790 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
4791 of the problem. Specifically, suggested +O2 rather than +O1 for the
4794 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
4796 PH/20 If a delivery was routed to a non-standard port by means of an SRV
4797 record, the port was not correctly logged when the outgoing_port log
4798 selector was set (it logged the transort's default port).
4800 PH/21 Added support for host-specific ports to manualroute, queryprogram,
4801 fallback_hosts, and "hosts" in the smtp transport.
4803 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
4804 host errors such as "Connection refused".
4806 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
4807 authentication with radiusclient 0.4.9:
4809 - Error returned from rc_read_config was caught wrongly
4810 - Username/password not passed on to radius server due to wrong length.
4812 The presumption is that some radiusclient API changes for 4.51/PH/17
4813 were not taken care of correctly. The code is still untested by me (my
4814 Linux distribution still has 0.3.2 of radiusclient), but it was
4815 contributed by a Radius user.
4817 PH/24 When doing a callout, the value of $domain wasn't set correctly when
4818 expanding the "port" option of the smtp transport.
4820 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
4821 while reading a MIME header. Thanks to Tom Hughes for a patch.
4823 PH/24 Include config.h inside local_scan.h so that configuration settings are
4826 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
4827 that in an ACL for RCPT (for example), you can examine exactly what was
4830 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
4831 commands, but it was not correctly comparing the address with the actual
4832 client host address. Thus, it would show the EHLO address in Received:
4833 header lines when this was not necessary.
4835 PH/27 Added the % operator to ${eval:}.
4837 PH/28 Exim tries to create and chdir to its spool directory when it starts;
4838 it should be ignoring failures (because with -C, for example, it has lost
4839 privilege). It wasn't ignoring creation failures other than "already
4842 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
4845 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
4846 because an input line was too long, either on its own, or by virtue of
4847 too many continuations, the temporary file was not being removed, and the
4848 return code was incorrect.
4850 PH/31 Missing "BOOL" in function definition in filtertest.c.
4852 PH/32 Applied Sieve patches from the maintainer.
4854 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
4856 PH/33 Added "verify = not_blind".
4858 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
4859 Local/Makefile (with some defaults set). These are used in built scripts
4860 such as exicyclog, but they have never been used in the exim_install
4861 script (though there are many overriding facilities there). I have
4862 arranged that the exim_install script now takes note of these two
4865 PH/35 Installed configuration files for Dragonfly.
4867 PH/36 When a locally submitted message by a trusted user did not contain a
4868 From: header, and the sender address was obtained from -f or from an SMTP
4869 MAIL command, and the trusted user did not use -F to supply a sender
4870 name, $originator_name was incorrectly used when constructing a From:
4871 header. Furthermore, $originator_name was used for submission mode
4872 messages from external hosts without From: headers in a similar way,
4873 which is clearly wrong.
4875 PH/37 Added control=suppress_local_fixups.
4877 PH/38 When log_selector = +received_sender was set, and the addition of the
4878 sender made the log line's construction buffer exactly full, or one byte
4879 less than full, an overflow happened when the terminating "\n" was
4882 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
4883 when the result of a list match is failure because a DNS lookup failed.
4885 PH/40 RM_COMMAND is now used in the building process.
4887 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
4888 the "build-* directories that it finds.
4890 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
4891 address in a domain literal was a prefix of an interface address.
4893 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
4894 when verifying a sender address, unless rewrite_headers is false.
4896 PH/44 Wrote a long comment about why errors_to addresses are verified as
4897 recipients, not senders.
4899 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
4900 the ratelimit ACL was added.
4902 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
4904 PH/46 Added extra information about PostgreSQL errors to the error string.
4906 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
4907 avoiding the need to install special zones in a real server. This is
4908 backwards compatible; if it can't find the fake resolver, it drops back.
4909 Thus, both old and new test suites can be run.
4911 TF/02 Added util/ratelimit.pl
4913 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
4914 clock is set back in time.
4916 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
4917 Candler <B.Candler@pobox.com>.
4919 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
4920 OK for addresses that are the result of redirections.
4922 PH/48 A number of further additions for the benefit of the new test suite,
4923 including a fake gethostbyname() that interfaces to the fake DNS resolver
4926 TF/06 The fix for widen_domains has also been applied to qualify_single and
4927 search_parents which are the other dnslookup options that can cause
4930 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
4933 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
4935 TF/07 Exim produced the error message "an SRV record indicated no SMTP
4936 service" if it encountered an MX record with an empty target hostname.
4937 The message is now "an MX or SRV record indicated no SMTP service".
4939 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
4940 if the DNS of the sending site is misconfigured. This is quite a
4941 common situation. This change restores the behaviour of treating a
4942 helo verification defer as a failure.
4944 PH/51 If self=fail was set on a router, the bounce message did not include the
4945 actual error message.
4951 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
4953 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
4954 command itself ended in error, the underlying message about the transport
4955 filter timeout was being overwritten with the pipe command error. Now the
4956 underlying error message should be appended to the second error message.
4958 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
4960 PH/02 Reduce the amount of output that "make" produces by default. Full output
4961 can still be requested.
4963 PH/03 The warning log line about a condition test deferring for a "warn" verb
4964 was being output only once per connection, rather than after each
4965 occurrence (because it was using the same function as for successful
4966 "warn" verbs). This seems wrong, so I have changed it.
4968 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4969 it should not have, which might have caused a crash in the right
4970 circumstances, but probably never did.
4972 PH/04 Installed a modified version of Tony Finch's patch to make submission
4973 mode fix the return path as well as the Sender: header line, and to
4974 add a /name= option so that you can make the user's friendly name appear
4977 TF/03 Added the control = fakedefer ACL modifier.
4979 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4980 Mark Lowes for thorough testing.
4982 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4984 TK/03 Merged latest SRS patch from Miles Wilton.
4986 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4987 with the definition in sysexits.h (which is #included earlier).
4988 Fortunately, Exim does not actually use EX_OK. The code used to try to
4989 preserve the sysexits.h value, by assuming that macro definitions were
4990 scanned for macro replacements. I have been disabused of this notion,
4991 so now the code just undefines EX_OK before #including unistd.h.
4993 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4994 in the smtp transport. When a block could not be written in a single
4995 write() function, the timeout was being re-applied to each part-write.
4996 This seems wrong - if the receiver was accepting one byte at a time it
4997 would take for ever. The timeout is now adjusted when this happens. It
4998 doesn't have to be particularly precise.
5000 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
5001 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
5003 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
5004 without a domain if the check to <postmaster@domain> fails.
5006 SC/01 Eximstats: added -xls and the ability to specify output files
5007 (patch written by Frank Heydlauf).
5009 SC/02 Eximstats: use FileHandles for outputting results.
5011 SC/03 Eximstats: allow any combination of xls, txt, and html output.
5013 SC/04 Eximstats: fixed display of large numbers with -nvr option
5015 SC/05 Eximstats: fixed merging of reports with empty tables.
5017 SC/06 Eximstats: added the -include_original_destination flag
5019 SC/07 Eximstats: removed tabs and trailing whitespace.
5021 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
5023 TK/06 MBOX spool code: Add real "From " MBOX separator line
5024 so the .eml file is really in mbox format (even though
5025 most programs do not really care). Patch from Alex Miller.
5027 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
5028 The latter is generated from $received_to and is only set if the
5029 message has one envelope recipient. SA can use these headers,
5030 obviously out-of-the-box. Patch from Alex Miller.
5032 PH/08 The ${def test on a variable was returning false if the variable's
5033 value was "0", contrary to what the specification has always said!
5034 The result should be true unless the variable is empty.
5036 PH/09 The syntax error of a character other than { following "${if
5037 def:variable_name" (after optional whitespace) was not being diagnosed.
5038 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
5039 accidental colon was present, for example, could give incorrect results.
5041 PH/10 Tidied the code in a number of places where the st_size field of a stat()
5042 result is used (not including appendfile, where other changes are about
5045 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
5046 This involved changing a lot of size variables from int to off_t. It
5047 should work with maildirs and everything.
5049 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
5050 spamd dying while we are connected to it.
5052 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
5055 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
5056 function for BDB 4.3.
5058 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
5060 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
5061 the sources for PCRE have been re-arranged and more files are now
5064 PH/15 The code I had for printing potentially long long variables in PH/11
5065 above was not the best (it lost precision). The length of off_t variables
5066 is now inspected at build time, and an appropriate printing format (%ld
5067 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
5068 to be "long long int" or "long int". This is needed for the internal
5069 formatting function string_vformat().
5071 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
5072 the configuration file to be ":syslog", then the script "guesses" where
5073 the logs files are, rather than using the compiled in default. In our
5074 case the guess is not the same as the compiled default, so the script
5075 suddenly stopped working when I started to use syslog. The patch checks
5076 to see if log_file_path is "". If so, it attempts to read it from exim
5077 with no configuration file to get the compiled in version, before it
5078 falls back to the previous guessing code."
5080 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
5081 implementing BATV in an Exim configuration. See NewStuff for the gory
5084 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
5085 Makefile that are specific to HP-UX.
5087 PH/18 If the "use_postmaster" option was set for a recipient callout together
5088 with the "random" option, the postmaster address was used as the MAIL
5089 FROM address for the random test, but not for the subsequent recipient
5090 test. It is now used for both.
5092 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
5093 patch removes a few documentation additions to RFC 3028, because the
5094 latest draft now contains them. It adds the new en;ascii-case comparator
5095 and a new error check for 8bit text in MIME parts. Comparator and
5096 require names are now matched exactly. I enabled the subaddress
5097 extension, but it is not well tested yet (read: it works for me)."
5099 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
5100 rework some of the code of TK/09 above to avoid the hardwired use of
5101 "%lld" and "long long". Replaced the call to snprintf() with a call to
5104 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
5105 records point to non-existent hosts", "retry timeout exceeded", and
5106 "retry time not reached for any host after a long failure period".
5108 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
5109 experimental DomainKeys support:
5111 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
5112 (2) On an error such as an illegally used "control", the wrong name for
5113 the control was given.
5115 These problems did NOT occur unless DomainKeys support was compiled.
5117 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
5119 PH/24 Added ${if match_ip condition.
5121 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
5122 databases so that it will be absolutely obvious if a crash occurs in the
5123 DB library. This is a regular occurrence (often caused by mis-matched
5126 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
5127 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
5128 picked up on a user's system that detects such things. There doesn't seem
5129 to be a gcc warning option for this - only an attribute that has to be
5130 put on the function's prototype. It seems that in Fedora Core 4 they have
5131 set this on a number of new functions. No doubt there will be more in due
5134 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
5135 specify a transport. However, if an address that was verified by such a
5136 router was the subject of a callout, Exim crashed because it tried to
5137 read the rcpt_include_affixes from the non-existent transport. Now it
5138 just assumes that the setting of that option is false. This bug was
5139 introduced by 4.51/PH/31.
5141 PH/28 Changed -d+all to exclude +memory, because that information is very
5142 rarely of interest, but it makes the output a lot bigger. People tend to
5143 do -d+all out of habit.
5145 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
5146 code in os-type was giving problems when libc.so lives in lib64, like on
5149 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
5150 aren't the modern standard, and it seems that some systems' include files
5151 don't always have them. Exim was already checking for some of the newer
5152 ones like T_AAAA, and defining it itself. I've added checks for all the
5153 record types that Exim uses.
5155 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
5156 not automatically generating a new one, as it is supposed to. This
5157 prevented TLS from working. If the file did exist, but contained invalid
5158 data, a new version was generated, as expected. It was only the case of a
5159 non-existent file that was broken.
5161 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
5162 with a change in libdomainkeys > 0.64.
5164 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
5165 from DNS. If the selector record carries the flag, it now has
5166 precedence over the domain-wide flag.
5168 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
5170 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
5171 the use of a port name that isn't defined in /etc/services) occurred, the
5172 message was deferred as in a normal delivery, and thus remained on the
5173 spool, instead of being failed because of the mua_wrapper setting. This
5174 is now fixed, and I tidied up some of the mua_wrapper messages at the
5177 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
5178 the messages in a hash of arrays rather than using individual hashes.
5179 This is a bit cleaner and results in dramatic memory savings, albeit
5180 at a slight CPU cost.
5182 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
5183 as requested by Marc Sherman.
5185 SC/10 Eximstats: added histograms for user specified patterns as requested
5188 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
5190 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
5191 fopen() in the content-scanning modules that did not already have it.
5197 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
5198 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
5200 TK/02 Fix ACL "control" statement not being available in MIME ACL.
5202 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
5204 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
5205 to test Sieve filters that use "vacation".
5207 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
5208 that changes the way the GnuTLS parameters are stored in the cache file.
5209 The new format can be generated externally. For backward compatibility,
5210 if the data in the cache doesn't make sense, Exim assumes it has read an
5211 old-format file, and it generates new data and writes a new file. This
5212 means that you can't go back to an older release without removing the
5215 PH/03 A redirect router that has both "unseen" and "one_time" set does not
5216 work if there are any delivery delays because "one_time" forces the
5217 parent to be marked "delivered", so its unseen clone is never tried
5218 again. For this reason, Exim now forbids the simultaneous setting of
5221 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
5222 redirected to themselves ("homonym" addresses). Read the long ChangeLog
5223 entry if you want to know the details. The fix, however, neglected to
5224 consider the case when local delivery batching is involved. The test for
5225 "previously delivered" was not happening when checking to see if an
5226 address could be batched with a previous (undelivered) one; under
5227 certain circumstances this could lead to multiple deliveries to the same
5230 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
5231 in its include files, and this causes problems building Exim.
5233 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
5234 header_syntax) but Exim was just ignoring anything given after a slash.
5235 In particular, this caused confusion with an attempt to use "verify =
5236 reverse_host_lookup/defer_ok". An error is now given when options are
5237 supplied for verify items that do not have them. (Maybe reverse_host_
5238 lookup should have a defer_ok option, but that's a different point.)
5240 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
5241 defined by RFC 821) to 2048, because there were problems with some AUTH
5242 commands, and RFC 1869 says the size should be increased for extended
5243 SMTP commands that take arguments.
5245 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
5248 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
5249 "unknown" error; now it says that the functionality isn't in the binary.
5251 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
5252 an address' error message when a string expansion fails (syntax or
5253 whatever). Otherwise the password may appear in the log. Following change
5254 PH/42 below, there is no longer a chance of it appearing in a bounce
5257 PH/11 Installed exipick version 20050225.0 from John Jetmore.
5259 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
5260 of its addresses was ever tried. (Bugzilla bug #2.)
5262 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
5263 the result incorrectly in the debug output. (It correctly added a newline
5264 to what was transported.)
5266 TF/01 Added $received_time.
5268 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
5269 commented out examples of how to interface to a virus scanner and to
5270 SpamAssassin. Also added commented examples of av_scanner and
5271 spamd_address settings.
5273 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
5274 and controls are allowed in which ACLs. There were a couple of minor
5275 errors. Some of the entries in the conditions table (which is a table of
5276 where they are NOT allowed) were getting very unwieldy; rewrote them as a
5277 negation of where the condition IS allowed.
5279 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
5281 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
5282 header file does not have a version number, so I've had to invent a new
5283 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
5284 API. The code is untested by me (my Linux distribution still has 0.3.2 of
5285 radiusclient), but it was contributed by a Radius user.
5287 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
5288 files or directories, for OpenSSL.
5290 PH/19 When an Exim process that is running as root has to create an Exim log
5291 file, it does so in a subprocess that runs as exim:exim so as to get the
5292 ownership right at creation (otherwise, other Exim processes might see
5293 the file with the wrong ownership). There was no test for failure of this
5294 fork() call, which would lead to the process getting stuck as it waited
5295 for a non-existent subprocess. Forks do occasionally fail when resources
5296 run out. I reviewed all the other calls to fork(); they all seem to check
5299 PH/20 When checking for unexpected SMTP input at connect time (before writing
5300 the banner), Exim was not dealing correctly with a non-positive return
5301 from the read() function. If the client had disconnected by this time,
5302 the result was a log entry for a synchronization error with an empty
5303 string after "input=" when read() returned zero. If read() returned -1
5304 (an event I could not check), uninitialized data bytes were printed.
5305 There were reports of junk text (parts of files, etc) appearing after
5308 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
5310 PH/22 Added support for macro redefinition, and (re)definition in between
5311 driver and ACL definitions.
5313 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
5314 forgetting to use the resulting value; it was using the unexpanded value.
5316 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
5317 hadn't been configured. The fix is from Juergen Kreileder, who
5318 understands it better than I do:
5320 "Here's what I see happening with three configured cyrus_sasl
5321 authenticators configured (plain, login, cram-md5):
5323 On startup auth_cyrus_sasl_init() gets called for each of these.
5324 This means three calls to sasl_listmech() without a specified mech_list.
5325 => SASL tests which mechs of all available mechs actually work
5326 => three warnings about OTP not working
5327 => the returned list contains: plain, login, cram-md5, digest-md5, ...
5329 With the patch, sasl_listmech() also gets called three times. But now
5330 SASL's mech_list option is set to the server_mech specified in the the
5331 authenticator. Or in other words, the answer from sasl_listmech()
5332 gets limited to just the mech you're testing for (which is different
5334 => the return list contains just 'plain' or 'login', 'cram-md5' or
5335 nothing depending on the value of ob->server_mech.
5337 I've just tested the patch: Authentication still works fine,
5338 unavailable mechs specified in the exim configuration are still
5339 caught, and the auth.log warnings about OTP are gone."
5341 PH/25 When debugging is enabled, the contents of the command line are added
5342 to the debugging output, even when log_selector=+arguments is not
5345 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
5346 answer is "GNU", and only if the return is "GNU/something" is the answer
5349 PH/27 $acl_verify_message is now set immediately after the failure of a
5350 verification in an ACL, and so is available in subsequent modifiers. In
5351 particular, the message can be preserved by coding like this:
5353 warn !verify = sender
5354 set acl_m0 = $acl_verify_message
5356 Previously, $acl_verify_message was set only while expanding "message"
5357 and "log_message" when a very denied access.
5359 PH/28 Modified OS/os.c-Linux with
5361 -#ifndef OS_LOAD_AVERAGE
5362 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
5364 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
5365 nomenclature these days.)
5367 PH/29 Installed patch from the Sieve maintainer that adds the options
5368 sieve_useraddress and sieve_subaddress to the redirect router.
5370 PH/30 In these circumstances:
5371 . Two addresses routed to the same list of hosts;
5372 . First host does not offer TLS;
5373 . First host accepts first address;
5374 . First host gives temporary error to second address;
5375 . Second host offers TLS and a TLS session is established;
5376 . Second host accepts second address.
5377 Exim incorrectly logged both deliveries with the TLS parameters (cipher
5378 and peerdn, if requested) that were in fact used only for the second
5381 PH/31 When doing a callout as part of verifying an address, Exim was not paying
5382 attention to any local part prefix or suffix that was matched by the
5383 router that accepted the address. It now behaves in the same way as it
5384 does for delivery: the affixes are removed from the local part unless
5385 rcpt_include_affixes is set on the transport.
5387 PH/32 Add the sender address, as F=<...>, to the log line when logging a
5388 timeout during the DATA phase of an incoming message.
5390 PH/33 Sieve envelope tests were broken for match types other than :is. I have
5391 applied a patch sanctioned by the Sieve maintainer.
5393 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
5394 the uid or gid is negative. A case of a negative gid caused this to be
5395 noticed. The fix allows for either to be negative.
5397 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
5398 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
5401 PH/36 The change PH/12 above was broken. Fixed it.
5403 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
5404 the grounds that routing the same address twice would always produce the
5405 same answer. This might have been true once, but it is certainly no
5406 longer true now. Routing a child address may depend on the previous
5407 routing that produced that child. Some complicated redirection strategies
5408 went wrong when messages had multiple recipients, and made Exim's
5409 behaviour dependent on the order in which the addresses were given.
5411 I have moved the duplicate checking until after the routing is complete.
5412 Exim scans the addresses that are assigned to local and remote
5413 transports, and removes any duplicates. This means that more work will be
5414 done, as duplicates will always all be routed, but duplicates are
5415 presumably rare, so I don't expect this is of any significance.
5417 For deliveries to pipes, files, and autoreplies, the duplicate checking
5418 still happens during the routing process, since they are not going to be
5421 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
5422 It corrects a timeout issue with spamd. This is Ian's comment: "The
5423 background is that sometimes spamd either never reads data from a
5424 connection it has accepted, or it never writes response data. The exiscan
5425 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
5426 blindly assumes that writes won't block so it may never time out."
5428 PH/39 Allow G after quota size as well as K and M.
5430 PH/40 The value set for $authenticated_id in an authenticator may not contain
5431 binary zeroes or newlines because the value is written to log lines and
5432 to spool files. There was no check on this. Now the value is run through
5433 the string_printing() function so that such characters are converted to
5434 printable escape sequences.
5436 PH/41 $message_linecount is a new variable that contains the total number of
5437 lines in the message. Compare $body_linecount, which is the count for the
5440 PH/42 Exim no longer gives details of delivery errors for specific addresses in
5441 bounce and delay warning messages, except in certain special cases, which
5444 (a) An SMTP error message from a remote host;
5445 (b) A message specified in a :fail: redirection;
5446 (c) A message specified in a "fail" command in a system filter;
5447 (d) A message specified in a FAIL return from the queryprogram router;
5448 (e) A message specified by the cannot_route_message router option.
5450 In these cases only, Exim does include the error details in bounce and
5451 warning messages. There are also a few cases where bland messages such
5452 as "unrouteable address" or "local delivery error" are given.
5454 PH/43 $value is now also set for the "else" part of a ${run expansion.
5456 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
5457 being worked on, but at least Exim now implements the latest version to
5460 PH/45 In a pipe transport, although a timeout while waiting for the pipe
5461 process to complete was treated as a delivery failure, a timeout while
5462 writing the message to the pipe was logged, but erroneously treated as a
5463 successful delivery. Such timeouts include transport filter timeouts. For
5464 consistency with the overall process timeout, these timeouts are now
5465 treated as errors, giving rise to delivery failures by default. However,
5466 there is now a new Boolean option for the pipe transport called
5467 timeout_defer, which, if set TRUE, converts the failures into defers for
5468 both kinds of timeout. A transport filter timeout is now identified in
5471 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
5472 systems where "make" and "gmake" are different, calling "gmake" at top
5473 level broke things. I've arranged for the value of $(MAKE) to be passed
5474 from the Makefile to this script so that it can call the same version of
5478 A note about Exim versions 4.44 and 4.50
5479 ----------------------------------------
5481 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
5482 changes of various kinds. As a consequence, a big documentation update was
5483 needed. This delayed the release for rather longer than seemed good, especially
5484 in the light of a couple of (minor) security issues. Therefore, the changes
5485 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
5486 release. So 4.44 and 4.50 are in effect two different branches that both start
5489 I have left the 4.50 change log unchanged; it contains all the changes since
5490 4.43. The change log for 4.44 is below; many of its items are identical to
5491 those for 4.50. This seems to be the most sensible way to preserve the
5492 historical information.
5498 1. Minor wording change to the doc/README.SIEVE file.
5500 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
5501 computation of the current number of files was incorrect.
5503 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5504 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5507 4. Give more explanation in the error message when the command for a transport
5508 filter fails to execute.
5510 5. There are several places where Exim runs a non-Exim command in a
5511 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5512 was being done only for the command run by the queryprogram router. It is
5513 now done for all such subprocesses. The other cases are: ${run, transport
5514 filters, and the commands run by the lmtp and pipe transports.
5516 6. Added CONFIGURE_GROUP build-time option.
5518 7. Some older OS have a limit of 256 on the maximum number of file
5519 descriptors. Exim was using setrlimit() to set 1000 as a large value
5520 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5521 systems. I've change it so that if it can't get 1000, it tries for 256.
5523 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5524 was an oversight, and furthermore, ever since the addition of extra
5525 controls (e.g. 4.43/32), the checks on when to allow different forms of
5526 "control" were broken. There should now be diagnostics for all cases when a
5527 control that does not make sense is encountered.
5529 9. Added the /retain_sender option to "control=submission".
5531 10. $recipients is now available in the predata ACL (oversight).
5533 11. Tidy the search cache before the fork to do a delivery from a message
5534 received from the command line. Otherwise the child will trigger a lookup
5535 failure and thereby defer the delivery if it tries to use (for example) a
5536 cached ldap connection that the parent has called unbind on.
5538 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5539 of $address_data from the recipient verification was clobbered by the
5540 sender verification.
5542 13. The value of address_data from a sender verification is now available in
5543 $sender_address_data in subsequent conditions in the ACL statement.
5545 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
5547 15. Added a new option "connect=<time>" to callout options, to set a different
5550 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5551 was its contents. (It was OK if the option was not defined at all.)
5553 17. A "Completed" log line is now written for messages that are removed from
5554 the spool by the -Mrm option.
5556 18. New variables $sender_verify_failure and $recipient_verify_failure contain
5557 information about exactly what failed.
5559 19. Added -dd to debug only the daemon process.
5561 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
5562 handles timeouts, both on the server side and network timeouts. Renamed the
5563 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
5565 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
5566 It is now set to "smtps".
5568 22. $host_address is now set to the target address during the checking of
5569 ignore_target_hosts.
5571 23. When checking ignore_target_hosts for an ipliteral router, no host name was
5572 being passed; this would have caused $sender_host_name to have been used if
5573 matching the list had actually called for a host name (not very likely,
5574 since this list is usually IP addresses). A host name is now passed as
5577 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5578 code that specifies a non-restarting handler (typically sigaction() in
5579 modern systems) in an attempt to fix a rare and obscure crash bug.
5581 25. Narrowed the window for a race in the daemon that could cause it to ignore
5582 SIGCHLD signals. This is not a major problem, because they are used only to
5583 wake it up if nothing else does.
5585 26. A malformed maildirsize file could cause Exim to calculate negative values
5586 for the mailbox size or file count. Odd effects could occur as a result.
5587 The maildirsize information is now recalculated if the size or filecount
5590 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5591 support for a long time. Removed HAVE_SYS_VFS_H.
5593 28. Installed the latest version of exipick from John Jetmore.
5595 29. In an address list, if the pattern was not a regular expression, an empty
5596 subject address (from a bounce message) matched only if the pattern was an
5597 empty string. Non-empty patterns were not even tested. This was the wrong
5598 because it is perfectly reasonable to use an empty address as part of a
5599 database query. An empty address is now tested by patterns that are
5600 lookups. However, all the other forms of pattern expect the subject to
5601 contain a local part and a domain, and therefore, for them, an empty
5602 address still always fails if the pattern is not itself empty.
5604 30. Exim went into a mad DNS loop when attempting to do a callout where the
5605 host was specified on an smtp transport, and looking it up yielded more
5606 than one IP address.
5608 31. Re-factored the code for checking spool and log partition space into a
5609 function that finds that data and another that does the check. The former
5610 is then used to implement four new variables: $spool_space, $log_space,
5611 $spool_inodes, and $log_inodes.
5613 32. The RFC2047 encoding function was originally intended for short strings
5614 such as real names; it was not keeping to the 75-character limit for
5615 encoded words that the RFC imposes. It now respects the limit, and
5616 generates multiple encoded words if necessary. To be on the safe side, I
5617 have increased the buffer size for the ${rfc2047: expansion operator from
5620 33. It is now permitted to omit both strings after an "if" condition; if the
5621 condition is true, the result is "true". As before, when the second string
5622 is omitted, a false condition yields an empty string. This makes it less
5623 cumbersome to write custom ACL and router conditions.
5625 34. Failure to deliver a bounce message always caused it to be frozen, even if
5626 there was an errors_to setting on the router. The errors_to setting is now
5629 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5630 canonical form (fully expanded) before being placed in
5631 $sender_host_address.
5633 36. The table in the code that translates DNS record types into text (T_A to
5634 "A" for instance) was missing entries for NS and CNAME. It is just possible
5635 that this could have caused confusion if both these types were looked up
5636 for the same domain, because the text type is used as part of Exim's
5637 per-process caching. But the chance of anyone hitting this buglet seems
5640 37. The dnsdb lookup has been extended in a number of ways.
5642 (1) There is a new type, "zns", which walks up the domain tree until it
5643 finds some nameserver records. It should be used with care.
5645 (2) There is a new type, "mxh", which is like "mx" except that it returns
5646 just the host names, not the priorities.
5648 (3) It is now possible to give a list of domains (or IP addresses) to be
5649 looked up. The behaviour when one of the lookups defers can be
5650 controlled by a keyword.
5652 (4) It is now possible to specify the separator character for use when
5653 multiple records are returned.
5655 38. The dnslists ACL condition has been extended: it is now possible to supply
5656 a list of IP addresses and/or domains to be looked up in a particular DNS
5659 39. Added log_selector=+queue_time_overall.
5661 40. When running the queue in the test harness, wait just a tad after forking a
5662 delivery process, to get repeatability of debugging output.
5664 41. Include certificate and key file names in error message when GnuTLS fails
5665 to set them up, because the GnuTLS error message doesn't include the name
5666 of the failing file when there is a problem reading it.
5668 42. Allow both -bf and -bF in the same test run.
5670 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
5672 44. The "Exiscan patch" is now merged into the mainline Exim source.
5674 45. Sometimes the final signoff response after QUIT could fail to get
5675 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5676 < 0 before doing a fflush(). This bug looks as though it goes back to the
5677 introduction of TLS in release 3.20, but "sometimes" must have been rare
5678 because the tests only now provoked it.
5680 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
5681 (this can affect the format of dates).
5683 47. exim_tidydb, when checking for the continued existence of a message for
5684 which it has found a message-specific retry record, was not finding
5685 messages that were in split spool directories. Consequently, it was
5686 deleting retry records that should have stayed in existence.
5688 48. Steve fixed some bugs in eximstats.
5690 49. The SPA authentication driver was not abandoning authentication and moving
5691 on to the next authenticator when an expansion was forced to fail,
5692 contradicting the general specification for all authenticators. Instead it
5693 was generating a temporary error. It now behaves as specified.
5695 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
5696 (the order specifies the preference for clients). The order is now AES256,
5697 AES128, 3DES, ARCFOUR128.
5699 51. Small patch to Sieve code - explicitly set From: when generating an
5702 52. Exim crashed if a remote delivery caused a very long error message to be
5703 recorded - for instance if somebody sent an entire SpamAssassin report back
5704 as a large number of 550 error lines. This bug was coincidentally fixed by
5705 increasing the size of one of Exim's internal buffers (big_buffer) that
5706 happened as part of the Exiscan merge. However, to be on the safe side, I
5707 have made the code more robust (and fixed the comments that describe what
5710 53. Now that there can be additional text after "Completed" in log lines (if
5711 the queue_time_overall log selector is set), a one-byte patch to exigrep
5712 was needed to allow it to recognize "Completed" as not the last thing in
5715 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
5716 patch that reportedly fixes this has been added. I am not expert enough to
5717 create a test for it. This is what the patch creator wrote:
5719 "I found a little strange behaviour of ldap code when working with
5720 Windows 2003 AD Domain, where users was placed in more than one
5721 Organization Units. When I tried to give exim partial DN, the exit code
5722 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
5723 But simultaneously result of request was absolutely normal ldap result,
5724 so I produce this patch..."
5726 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
5727 REFERENCE, so I have modified the code to exclude the patch when that macro
5730 55. Some experimental protocols are using DNS PTR records for new purposes. The
5731 keys for these records are domain names, not reversed IP addresses. The
5732 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5733 leaves it alone. Component reversal etc. now happens only for IP addresses.
5736 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
5738 57. Double the size of the debug message buffer (to 2048) so that more of very
5739 long debug lines gets shown.
5741 58. The exicyclog utility now does better if the number of log files to keep
5742 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
5744 59. Two changes related to the smtp_active_hostname option:
5746 (1) $smtp_active_hostname is now available as a variable.
5747 (2) The default for smtp_banner uses $smtp_active_hostname instead
5748 of $primary_hostname.
5750 60. The host_aton() function is supposed to be passed a string that is known
5751 to be a valid IP address. However, in the case of IPv6 addresses, it was
5752 not checking this. This is a hostage to fortune. Exim now panics and dies
5753 if the condition is not met. A case was found where this could be provoked
5754 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5755 components; fortuitously, this particular loophole had already been fixed
5756 by change 4.50/55 above.
5758 If there are any other similar loopholes, the new check in host_aton()
5759 itself should stop them being exploited. The report I received stated that
5760 data on the command line could provoke the exploit when Exim was running as
5761 exim, but did not say which command line option was involved. All I could
5762 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5763 running as the user.
5766 61. There was a buffer overflow vulnerability in the SPA authentication code
5767 (which came originally from the Samba project). I have added a test to the
5768 spa_base64_to_bits() function which I hope fixes it.
5771 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
5772 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
5774 63. The daemon start-up calls getloadavg() while still root for those OS that
5775 need the first call to be done as root, but it missed one case: when
5776 deliver_queue_load_max is set with deliver_drop_privilege. This is
5777 necessary for the benefit of the queue runner, because there is no re-exec
5778 when deliver_drop_privilege is set.
5780 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
5781 This has been fixed.
5783 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5784 was in use, was not putting the data itself into the right store pool;
5785 consequently, it could be overwritten for a subsequent message in the same
5786 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5789 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
5791 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
5792 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
5793 the function were treating the return as a boolean value, which happened to
5794 work because 0=false and not-0=true, but is not correct code.
5796 68. The host_aton() function was not handling scoped IPv6 addresses (those
5797 with, for example, "%eth0" on the end) correctly.
5799 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
5800 negated items (that is, ~something) in unsigned ints. Some compilers
5801 apparently mutter when there is no cast.
5803 70. If an address verification called from an ACL failed, and did not produce a
5804 user-specific message (i.e. there was only a "system" message), nothing was
5805 put in $acl_verify_message. In this situation, it now puts the system
5808 71. Change 4.23/11 added synchronization checking at the start of an SMTP
5809 session; change 4.31/43 added the unwanted input to the log line - except
5810 that it did not do this in the start of session case. It now does.
5812 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
5813 This is wrong and can cause the other end to generate a synchronization
5814 error if it is another Exim or anything else that does the synchronization
5815 check. A QUIT command is no longer sent after a timeout.
5817 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
5818 during host lookups.
5820 74. The defer_ok option of callout verification was not working if it was used
5821 when verifying addresses in header lines, that is, for this case:
5823 verify = header_sender/callout=defer_ok
5825 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
5826 those file descriptors could be used for SMTP connections. If anything
5827 wrote to stderr (the example that came up was "warn" in embedded Perl), it
5828 could be sent to the SMTP client, causing chaos. The daemon now opens
5829 stdin, stdout, and stderr to /dev/null when it puts itself into the
5832 76. Arrange for output from Perl's "warn" command to be written to Exim's main
5833 log by default. The user can override this with suitable Perl magic.
5835 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
5836 the log message when discard triggers, was not working for the DATA ACL or
5837 for the non-SMTP ACL.
5839 78. Error message wording change in sieve.c.
5841 79. If smtp_accept_max_per_host was set, the number of connections could be
5842 restricted to fewer than expected, because the daemon was trying to set up
5843 a new connection before checking whether the processes handling previous
5844 connections had finished. The check for completed processes is now done
5845 earlier. On busy systems, this bug wouldn't be noticed because something
5846 else would have woken the daemon, and it would have reaped the completed
5849 80. If a message was submitted locally by a user whose login name contained one
5850 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
5851 It caused a spool format error. I have fixed the spool reading code. A
5852 related problem was that the "from" clause in the Received: line became
5853 illegal because of the space(s). It is now covered by ${quote_local_part.
5855 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
5858 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
5860 83. Patch from Sieve maintainer to fix unterminated string problem in
5861 "vacation" handling.
5863 84. Some minor changes to the Linux configuration files to help with other
5864 OS variants using glibc.
5866 85. One more patch for Sieve to update vacation handling to latest spec.
5869 ----------------------------------------------------
5870 See the note above about the 4.44 and 4.50 releases.
5871 ----------------------------------------------------
5877 1. Change 4.43/35 introduced a bug that caused file counts to be
5878 incorrectly computed when quota_filecount was set in an appendfile
5881 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5882 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5885 3. Give more explanation in the error message when the command for a transport
5886 filter fails to execute.
5888 4. There are several places where Exim runs a non-Exim command in a
5889 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5890 was being done only for the command run by the queryprogram router. It is
5891 now done for all such subprocesses. The other cases are: ${run, transport
5892 filters, and the commands run by the lmtp and pipe transports.
5894 5. Some older OS have a limit of 256 on the maximum number of file
5895 descriptors. Exim was using setrlimit() to set 1000 as a large value
5896 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5897 systems. I've change it so that if it can't get 1000, it tries for 256.
5899 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5900 was an oversight, and furthermore, ever since the addition of extra
5901 controls (e.g. 4.43/32), the checks on when to allow different forms of
5902 "control" were broken. There should now be diagnostics for all cases when a
5903 control that does not make sense is encountered.
5905 7. $recipients is now available in the predata ACL (oversight).
5907 8. Tidy the search cache before the fork to do a delivery from a message
5908 received from the command line. Otherwise the child will trigger a lookup
5909 failure and thereby defer the delivery if it tries to use (for example) a
5910 cached ldap connection that the parent has called unbind on.
5912 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5913 of $address_data from the recipient verification was clobbered by the
5914 sender verification.
5916 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5917 was its contents. (It was OK if the option was not defined at all.)
5919 11. A "Completed" log line is now written for messages that are removed from
5920 the spool by the -Mrm option.
5922 12. $host_address is now set to the target address during the checking of
5923 ignore_target_hosts.
5925 13. When checking ignore_target_hosts for an ipliteral router, no host name was
5926 being passed; this would have caused $sender_host_name to have been used if
5927 matching the list had actually called for a host name (not very likely,
5928 since this list is usually IP addresses). A host name is now passed as
5931 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5932 code that specifies a non-restarting handler (typically sigaction() in
5933 modern systems) in an attempt to fix a rare and obscure crash bug.
5935 15. Narrowed the window for a race in the daemon that could cause it to ignore
5936 SIGCHLD signals. This is not a major problem, because they are used only to
5937 wake it up if nothing else does.
5939 16. A malformed maildirsize file could cause Exim to calculate negative values
5940 for the mailbox size or file count. Odd effects could occur as a result.
5941 The maildirsize information is now recalculated if the size or filecount
5944 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5945 support for a long time. Removed HAVE_SYS_VFS_H.
5947 18. Updated exipick to current release from John Jetmore.
5949 19. Allow an empty sender to be matched against a lookup in an address list.
5950 Previously the only cases considered were a regular expression, or an
5953 20. Exim went into a mad DNS lookup loop when doing a callout where the
5954 host was specified on the transport, if the DNS lookup yielded more than
5957 21. The RFC2047 encoding function was originally intended for short strings
5958 such as real names; it was not keeping to the 75-character limit for
5959 encoded words that the RFC imposes. It now respects the limit, and
5960 generates multiple encoded words if necessary. To be on the safe side, I
5961 have increased the buffer size for the ${rfc2047: expansion operator from
5964 22. Failure to deliver a bounce message always caused it to be frozen, even if
5965 there was an errors_to setting on the router. The errors_to setting is now
5968 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5969 canonical form (fully expanded) before being placed in
5970 $sender_host_address.
5972 24. Updated eximstats to version 1.33
5974 25. Include certificate and key file names in error message when GnuTLS fails
5975 to set them up, because the GnuTLS error message doesn't include the name
5976 of the failing file when there is a problem reading it.
5978 26. Expand error message when OpenSSL has problems setting up cert/key files.
5981 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5982 (this can affect the format of dates).
5984 28. exim_tidydb, when checking for the continued existence of a message for
5985 which it has found a message-specific retry record, was not finding
5986 messages that were in split spool directories. Consequently, it was
5987 deleting retry records that should have stayed in existence.
5989 29. eximstats updated to version 1.35
5990 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5991 1.35 - bugfix such that pie charts by volume are generated correctly
5993 30. The SPA authentication driver was not abandoning authentication and moving
5994 on to the next authenticator when an expansion was forced to fail,
5995 contradicting the general specification for all authenticators. Instead it
5996 was generating a temporary error. It now behaves as specified.
5998 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5999 (the order specifies the preference for clients). The order is now AES256,
6000 AES128, 3DES, ARCFOUR128.
6002 31. Small patch to Sieve code - explicitly set From: when generating an
6005 32. Exim crashed if a remote delivery caused a very long error message to be
6006 recorded - for instance if somebody sent an entire SpamAssassin report back
6007 as a large number of 550 error lines. This bug was coincidentally fixed by
6008 increasing the size of one of Exim's internal buffers (big_buffer) that
6009 happened as part of the Exiscan merge. However, to be on the safe side, I
6010 have made the code more robust (and fixed the comments that describe what
6013 33. Some experimental protocols are using DNS PTR records for new purposes. The
6014 keys for these records are domain names, not reversed IP addresses. The
6015 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
6016 leaves it alone. Component reversal etc. now happens only for IP addresses.
6019 34. The host_aton() function is supposed to be passed a string that is known
6020 to be a valid IP address. However, in the case of IPv6 addresses, it was
6021 not checking this. This is a hostage to fortune. Exim now panics and dies
6022 if the condition is not met. A case was found where this could be provoked
6023 from a dnsdb PTR lookup with an IPv6 address that had more than 8
6024 components; fortuitously, this particular loophole had already been fixed
6025 by change 4.50/55 or 4.44/33 above.
6027 If there are any other similar loopholes, the new check in host_aton()
6028 itself should stop them being exploited. The report I received stated that
6029 data on the command line could provoke the exploit when Exim was running as
6030 exim, but did not say which command line option was involved. All I could
6031 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
6032 running as the user.
6035 35. There was a buffer overflow vulnerability in the SPA authentication code
6036 (which came originally from the Samba project). I have added a test to the
6037 spa_base64_to_bits() function which I hope fixes it.
6040 36. The daemon start-up calls getloadavg() while still root for those OS that
6041 need the first call to be done as root, but it missed one case: when
6042 deliver_queue_load_max is set with deliver_drop_privilege. This is
6043 necessary for the benefit of the queue runner, because there is no re-exec
6044 when deliver_drop_privilege is set.
6046 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
6047 was in use, was not putting the data itself into the right store pool;
6048 consequently, it could be overwritten for a subsequent message in the same
6049 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
6052 38. Sometimes the final signoff response after QUIT could fail to get
6053 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
6054 < 0 before doing a fflush(). This bug looks as though it goes back to the
6055 introduction of TLS in release 3.20, but "sometimes" must have been rare
6056 because the tests only now provoked it.
6062 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
6063 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
6064 Now it returns an integer. A number of calls were still expecting a T/F
6065 return. Fortuitously, in all cases, the tests worked in OK situations,
6066 which is the norm. However, things would have gone wrong on any write
6067 failures on the smtp file descriptor. This function is used when sending
6068 messages over SMTP and also when doing verify callouts.
6070 2. When Exim is called to do synchronous delivery of a locally submitted
6071 message (the -odf or -odi options), it no longer closes stderr before doing
6074 3. Implemented the mua_wrapper option.
6076 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
6078 5. Implemented the functions header_remove(), header_testname(),
6079 header_add_at_position(), and receive_remove_recipient(), and exported them
6082 6. If an ACL "warn" statement specified the addition of headers, Exim already
6083 inserted X-ACL-Warn: at the start if there was no header name. However, it
6084 was not making this test for the second and subsequent header lines if
6085 there were newlines in the string. This meant that an invalid header could
6086 be inserted if Exim was badly configured.
6088 7. Allow an ACL "warn" statement to add header lines at the start or after all
6089 the Received: headers, as well as at the end.
6091 8. Added the rcpt_4xx retry error code.
6093 9. Added postmaster_mailfrom=xxx to callout verification option.
6095 10. Added mailfrom=xxxx to the callout verification option, for verify=
6098 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
6099 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
6101 12. Inserted some casts to stop certain compilers warning when using pointer
6102 differences as field lengths or precisions in printf-type calls (mostly
6103 affecting debugging statements).
6105 13. Added optional readline() support for -be (dynamically loaded).
6107 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
6108 same clock tick as a message's arrival, so that its received time was the
6109 same as the "first fail" time on the retry record, and that message
6110 remained on the queue past the ultimate address timeout, every queue runner
6111 would try a delivery (because it was past the ultimate address timeout) but
6112 after another failure, the ultimate address timeout, which should have then
6113 bounced the address, did not kick in. This was a "< instead of <=" error;
6114 in most cases the first failure would have been in the next clock tick
6115 after the received time, and all would be well.
6117 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
6118 being recognized when the domain list was tested by the match_domain
6119 condition in an expansion string.
6121 16. Added the ${str2b64: operator.
6123 17. Exim was always calling setrlimit() to set a large limit for the number of
6124 processes, without checking whether the existing limit was already
6125 adequate. (It did check for the limit on file descriptors.) Furthermore,
6126 errors from getrlimit() and setrlimit() were being ignored. Now they are
6127 logged to the main and panic logs, but Exim does carry on, to try to do its
6128 job under whatever limits there are.
6130 18. Imported PCRE 5.0.
6132 19. Trivial typo in log message " temporarily refused connection" (the leading
6135 20. If the log selector return_path_on_delivery was set and an address was
6136 redirected to /dev/null, the delivery process crashed because it assumed
6137 that a return path would always be set for a "successful" delivery. In this
6138 case, the whole delivery is bypassed as an optimization, and therefore no
6141 21. Internal re-arrangement: the function for sending a challenge and reading
6142 a response while authentication was assuming a zero-terminated challenge
6143 string. It's now changed to take a pointer and a length, to allow for
6144 binary data in such strings.
6146 22. Added the cyrus_sasl authenticator (code supplied by MBM).
6148 23. Exim was not respecting finduser_retries when seeking the login of the
6149 uid under which it was called; it was always trying 10 times. (The default
6150 setting of finduser_retries is zero.) Also, it was sleeping after the final
6151 failure, which is pointless.
6153 24. Implemented tls_on_connect_ports.
6155 25. Implemented acl_smtp_predata.
6157 26. If the domain in control=submission is set empty, Exim assumes that the
6158 authenticated id is a complete email address when it generates From: or
6159 Sender: header lines.
6161 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
6162 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
6163 chown and chgrp in /bin and hostname in /usr/bin.
6165 28. Exim was keeping the "process log" file open after each use, just as it
6166 does for the main log. This opens the possibility of it remaining open for
6167 long periods when the USR1 signal hits a daemon. Occasional processlog
6168 errors were reported, that could have been caused by this. Anyway, it seems
6169 much more sensible not to leave this file open at all, so that is what now
6172 29. The long-running daemon process does not normally write to the log once it
6173 has entered its main loop, and it closes the log before doing so. This is
6174 so that log files can straightforwardly be renamed and moved. However,
6175 there are a couple of unusual error situations where the daemon does write
6176 log entries, and I had neglected to close the log afterwards.
6178 30. The text of an SMTP error response that was received during a remote
6179 delivery was being truncated at 512 bytes. This is too short for some of
6180 the long messages that one sometimes sees. I've increased the limit to
6183 31. It is now possible to make retry rules that apply only when a message has a
6184 specific sender, in particular, an empty sender.
6186 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
6187 it possible to be selective about when SMTP synchronization is enforced.
6189 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
6191 32. Implemented hosts_connection_nolog.
6193 33. Added an ACL for QUIT.
6195 34. Setting "delay_warning=" to disable warnings was not working; it gave a
6198 35. Added mailbox_size and mailbox_filecount to appendfile.
6200 36. Added control = no_multiline_responses to ACLs.
6202 37. There was a bug in the logic of the code that waits for the clock to tick
6203 in the case where the clock went backwards by a substantial amount such
6204 that the microsecond fraction of "now" was more than the microsecond
6205 fraction of "then" (but the whole seconds number was less).
6207 38. Added support for the libradius Radius client library this is found on
6208 FreeBSD (previously only the radiusclient library was supported).
6214 1. When certain lookups returned multiple values in the form name=value, the
6215 quoting of the values was not always being done properly. Specifically:
6216 (a) If the value started with a double quote, but contained no whitespace,
6218 (b) If the value contained whitespace other than a space character (i.e.
6219 tabs or newlines or carriage returns) it was not quoted.
6220 This fix has been applied to the mysql and pgsql lookups by writing a
6221 separate quoting function and calling it from the lookup code. The fix
6222 should probably also be applied to nisplus, ibase and oracle lookups, but
6223 since I cannot test any of those, I have not disturbed their existing code.
6225 2. A hit in the callout cache for a specific address caused a log line with no
6226 reason for rejecting RCPT. Now it says "Previous (cached) callout
6227 verification failure".
6229 3. There was an off-by-one bug in the queryprogram router. An over-long
6230 return line was truncated at 256 instead of 255 characters, thereby
6231 overflowing its buffer with the terminating zero. As well as fixing this, I
6232 have increased the buffer size to 1024 (and made a note to document this).
6234 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
6235 when Exim is waiting for an SMTP response from a remote server, Exim
6236 restarts its select() call on the socket, thereby resetting its timeout.
6237 This is not a problem when such interrupts are rare. Somebody set up a cron
6238 job to run exiwhat every 2 minutes, which is less than the normal select()
6239 timeout (5 or 10 minutes). This meant that the select() timeout never
6240 kicked in because it was always reset. I have fixed this by comparing the
6241 time when an interrupt arrives with the time at the start of the first call
6242 to select(). If more time than the timeout has elapsed, the interrupt is
6243 treated as a timeout.
6245 5. Some internal re-factoring in preparation for the addition of Sieve
6246 extensions (by MH). In particular, the "personal" test is moved to a
6247 separate function, and given an option for scanning Cc: and Bcc: (which is
6248 not set for Exim filters).
6250 6. When Exim created an email address using the login of the caller as the
6251 local part (e.g. when creating a From: or Sender: header line), it was not
6252 quoting the local part when it contained special characters such as @.
6254 7. Installed new OpenBSD configuration files.
6256 8. Reworded some messages for syntax errors in "and" and "or" conditions to
6257 try to make them clearer.
6259 9. Callout options, other than the timeout value, were being ignored when
6260 verifying sender addresses in header lines. For example, when using
6262 verify = header_sender/callout=no_cache
6264 the cache was (incorrectly) being used.
6266 10. Added a missing instance of ${EXE} to the exim_install script; this affects
6267 only the Cygwin environment.
6269 11. When return_path_on_delivery was set as a log selector, if different remote
6270 addresses in the same message used different return paths and parallel
6271 remote delivery occurred, the wrong values would sometimes be logged.
6272 (Whenever a remote delivery process finished, the return path value from
6273 the most recently started remote delivery process was logged.)
6275 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
6276 lines when AUTH and/or TLS are in use. This is the "received protocol"
6277 field. Exim used to use "asmtp" for authenticated SMTP, without any
6278 indication (in the protocol name) for TLS use. Now it follows the RFC and
6279 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
6280 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
6281 names appear in log lines as well as in Received: header lines.
6283 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
6284 extensions, and comparison tests, and to fix some bugs.
6286 14. Changes to the "personal" filter test:
6288 (1) The test was buggy in that it was just doing the equivalent of
6289 "contains" tests on header lines. For example, if a user's address was
6290 anne@some.where, the "personal" test would incorrectly be true for
6292 To: susanne@some.where
6294 This test is now done by extracting each address from the header in turn,
6295 and checking the entire address. Other tests that are part of "personal"
6296 are now done using regular expressions (for example, to check local parts
6297 of addresses in From: header lines).
6299 (2) The list of non-personal local parts in From: addresses has been
6300 extended to include "listserv", "majordomo", "*-request", and "owner-*",
6301 taken from the Sieve specification recommendations.
6303 (3) If the message contains any header line starting with "List-" it is
6304 treated as non-personal.
6306 (4) The test for "circular" in the Subject: header line has been removed
6307 because it now seems ill-conceived.
6309 15. Minor typos in src/EDITME comments corrected.
6311 16. Installed latest exipick from John Jetmore.
6313 17. If headers_add on a router specified a text string that was too long for
6314 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
6315 of string_sprintf() is now avoided.
6317 18. $message_body_size was not set (it was always zero) when running the DATA
6318 ACL and the local_scan() function.
6320 19. For the "mail" command in an Exim filter, no default was being set for
6321 the once_repeat time, causing a random time value to be used if "once" was
6322 specified. (If the value happened to be <= 0, no repeat happened.) The
6323 default is now 0s, meaning "never repeat". The "vacation" command was OK
6324 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
6325 (I found it when inspecting the code).
6327 20. There is now an overall timeout for performing a callout verification. It
6328 defaults to 4 times the callout timeout, which applies to individual SMTP
6329 commands during the callout. The overall timeout applies when there is more
6330 than one host that can be tried. The timeout is checked before trying the
6331 next host. This prevents very long delays if there are a large number of
6332 hosts and all are timing out (e.g. when the network connections are timing
6333 out). The value of the overall timeout can be changed by specifying an
6334 additional sub-option for "callout", called "maxwait". For example:
6336 verify = sender/callout=5s,maxwait=20s
6338 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
6339 to the end before writing, but this should make it even safer).
6341 22. Exim was forgetting that it had advertised PIPELINING for the second and
6342 subsequent messages on an SMTP connection. It was also not resetting its
6343 memory on STARTTLS and an internal HELO.
6345 23. When Exim logs an SMTP synchronization error within a session, it now
6346 records whether PIPELINING has been advertised or not.
6348 24. Added 3 instances of "(long int)" casts to time_t variables that were being
6349 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
6350 rather than long int.
6352 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
6354 26. Added the never_mail option to autoreply.
6360 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
6361 crash if the getsockname() call failed; this can happen if a connection is
6362 closed very soon after it is established. The problem was simply in the
6363 order in which certain operations were done, causing Exim to try to write
6364 to the SMTP stream before it had set up the file descriptor. The bug has
6365 been fixed by making things happen in the correct order.
6371 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
6372 before the connection was closed, thus losing the rejection response.
6374 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
6375 some early Solaris releases, but causes trouble in current releases where
6376 socklen_t is defined.
6378 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
6381 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
6384 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
6385 that releasing the top of it at the end releases what was used for sub-
6386 expansions (unless the block got too big). However, discard this block if
6387 the first thing is a variable or header, so that we can use its block when
6388 it is dynamic (useful for very large $message_headers, for example).
6390 6. Lookups now cache *every* query, not just the most recent. A new, separate
6391 store pool is used for this. It can be recovered when all lookup caches are
6392 flushed. Lookups now release memory at the end of their result strings.
6393 This has involved some general refactoring of the lookup sources.
6395 7. Some code has been added to the store_xxx() functions to reduce the amount
6396 of flapping under certain conditions.
6398 8. log_incoming_interface used to affect only the <= reception log lines. Now
6399 it causes the local interface and port to be added to several more SMTP log
6400 lines, for example "SMTP connection from", and rejection lines.
6402 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
6404 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
6406 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
6407 could be overwritten at the end of the current message (or the start of a
6408 new message if it was set in a HELO ACL). The value is now preserved for
6409 the duration of the SMTP connection.
6411 12. If a transport had a headers_rewrite setting, and a matching header line
6412 contained an unqualified address, that address was qualified, even if it
6413 did not match any rewriting rules. The underlying bug was that the values
6414 of the flags that permit the existence of unqualified sender and recipient
6415 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
6416 non-local messages, and by -bnq for local messages) were not being
6417 preserved with the message after it was received.
6419 13. When Exim was logging an SMTP synchronization error, it could sometimes log
6420 "next input=" as part of the text comprising the host identity instead of
6421 the correct text. The code was using the same buffer for two different
6422 strings. However, depending on which order the printing function evaluated
6423 its arguments, the bug did not always show up. Under Linux, for example, my
6424 test suite worked just fine.
6426 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
6427 doesn't work with some older versions of Perl. It has been changed to "my",
6428 which in any case is probably the better facility to use.
6430 15. A really picky compiler found some instances of statements for creating
6431 error messages that either had too many or two few arguments for the format
6434 16. The size of the buffer for calls to the DNS resolver has been increased
6435 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
6436 for addresses that have a lot of PTR records. This alleviates a problem; it
6437 does not fully solve it.
6439 17. A dnsdb lookup for PTR records that receives more data than will fit in the
6440 buffer now truncates the list and logs the incident, which is the same
6441 action as happens when Exim is looking up a host name and its aliases.
6442 Previously in this situation something unpredictable would happen;
6443 sometimes it was "internal error: store_reset failed".
6445 18. If a server dropped the connection unexpectedly when an Exim client was
6446 using GnuTLS and trying to read a response, the client delivery process
6447 crashed while trying to generate an error log message.
6449 19. If a "warn" verb in an ACL added multiple headers to a message in a single
6450 string, for example:
6452 warn message = H1: something\nH2: something
6454 the text was added as a single header line from Exim's point of view
6455 though it ended up OK in the delivered message. However, searching for the
6456 second and subsequent header lines using $h_h2: did not work. This has been
6457 fixed. Similarly, if a system filter added multiple headers in this way,
6458 the routers could not see them.
6460 20. Expanded the error message when iplsearch is called with an invalid key to
6461 suggest using net-iplsearch in a host list.
6463 21. When running tests using -bh, any delays imposed by "delay" modifiers in
6464 ACLs are no longer actually imposed (and a message to that effect is
6467 22. If a "gecos" field in a passwd entry contained escaped characters, in
6468 particular, if it contained a \" sequence, Exim got it wrong when building
6469 a From: or a Sender: header from that name. A second bug also caused
6470 incorrect handling when an unquoted " was present following a character
6471 that needed quoting.
6473 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
6474 was not being matched caselessly.
6476 24. Arranged for all hyphens in the exim.8 source to be escaped with
6479 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
6480 properly. Recipient callout cache records were still being keyed to include
6481 the sender, even when use_sender was set false. This led to far more
6482 callouts that were necessary. The sender is no longer included in the key
6483 when use_sender is false.
6485 26. Added "control = submission" modifier to ACLs.
6487 27. Added the ${base62d: operator to decode base 62 numbers.
6489 28. dnsdb lookups can now access SRV records.
6491 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
6492 the configuration file.
6494 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
6495 (-v) mode. This makes the output for a verbose queue run more intelligible.
6497 31. Added a use_postmaster feature to recipient callouts.
6499 32. Added the $body_zerocount variable, containing the number of binary zero
6500 bytes in the message body.
6502 33. The time of last modification of the "new" subdirectory is now used as the
6503 "mailbox time last read" when there is a quota error for a maildir
6506 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
6508 35. Added +ignore_unknown as a special item in host lists.
6510 36. Code for decoding IPv6 addresses in host lists is now included, even if
6511 IPv6 support is not being compiled. This fixes a bug in which an IPv6
6512 address was recognized as an IP address, but was then not correctly decoded
6513 into binary, causing unexpected and incorrect effects when compared with
6520 1. Very minor rewording of debugging text in manualroute to say "list of
6521 hosts" instead of "hostlist".
6523 2. If verify=header_syntax was set, and a header line with an unqualified
6524 address (no domain) and a large number of spaces between the end of the
6525 name and the colon was received, the reception process suffered a buffer
6526 overflow, and (when I tested it) crashed. This was caused by some obsolete
6527 code that should have been removed. The fix is to remove it!
6529 3. When running in the test harness, delay a bit after writing a bounce
6530 message to get a bit more predictability in the log output.
6532 4. Added a call to search_tidyup() just before forking a reception process. In
6533 theory, someone could use a lookup in the expansion of smtp_accept_max_
6534 per_host which, without the tidyup, could leave open a database connection.
6536 5. Added the variables $recipient_data and $sender_data which get set from a
6537 lookup success in an ACL "recipients" or "senders" condition, or a router
6538 "senders" option, similar to $domain_data and $local_part_data.
6540 6. Moved the writing of debug_print from before to after the "senders" test
6543 7. Change 4.31/66 (moving the time when the Received: is generated) caused
6544 problems for message scanning, either using a data ACL, or using
6545 local_scan() because the Received: header was not generated till after they
6546 were called (in order to set the time as the time of reception completion).
6547 I have revised the way this works. The header is now generated after the
6548 body is received, but before the ACL or local_scan() are called. After they
6549 are run, the timestamp in the header is updated.
6555 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
6556 before starting a queue runner without re-exec. This happened only when
6557 deliver_drop_privilege was set or when the Exim user was set to root. The
6558 effect of the bug was that timeouts during subsequent deliveries caused
6559 crashes instead of being properly handled. The handler is now left at its
6560 default (and expected) setting.
6562 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
6563 message, again when deliver_drop_privilege is set or Exim is run as root.
6564 The bug described in (1) was not present in this case, but the tidying up
6565 of the other signals was missing. I have made the two cases consistent.
6567 3. The ignore_target_hosts setting on a manualroute router was being ignored
6568 for hosts that were looked up using the /MX notation.
6570 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
6573 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
6574 operated on the sender address. After changing the $sender_address to <>
6575 for the sender address verify, Exim was re-instated it as the original
6576 (before rewriting) address, but remembering that it had rewritten it, so it
6577 wasn't rewriting it again. This bug also had the effect of breaking the
6578 sender address verification caching when the sender address was rewritten.
6580 6. The ignore_target_hosts option was being ignored by the ipliteral router.
6581 This has been changed so that if the ip literal address matches
6582 ignore_target_hosts, the router declines.
6584 7. Added expansion conditions match_domain, match_address, and match_local_
6585 part (NOT match_host).
6587 8. The placeholder for the Received: header didn't have a length field set.
6589 9. Added code to Exim itself and to exim_lock to test for a specific race
6590 condition that could lead to file corruption when using MBX delivery. The
6591 issue is with the lockfile that is created in /tmp. If this file is removed
6592 after a process has opened it but before that process has acquired a lock,
6593 there is the potential for a second process to recreate the file and also
6594 acquire a lock. This could lead to two Exim processes writing to the file
6595 at the same time. The added code performs the same test as UW imapd; it
6596 checks after acquiring the lock that its file descriptor still refers to
6597 the same named file.
6599 10. The buffer for building added header lines was of fixed size, 8192 bytes.
6600 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
6603 11. Added the smtp_active_hostname option. If used, this will typically be made
6604 to depend on the incoming interface address. Because $interface_address is
6605 not set up until the daemon has forked a reception process, error responses
6606 that can happen earlier (such as "too many connections") no longer contain
6609 12. If an expansion in a condition on a "warn" statement fails because a lookup
6610 defers, the "warn" statement is abandoned, and the next ACL statement is
6611 processed. Previously this caused the whole ACL to be aborted.
6613 13. Added the iplsearch lookup type.
6615 14. Added ident_timeout as a log selector.
6617 15. Added tls_certificate_verified as a log selector.
6619 16. Added a global option tls_require_ciphers (compare the smtp transport
6620 option of the same name). This controls incoming TLS connections.
6622 17. I finally figured out how to make tls_require_ciphers do a similar thing
6623 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
6624 before starting the TLS session.
6626 18. Tabs are now shown as \t in -bP output.
6628 19. If the log selector return_path_on_delivery was set, Exim crashed when
6629 bouncing a message because it had too many Received: header lines.
6631 20. If two routers both had headers_remove settings, and the first one included
6632 a superfluous trailing colon, the final name in the first list and the
6633 first name in the second list were incorrectly joined into one item (with a
6634 colon in the middle).
6640 1. Added -C and -D options to the exinext utility, mainly to make it easier
6641 to include in the automated testing, but these could be helpful when
6642 multiple configurations are in use.
6644 2. The exinext utility was not formatting the output nicely when there was
6645 an alternate port involved in the retry record key, nor when there was a
6646 message id as well (for retries that were specific to a specific message
6647 and a specific host). It was also confused by IPv6 addresses, because of
6648 the additional colons they contain. I have fixed the IPv4 problem, and
6649 patched it up to do a reasonable job for IPv6.
6651 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
6652 delivery, the log line now contains "pipelined" if PIPELINING was used.
6654 4. An SMTP transport process used to panic and die if the bind() call to set
6655 an explicit outgoing interface failed. This has been changed; it is now
6656 treated in the same way as a connect() failure.
6658 5. A reference to $sender_host_name in the part of a conditional expansion
6659 that was being skipped was still causing a DNS lookup. This no longer
6662 6. The def: expansion condition was not recognizing references to header lines
6663 that used bh_ and bheader_.
6665 7. Added the _cache feature to named lists.
6667 8. The code for checking quota_filecount in the appendfile transport was
6668 allowing one more file than it should have been.
6670 9. For compatibility with Sendmail, the command line option
6678 and sets the incoming protocol and host name (for trusted callers). The
6679 host name and its colon can be omitted when only the protocol is to be set.
6680 Note the Exim already has two private options, -pd and -ps, that refer to
6681 embedded Perl. It is therefore impossible to set a protocol value of "d" or
6682 "s", but I don't think that's a major issue.
6684 10. A number of refactoring changes to the code, none of which should affect
6687 (a) The number of logging options was getting close to filling up the
6688 32-bit word that was used as a bit map. I have split them into two classes:
6689 those that are passed in the argument to log_write(), and those that are
6690 only ever tested independently outside of that function. These are now in
6691 separate 32-bit words, so there is plenty of room for expansion again.
6692 There is no change in the user interface or the logging behaviour.
6694 (b) When building, for example, log lines, the code previously used a
6695 macro that called string_cat() twice, in order to add two strings. This is
6696 not really sufficiently general. Furthermore, there was one instance where
6697 it was actually wrong because one of the argument was used twice, and in
6698 one call a function was used. (As it happened, calling the function twice
6699 did not affect the overall behaviour.) The macro has been replaced by a
6700 function that can join an arbitrary number of extra strings onto a growing
6703 (c) The code for expansion conditions now uses a table and a binary chop
6704 instead of a serial search (which was left over from when there were very
6705 few conditions). Also, it now recognizes conditions like "pam" even when
6706 the relevant support is not compiled in: a suitably worded error message is
6707 given if an attempt is made to use such a condition.
6709 11. Added ${time_interval:xxxxx}.
6711 12. A bug was causing one of the ddress fields not to be passed back correctly
6712 from remote delivery subprocesses. The field in question was not being
6713 subsequently used, so this caused to problems in practice.
6715 13. Added new log selectors queue_time and deliver_time.
6717 14. Might have fixed a bug in maildirsizefile handling that threw up
6718 "unexpected character" debug warnings, and recalculated the data
6719 unnecessarily. In any case, I expanded the warning message to give more
6722 15. Added the message "Restricted characters in address" to the statements in
6723 the default ACL that block characters like @ and % in local parts.
6725 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
6726 Three changes have been made:
6728 (a) There was a serious bug; a negative response to MAIL caused the whole
6729 recipient domain to be cached as invalid, thereby blocking all messages
6730 to all local parts at the same domain, from all senders. This bug has
6731 been fixed. The domain is no longer cached after a negative response to
6732 MAIL if the sender used is not empty.
6734 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
6737 (c) A new callout option, "use_sender" has been added for people who want
6738 the modified behaviour.
6744 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
6747 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
6748 indeed breaks things for older releases.
6750 3. Added additional logging to the case where there is a problem reading data
6751 from a filter that is running in a subprocess using a pipe, in order to
6752 try to track down a specific problem.
6754 4. Testing facility fudge: when running in the test harness and attempting
6755 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
6756 getting "No route to host". Convert this to a timeout.
6758 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
6761 6. Some OS don't have socklen_t but use size_t instead. This affects the
6762 fifth argument of getsockopt() amongst other things. This is now
6763 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
6764 can be set for individual OS. I have set it for SunOS5, OSF1, and
6765 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
6766 some earlier ones do not.
6768 7. Change 4.30/15 was not doing the test caselessly.
6770 8. The standard form for an IPv6 address literal was being rejected by address
6771 parsing in, for example, MAIL and RCPT commands. An example of this kind of
6772 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6773 this, as well as the form without the "IPv6" on the front (but only when
6774 address literals are enabled, of course).
6776 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
6778 10. Exim crashed if a message with an empty sender address specified by -f
6779 encountered a router with an errors_to setting. This could be provoked only
6780 by a command such as
6784 where an empty string was supplied; "<>" did not hit this bug.
6786 11. Installed PCRE release 4.5.
6788 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
6789 remained set. It is now erased.
6791 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
6792 times from message ids (which are base 36 rather than the normal 62).
6794 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
6795 were being counted as actual protocol errors, and logged if the log
6796 selector +smtp_protocol_error was set. One cannot be perfect in this test,
6797 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
6798 and DATA following a set of rejected RCPTs do not count as protocol errors.
6799 In other words, Exim assumes they were pipelined, though this may not
6800 actually be the case. Of course, in all cases the client gets an
6801 appropriate error code.
6803 15. If a lookup fails in an ACL condition, a message about the failure may
6804 be available; it is used if testing the ACL cannot continue, because most
6805 such messages specify what the cause of the deferral is. However, some
6806 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
6807 that caused an old message to be retained and used if a later statement
6808 caused a defer, replacing the real cause of the deferral.
6810 16. If an IP address had so many PTR records that the DNS lookup buffer
6811 was not large enough to hold them, Exim could crash while trying to process
6812 the truncated data. It now detects and logs this case.
6814 17. Further to 4.21/58, another change has been made: if (and only if) the
6815 first line of a message (the first header line) ends with CRLF, a bare LF
6816 in a subsequent header line has a space inserted after it, so as not to
6817 terminate the header.
6819 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
6820 unnecessarily, used atoi() instead of strtol(), and didn't check the
6821 termination when getting file sizes from file names by regex.
6823 19. Completely re-implemented the support for maildirsize files, in the light
6824 of a number of problems with the previous contributed implementation
6825 (4.30/29). In particular:
6827 . If the quota is zero, the maildirsize file is maintained, but no quota is
6830 . If the maildir directory does not exist, it is created before any attempt
6831 to write a maildirsize file.
6833 . The quota value in the file is just a cache; if the quota is changed in
6834 the transport, the new value overrides.
6836 . A regular expression is available for excluding directories from the
6839 20. The autoreply transport checks the characters in options that define the
6840 message's headers; it allows continued headers, but it was checking with
6841 isspace() after an embedded newline instead of explicitly looking for a
6844 21. If all the "regular" hosts to which an address was routed had passed their
6845 expiry times, and had not reached their retry times, the address was
6846 bounced, even if fallback hosts were defined. Now Exim should go on to try
6849 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
6850 equivalent code in the SMTP transport. Some hosts send humungous responses
6851 to HELO/EHLO, more than 1024 it seems.
6853 23. Refactoring: code in filter.c used (void *) for "any old type" but this
6854 gives compiler warnings in some environments. I've now done it "properly",
6857 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
6858 (because of problems with the built-in one) was declared to return uschar *
6859 instead of char *, causing compiler failure.
6861 25. Fixed a file descriptor leak when processing alias/forward files.
6863 26. Fixed a minor format string issue in dbfn.c.
6865 27. Typo in exim.c: ("dmbnz" for "dbmnz").
6867 28. If a filter file refered to $h_xxx or $message_headers, and the headers
6868 contained RFC 2047 "words", Exim's memory could, under certain conditions,
6871 29. When a sender address is verified, it is cached, to save repeating the test
6872 when there is more than one recipient in a message. However, when the
6873 verification involves a callout, it is possible for different callout
6874 options to be set for different recipients. It is too complicated to keep
6875 track of this in the cache, so now Exim always runs a verification when a
6876 callout is required, relying on the callout cache for the optimization.
6877 The overhead is duplication of the address routing, but this should not be
6880 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
6881 to be verified with callout=postmaster, and the main callout worked but the
6882 postmaster check failed, the verification correctly failed. However, if a
6883 subsequent RCPT command asked for sender verification *without* the
6884 postmaster check, incorrect caching caused this verification also to fail,
6887 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
6888 it was not caching the DNS options (qualify_single, search_parents) that
6889 were used when the lookup failed. A subsequent lookup with different
6890 options therefore always gave the same answer, though there were cases
6891 where it should not have. (Example: a "domains = !$mx_any" option on a
6892 dnslookup router: the "domains" option is always processed without any
6893 widening, but the router might have qualify_single set.) Now Exim uses the
6894 cached value only when the same options are set.
6896 32. Added John Jetmore's "exipick" utility to the distribution.
6898 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
6899 than a timeout (e.g. a certificate is required, and is not provided), an
6900 Exim server now closes the connection immediately. Previously it waited for
6901 the client to close - but if the client is SSL, it seems that they each
6902 wait for each other, leading to a delay before one of them times out.
6904 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
6905 maintained 0.8.x compatibility because I don't think many are using it, and
6906 it is clearly obsolete.
6908 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
6911 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
6912 client certificate was expired. A simple patch fixes this, though I don't
6913 understand the full logic of why the verify callback is called multiple
6916 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
6917 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
6918 which causes problems with some clients (such as the Certicom SSL Plus
6919 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
6920 disables the coutermeasure allowing Eudora to connect."
6922 38. Exim was not checking that a write() to a log file succeeded. This could
6923 lead to Bad Things if a log got too big, in particular if it hit a file
6924 size limit. Exim now panics and dies if it cannot write to a log file, just
6925 as it does if it cannot open a log file.
6927 39. Modified OS/Makefile-Linux so that it now contains
6929 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
6931 The two -D definitions ensure that Exim is compiled with large file
6932 support, which makes it possible to handle log files that are bigger than
6935 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
6936 instance) a domain was checked against a named list that involved a lookup,
6937 causing $domain_data to be set, then another domain was checked against the
6938 same list, then the first domain was re-checked, the value of $domain_data
6939 after the final check could be wrong. In particular, if the second check
6940 failed, it could be set empty. This bug probably also applied to
6943 41. The strip_trailing_dot option was not being applied to the address given
6944 with the -f command-line option.
6946 42. The code for reading a message's header from the spool was incrementing
6947 $received_count, but never initializing it. This meant that the value was
6948 incorrect (doubled) while delivering a message in the same process in which
6949 it was received. In the most common configuration of Exim, this never
6950 happens - a fresh exec is done - but it can happen when
6951 deliver_drop_privilege is set.
6953 43. When Exim logs an SMTP synchronization error - client data sent too soon -
6954 it now includes up to 150 characters of the unexpected data in the log
6957 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
6958 and building data strings. The size of both of these buffers was 10 000
6959 bytes - far larger than anybody would *ever* want, thought I. Needless to
6960 say, somebody hit the limit. I have increased the maximum line length to
6961 20 000 and the maximum data length of concatenated lines to 100 000. I have
6962 also fixed two bugs, because there was no checking on these buffers. Tsk,
6963 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
6964 buffer is too small.
6966 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6967 lsearch lookups. Now it does.
6969 46. When parsing a route_list item in a manualroute router, a fixed-length
6970 buffer was used for the list of hosts. I made this 1024 bytes long,
6971 thinking that nobody would ever have a list of hosts that long. Wrong.
6972 Somebody had a whole pile of complicated expansion conditions, and the
6973 string was silently truncated, leading to an expansion error. It turns out
6974 that it is easier to change to an unlimited length (owing to other changes
6975 that have happened since this code was originally written) than to build
6976 structure for giving a limitation error. The length of the item that
6977 expands into the list of hosts is now unlimited.
6979 47. The lsearch lookup could not handle data where the length of text line was
6980 more than 4095 characters. Such lines were truncated, leading to shortened
6981 data being returned. It should now handle lines of any length.
6983 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6984 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6987 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6988 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6989 no longer happens in comment lines. A list of replacements is now placed
6990 at the head of all of the source files, except those whose only change is
6991 to replace PERL_COMMAND in the very first #! line.
6993 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6994 messages on the queue, with a bottom-up merge sort, using code contributed
6995 by Michael Haardt. This should make operations like -bp somewhat faster on
6996 large queues. It won't affect queue runners, except when queue_run_in_order
6999 51. Installed eximstats 1.31 in the distribution.
7001 52. Added support for SRV lookups to the dnslookup router.
7003 53. If an ACL referred to $message_body or $message_body_end, the value was not
7004 reset for any messages that followed in the same SMTP session.
7006 54. The store-handling optimization for building very long strings was not
7007 differentiating between the different store pools. I don't think this
7008 actually made any difference in practice, but I've tidied it.
7010 55. While running the routers to verify a sender address, $sender_address
7011 was still set to the sender address. This is wrong, because when routing to
7012 send a bounce to the sender, it would be empty. Therefore, I have changed
7013 it so that, while verifying a sender address, $sender_address is set to <>.
7014 (There is no change to what happens when verifying a recipient address.)
7016 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
7017 target A or AAAA records (if not already returned) without resetting the
7018 qualify_single or search_parents options of the DNS resolver. These are
7019 inappropriate in this case because the targets of MX and SRV records must
7020 be FQDNs. A broken DNS record could cause trouble if it happened to have a
7021 target that, when qualified, matched something in the local domain. These
7022 two options are now turned off when doing these lookups.
7024 57. It seems that at least some releases of Reiserfs (which does not have the
7025 concept of a fixed number of inodes) returns zero and not -1 for the
7026 number of available inodes. This interacted badly with check_spool_inodes,
7027 which assumed that -1 was the "no such thing" setting. What I have done is
7028 to check that the total number of inodes is greater than zero before doing
7029 the test of how many are available.
7031 58. When a "warn" ACL statement has a log_message modifier, the message is
7032 remembered, and not repeated. This is to avoid a lot of repetition when a
7033 message has many recipients that cause the same warning to be written.
7034 However, Exim was preserving the list of already written lines for an
7035 entire SMTP session, which doesn't seem right. The memory is now reset if a
7036 new message is started.
7038 59. The "rewrite" debugging flag was not showing the result of rewriting in the
7039 debugging output unless log_rewrite was also set.
7041 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
7042 of (int)(handle) when we know that handle contains (void *)(-1).
7044 61. The Exim daemon panic-logs an error return when it closes the incoming
7045 connection. However "connection reset by peer" seems to be common, and
7046 isn't really an error worthy of noting specially, so that particular error
7049 62. When Exim is trying to find all the local interfaces, it used to panic and
7050 die if the ioctl to get the interface flags failed. However, it seems that
7051 on at least one OS (Solaris 9) it is possible to have an interface that is
7052 included in the list of interfaces, but for which you get a failure error
7053 for this call. This happens when the interface is not "plumbed" into a
7054 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
7055 failure of the "get flags" call assumes that the interface is down.
7057 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
7058 makes life easier for people who are doing arithmetic on fields extracted
7059 from dates, where you often get leading zeros that should not be
7060 interpreted as octal.
7062 64. Added qualify_domain to the redirect router, to override the global
7065 65. If a pathologically long header line contained very many addresses (the
7066 report of this problem mentioned 10 000) and each of them was rewritten,
7067 Exim could use up a very large amount of memory. (It kept on making new
7068 copies of the header line as it rewrote, and never released the old ones.)
7069 At the expense of a bit more processing, the header rewriting function has
7070 been changed so that it no longer eats memory in this way.
7072 66. The generation of the Received: header has been moved from the time that a
7073 message starts to be received, to the time that it finishes. The timestamp
7074 in the Received: header should now be very close to that of the <= log
7075 line. There are two side-effects of this change:
7077 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
7078 logged header lines no longer include the local Received: line, because
7079 it has not yet been created. The same applies to a copy of the message
7080 that is returned to a non-SMTP sender when a message is rejected.
7082 (b) When a filter file is tested using -bf, no additional Received: header
7083 is added to the test message. After some thought, I decided that this
7086 This change does not affect the value of $received_for. It is still set
7087 after address rewriting, but before local_scan() is called.
7089 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
7091 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
7092 gave an unhelpful panic error message, and a defer error. I have managed to
7093 change this behaviour so that it now rejects any supplied certificate,
7094 which seems right, as the list of acceptable certificates is empty.
7096 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
7097 gave an unhelpful defer error. I have not managed to make this reject any
7098 supplied certificates, but the error message it gives is "no certificate
7099 supplied", which is not helpful.
7101 70. exigrep's output now also includes lines that are not associated with any
7102 message, but which match the given pattern. Implemented by a patch from
7103 Martin Sluka, which also tidied up the Perl a bit.
7105 71. Recipient callout verification, like sender verification, was using <> in
7106 the MAIL FROM command. This isn't really the right thing, since the actual
7107 sender may affect whether the remote host accepts the recipient or not. I
7108 have changed it to use the actual sender in the callout; this means that
7109 the cache record is now keyed on a recipient/sender pair, not just the
7110 recipient address. There doesn't seem to be a real danger of callout loops,
7111 since a callout by the remote host to check the sender would use <>.
7112 [SEE ABOVE: changed after hitting problems.]
7114 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
7115 temporary errors. However, in the case of such a code being given after
7116 the end of a data transmission (i.e. after ".") Exim was failing to write
7117 a retry record for the message. (Yes, there was some broken host that was
7118 actually sending 8xx at this point.)
7120 73. An unknown lookup type in a host list could cause Exim to panic-die when
7121 the list was checked. (An example that provoked this was putting <; in the
7122 middle of a list instead of at the start.) If this happened during a DATA
7123 ACL check, a -D file could be left lying around. This kind of configuration
7124 error no longer causes Exim to die; instead it causes a defer error. The
7125 incident is still logged to the main and panic logs.
7127 74. Buglet left over from Exim 3 conversion. The message "too many messages
7128 in one connection" was written to the rejectlog but not the mainlog, except
7129 when address rewriting (yes!) was being logged.
7131 75. Added write_rejectlog option.
7133 76. When a system filter was run not as root (that is, when system_filter_user
7134 was set), the values of the $n variables were not being returned to the
7135 main process; thus, they were not subsequently available in the $sn
7138 77. Added +return_path_on_delivery log selector.
7140 78. A connection timeout was being treated differently from recipients deferred
7141 when testing hosts_max_try with a message that was older than the host's
7142 retry timeout. (The host should not be counted, thus allowing all hosts to
7143 be tried at least once before bouncing.) This may have been the cause of an
7144 occasionally reported bug whereby a message would remain on the queue
7145 longer than the retry timeout, but would be bounced if a delivery was
7146 forced. I say "may" because I never totally pinned down the problem;
7147 setting up timeout/retry tests is difficult. See also the next item.
7149 79. The ultimate address timeout was not being applied to errors that involved
7150 a combination of host plus message (for example, a timeout on a MAIL
7151 command). When an address resolved to a number of possible hosts, and they
7152 were not all tried for each delivery (e.g. because of hosts_max_try), a
7153 message could remain on the queue longer than the retry timeout.
7155 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
7158 81. Fixed an obscure SMTP outgoing bug which required at least the following
7159 conditions: (a) there was another message waiting for the same server;
7160 (b) the server returned 5xx to all RCPT commands in the first message so
7161 that the message was not completed; (c) the server dropped the connection
7162 or gave a negative response to the RSET that Exim sends to abort the
7163 transaction. The observed case was a dropped connection after DATA that had
7164 been sent in pipelining mode. That is, the server had advertised PIPELINING
7165 but was not implementing it correctly. The effect of the bug was incorrect
7166 behaviour, such as trying another host, and this could lead to a crash.
7172 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
7173 and daemon.c were passed as pointers to ints; they should have been
7174 pointers to socklen_t variables (which are typically unsigned ints).
7176 2. Some signed/unsigned type warnings in the os.c file for Linux have been
7179 3. Fixed a really odd bug that affected only the testing scheme; patching a
7180 certain fixed string in the binary changed the value of another string that
7181 happened to be identical to the end of the original first string.
7183 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
7184 name", it returns that address as the IP address. On some operating
7185 systems (e.g. Solaris), it also passes back the IP address string as the
7186 "host name". However, on others (e.g. Linux), it passes back an empty
7187 string. Exim wasn't checking for this, and was changing the host name to an
7188 empty string, assuming it had been canonicalized.
7190 5. Although rare, it is permitted to have more than one PTR record for a given
7191 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
7192 all the names associated with an address, because they do in Solaris.
7193 However, it seems that they do not in Linux for data that comes from the
7194 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
7195 I found this out when I moved to a new Linux workstation and tried to run
7196 the Exim test suite.
7198 To get round this problem I have changed the code so that it now does its
7199 own call to the DNS to look up PTR records when searching for a host name.
7200 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
7201 addresses that are only in /etc/hosts are still found.
7203 This behaviour is, however, controlled by an option called host_lookup_
7204 order, which defaults to "bydns:byaddr". If people want to use the other
7205 order, or indeed, just use one or the other means of lookup, they can
7206 specify it in this variable.
7208 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
7209 some operating systems, this comes back from gethostbyaddr() as an empty
7210 string, and this is what Exim used to test for. However, it seems that in
7211 other systems, "." is yielded. Exim now tests for this case too.
7213 7. The values of check_spool_space and check_log_space are now held internally
7214 as a number of kilobytes instead of an absolute number of bytes. If a
7215 numbers is specified without 'K' or 'M', it is rounded up to the nearest
7216 kilobyte. This means that much larger values can be stored.
7218 8. Exim monitor: an attempt to get the action menu when not actually pointing
7219 at a message produces an empty menu entitled "No message selected". This
7220 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
7221 no entries in it ("Shell widget menu has zero width and/or height"). So I
7222 have added a single, blank menu entry in this case.
7224 9. Added ${quote_local_part.
7226 10. MIME decoding is now applied to the contents of Subject: header lines when
7229 11. Now that a reference to $sender_host_address automatically causes a reverse
7230 lookup to occur if necessary (4.13/18), there is no need to arrange for a
7231 host lookup before query-style lookups in lists that might use this
7232 variable. This has therefore been abolished, and the "net-" prefix is no
7233 longer necessary for query-style lookups.
7235 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
7236 have been a typo for LFLAGS, so it has been changed.
7238 13. The install script calls Exim with "-C /dev/null" in order to find the
7239 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
7240 to be output. However, since Exim outputs its version number before the
7241 error, it didn't break the script. It just looked ugly. I fixed this by
7242 always allowing "-C /dev/null" if the caller is root.
7244 14. Ignore overlarge ACL variable number when reading spool file - insurance
7245 against a later release with more variables having written the file.
7247 15. The standard form for an IPv6 address literal was being rejected by EHLO.
7248 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
7249 this, as well as the form without the "IPv6" on the front.
7251 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
7252 OS/Makefile-Darwin file.
7254 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
7255 only to LDAP libraries that do not have LDAP_OPT_DEREF.
7257 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
7258 variable. However, some OS use ints for inodes. I've added cast to long int
7259 to get rid of the compiler warning.
7261 19. I had forgotten to lock out "/../" in configuration file names when
7262 ALT_CONFIG_PREFIX was set.
7264 20. Routers used for verification do not need to specify transports. However,
7265 if such a router generated a host list, and callout was configured, Exim
7266 crashed, because it could not find a port number from the (non-existent)
7267 transport. It now assumes port 25 in this circumstance.
7269 21. Added the -t option to exigrep.
7271 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
7272 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
7273 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
7274 them is set, LOOKUP_LSEARCH is forced.
7276 23. "exim -bV" now outputs a list of lookups that are included in the binary.
7278 24. Added sender and host information to the "rejected by local_scan()" log
7279 line; previously there was no indication of these.
7281 25. Added .include_if_exists.
7283 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
7284 Linux. It turns out that not all file systems support this. Apparently some
7285 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
7286 do it.) To cope with this, the error EINVAL, which means that sync-ing is
7287 not supported on the file descriptor, is now ignored when Exim is trying to
7288 sync a directory. This applies only to Linux.
7290 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
7292 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
7293 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
7296 29. Added support for maildirsize files from supplied patch (modified a bit).
7298 30. The use of :fail: followed by an empty string could lead Exim to respond to
7299 sender verification failures with (e.g.):
7301 550 Verification failed for <xxx>
7302 550 Sender verify failed
7304 where the first response line was missing the '-' that indicates it is not
7305 the final line of the response.
7307 31. The loop for finding the name of the user that called Exim had a hardwired
7308 limit of 10; it now uses the value of finduser_retries, which is used for
7309 all other user lookups.
7311 32. Added $received_count variable, available in data and not_smtp ACLs, and at
7314 33. Exim was neglecting to zero errno before one call of strtol() when
7315 expanding a string and expecting an integer value. On some systems this
7316 resulted in spurious "integer overflow" errors. Also, it was casting the
7317 result into an int without checking.
7319 34. Testing for a connection timeout using "timeout_connect" in the retry rules
7320 did not work. The code looks as if it has *never* worked, though it appears
7321 to have been documented since at least release 1.62. I have made it work.
7323 35. The "timeout_DNS" error in retry rules, also documented since at least
7324 1.62, also never worked. As it isn't clear exactly what this means, and
7325 clearly it isn't a major issue, I have abolished the feature by treating it
7326 as "timeout", and writing a warning to the main and panic logs.
7328 36. The display of retry rules for -brt wasn't always showing the error code
7331 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
7332 timeout_connect_A, timeout_connect_MX.
7334 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
7335 to the empty sender.
7337 39. The daemon was not analysing the content of -oX till after it had closed
7338 stderr and disconnected from the controlling terminal. This meant that any
7339 syntax errors were only noted on the panic log, and the return code from
7340 the command was 0. By re-arranging the code a little, I've made the
7341 decoding happen first, so such errors now appear on stderr, and the return
7342 code is 1. However, the actual setting up of the sockets still happens in
7343 the disconnected process, so errors there are still only recorded on the
7346 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
7347 connections (as happens on some IP stacks) was logged at start up time as
7348 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
7349 it from "IPv6 and IPv4", which means that two separate sockets are being
7352 41. The debug output for gethostbyname2() or getipnodebyname() failures now
7353 says whether AF_INET or AF_INET6 was passed as an argument.
7355 42. Exiwhat output was messed up when time zones were included in log
7358 43. Exiwhat now gives more information about the daemon's listening ports,
7359 and whether -tls-on-connect was used.
7361 44. The "port" option of the smtp transport is now expanded.
7363 45. A "message" modifier in a "warn" statement in a non-message ACL was being
7364 silently ignored. Now an error message is written to the main and panic
7367 46. There's a new ACL modifier called "logwrite" which writes to a log file
7368 as soon as it is encountered.
7370 47. Added $local_user_uid and $local_user_gid at routing time.
7372 48. Exim crashed when trying to verify a sender address that was being
7375 49. Exim was recognizing only a space character after ".include". It now also
7376 recognizes a tab character.
7378 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
7379 extracting the relevant information from the specification. The man page no
7380 longer contains scrambled data for the -d option, and I've added a section
7381 at the front about calling Exim under different names.
7383 51. Added "extra_headers" argument to the "mail" command in filter files.
7385 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
7388 53. Installed eximstats 1.29.
7390 54. Added transport_filter_timeout as a generic transport option.
7392 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
7393 Cc: header lines. This was required by RFC 822, but it not required by RFC
7396 56. Exim used to add From:, Date:, and Message-Id: header lines to any
7397 incoming messages that did not have them. Now it does so only if the
7398 message originates locally, that is, if there is no associated remote host
7399 address. When Resent- header lines are present, this applies to the Resent-
7400 lines rather than the non-Resent- lines.
7402 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
7403 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
7405 58. Messages for configuration errors now include the name of the main
7406 configuration file - useful now that there may be more than one file in a
7407 list (.included file names were always shown).
7409 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
7410 for those rare installations that do not start the daemon as root or run it
7411 setuid root. I've cut out the call to initgroups() if the daemon is not
7414 60. The Exim user and group can now be bound into the binary as text strings
7415 that are looked up at the start of Exim's processing.
7417 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
7419 62. Added $mailstore_basename variable.
7421 63. Installed patch to sieve.c from Michael Haardt.
7423 64. When Exim failed to open the panic log after failing to open the main log,
7424 the original message it was trying to log was written to stderr and debug
7425 output, but if they were not available (the usual case in production), it
7426 was lost. Now it is written to syslog before the two lines that record the
7427 failures to open the logs.
7429 65. Users' Exim filters run in subprocesses under the user's uid. It is
7430 possible for a "deliver" command or an alias in a "personal" command to
7431 provoke an address rewrite. If logging of address rewriting is configured,
7432 this fails because the process is not running as root or exim. There may be
7433 a better way of dealing with this, but for the moment (because 4.30 needs
7434 to be released), I have disabled address rewrite logging when running a
7435 filter in a non-root, non-exim process.
7441 1. The buildconfig auxiliary program wasn't quoting the value set for
7442 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
7443 not defined. This bug was masked in 4.22 by the effect that was fixed in
7446 2. Some messages that were rejected after a message id was allocated were
7447 shown as "incomplete" by exigrep. It no longer does this for messages that
7448 are rejected by local_scan() or the DATA or non-SMTP ACLs.
7450 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
7451 have allow_domain_literals set, the ID did not get logged in the <= line.
7452 Domain literals are now always recognized in Message-ID: header lines.
7454 4. The first argument for a ${extract expansion item is the key name or field
7455 number. Leading and trailing spaces in this item were not being ignored,
7456 causing some misleading effects.
7458 5. When deliver_drop_privilege was set, single queue runner processes started
7459 manually (i.e. by the command "exim -q") or by the daemon (which uses the
7460 same command in the process it spins off) were not dropping privilege.
7462 6. When the daemon running as "exim" started a queue runner, it always
7463 re-executed Exim in the spun-off process. This is a waste of effort when
7464 deliver_drop_privilege is set. The new process now just calls the
7465 queue-runner function directly.
7471 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
7474 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
7475 ignored. Though the use of -oP was forcing the writing of a pid file, it
7476 was always written to the default place.
7478 3. If the message "no IP address found for host xxxx" is generated during
7479 incoming verification, it is now followed by identification of the incoming
7480 connection (so you can more easily find what provoked it).
7482 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
7484 5. Added some features to "harden" Exim a bit more against certain attacks:
7486 (a) There is now a build-time option called FIXED_NEVER_USERS that can
7487 be put in Local/Makefile. This is like the never_users runtime option,
7488 but it cannot be overridden. The default setting is "root".
7490 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
7491 prefix string with which any file named in a -C command line option
7494 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
7495 is retained for -C and -D only if the caller of Exim is root. Without
7496 it, the exim user may also use -C and -D and retain privilege.
7498 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
7499 command line option is disabled.
7501 6. Macro names set by the -D option must start with an upper case letter, just
7502 like macro names defined in the configuration file.
7504 7. Added "dereference=" facility to LDAP.
7506 8. Two instances of the typo "uknown" in the source files are fixed.
7508 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
7509 the Configure-Makefile script screwed up while processing it.
7511 10. Incorporated PCRE 4.4.
7513 11. The SMTP synchronization check was not operating right at the start of an
7514 SMTP session. For example, it could not catch a HELO sent before the client
7515 waited for the greeting. There is now a check for outstanding input at the
7516 point when the greeting is written. Because of the duplex, asynchronous
7517 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
7518 way, but not yet received, when the check is performed.
7520 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
7521 on TCP/IP sockets, because this apparently causes some broken clients to
7524 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
7525 unchanged) from the Cygwin maintainer.
7527 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
7528 was supported instead of testing for mbx. Effectively a typo.
7530 15. The spa authenticator server code was not checking that the input it
7531 received was valid base64.
7533 16. The debug output line for the "set" modifier in ACLs was not showing the
7534 name of the variable that was being set.
7536 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
7538 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
7539 lookup on its own. Something else was needed to trigger the lookup. For
7540 example, a match in host_lookup or the need for a host name in a host list.
7541 Now, if $sender_host_name is referenced and the host name has not yet been
7542 looked up, a lookup is performed. If the lookup fails, the variable remains
7543 empty, and $host_lookup_failed is set to "1".
7545 19. Added "eqi" as a case-independent comparison operator.
7547 20. The saslauthd authentication condition could segfault if neither service
7548 nor realm was specified.
7550 21. If an overflowing value such as "2048M" was set for message_size_limit, the
7551 error message that was logged was misleading, and incoming SMTP
7552 connections were dropped. The message is now more accurate, and temporary
7553 errors are given to SMTP connections.
7555 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
7556 (except RSET) with a 421 error, until QUIT is received. However, it was
7557 failing to send a response to QUIT.
7559 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
7560 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
7561 ACL is now run after the helo_try_verify_hosts code.
7563 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
7564 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
7565 case-independent, so other case variants are also recognized). Apparently
7566 some systems use these upper case variants.
7568 25. If more than two messages were waiting for the same host, and a transport
7569 filter was specified for the transport, Exim sent two messages over the
7570 same TCP/IP connection, and then failed with "socket operation on non-
7571 socket" when it tried to send the third.
7573 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
7575 27. The extern definition of crypt16() in expand.c was not being excluded when
7576 the OS had its own crypt16() function.
7578 28. Added bounce_return_body as a new option, and bounce_return_size_limit
7579 as a preferred synonym for return_size_limit, both as an option and as an
7582 29. Added LIBS=-liconv to OS/Makefile-OSF1.
7584 30. Changed the default configuration ACL to relax the local part checking rule
7585 for addresses that are not in any local domains. For these addresses,
7586 slashes and pipe symbols are allowed within local parts, but the sequence
7587 /../ is explicitly forbidden.
7589 31. SPA server authentication was not clearing the challenge buffer before
7592 32. log_message in a "warn" ACL statement was writing to the reject log as
7593 well as to the main log, which contradicts the documentation and doesn't
7594 seem right (because no rejection is happening). So I have stopped it.
7596 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
7597 However, I am unable to do any testing of this.
7599 34. Fixed an infelicity in the appendfile transport. When checking directories
7600 for a mailbox, to see if any needed to be created, it was accidentally
7601 using path names with one or more superfluous leading slashes; tracing
7602 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
7604 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
7605 contents are added to the log line that is written for every discarded
7606 recipient. (Previously a log_message setting was ignored.)
7608 36. The ${quote: operator now quotes the string if it is empty.
7610 37. The install script runs exim in order to find its version number. If for
7611 some reason other than non-existence or emptiness, which it checks, it
7612 could not run './exim', it was installing it with an empty version number,
7613 i.e. as "exim-". This error state is now caught, and the installation is
7616 38. An argument was missing from the function that creates an error message
7617 when Exim fails to connect to the socket for saslauthd authentication.
7618 This could cause Exim to crash, or give a corrupted message.
7620 39. Added isip, isip4, and isip6 to ${if conditions.
7622 40. The ACL variables $acl_xx are now saved with the message, and can be
7623 accessed later in routers, transports, and filters.
7625 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
7626 strings in the file are not string-expanded.
7628 42. If a MAIL command specified a SIZE value that was too large to fit into an
7629 int variable, the check against message_size_limit failed. Such values are
7630 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
7631 day this will have to be increased, but I don't think I want to be around
7632 when emails are that large.
7639 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
7640 iconv() is not standard in FreeBSD.
7642 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
7643 IPv6 enabled. The observed symptom was a segmentation fault on return from
7644 the function os_common_find_running_interfaces() in src/os.c.
7646 3. In the check_special_case() function in daemon.c I had used "errno" as an
7647 argument name, which causes warnings on some systems. This was basically a
7648 typo, since it was named "eno" in the comments!
7650 4. The code that waits for the clock to tick (at a resolution of some fraction
7651 of a second) so as to ensure message-id uniqueness was always waiting for
7652 at least one whole tick, when it could have waited for less. [This is
7653 almost certainly not relevant at current processor speeds, where it is
7654 unlikely to ever wait at all. But we try to future-proof.]
7656 5. The function that sleeps for a time interval that includes fractions of a
7657 second contained a race. It did not block SIGALRM between setting the
7658 timer, and suspending (a couple of lines later). If the interval was short
7659 and the sigsuspend() was delayed until after it had expired, the suspension
7660 never ended. On busy systems this could lead to processes getting stuck for
7663 6. Some uncommon configurations may cause a lookup to happen in a queue runner
7664 process, before it forks any delivery processes. The open lookup caching
7665 mechanism meant that the open file or database connection was passed into
7666 the delivery process. The problem was that delivery processes always tidy
7667 up cached lookup data. This could cause a problem for the next delivery
7668 process started by the queue runner, because the external queue runner
7669 process does not know about the closure. So the next delivery process
7670 still has data in the lookup cache. In the case of a file lookup, there was
7671 no problem because closing a file descriptor in a subprocess doesn't affect
7672 the parent. However, if the lookup was caching a connection to a database,
7673 the connection was closed, and the second delivery process was likely to
7674 see errors such as "PGSQL: query failed: server closed the connection
7675 unexpectedly". The problem has been fixed by closing all cached lookups
7676 in a queue runner before running a delivery process.
7678 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
7679 seem to have the "const" qualifier which it has on other OS. I've
7682 8. Change 4.21/2 was too strict. It is only if there are two authenticators
7683 *of the same type* (client or server) with the same public name that an
7684 error should be diagnosed.
7686 9. When Exim looked up a host name for an IP address, but failed to find the
7687 original IP address when looking up the host name (a safety check), it
7688 output the message "<ip address> does not match any IP for NULL", which was
7689 confusing, to say the least. The bug was that the host name should have
7690 appeared instead of "NULL".
7692 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
7693 user that is built into the binary, and the -C or -D options is used, root
7694 privilege is dropped before the configuration file is read. In addition,
7695 logging is switched to stderr instead of the normal log files. If the
7696 configuration then re-defines the Exim user, the unprivileged environment
7697 is probably not what is expected, so Exim logs a panic warning message (but
7700 However, if deliver_drop_privilege is set, the unprivileged state may well
7701 be exactly what is intended, so the warning has been cut out in that case,
7702 and Exim is allowed to try to write to its normal log files.
7708 1. smtp_return_error_details was not giving details for temporary sender
7709 or receiver verification errors.
7711 2. Diagnose a configuration error if two authenticators have the same public
7714 3. Exim used not to create the message log file for a message until the first
7715 delivery attempt. This could be confusing when incoming messages were held
7716 for policy or load reasons. The message log file is now created at the time
7717 the message is received, and an initial "Received" line is written to it.
7719 4. The automatically generated man page for command line options had a minor
7720 bug that caused no ill effects; however, a more serious problem was that
7721 the procedure for building the man page automatically didn't always
7722 operate. Consequently, release 4.20 contains an out-of-date version. This
7723 shouldn't happen again.
7725 5. When building Exim with embedded Perl support, the script that builds the
7726 Makefile was calling 'perl' to find its compile-time parameters, ignoring
7727 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
7729 6. The freeze_tell option was not being used for messages that were frozen on
7730 arrival, either by an ACL or by local_scan().
7732 7. Added the smtp_incomplete_transaction log selector.
7734 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
7735 was accepting AUTH without a new EHLO.
7737 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
7738 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
7741 10. Logging of TCP/IP connections (when configured) now happens in the main
7742 daemon process instead of the child process, so that the TCP/IP connection
7743 count is more accurate (but it can never be perfect).
7745 11. The use of "drop" in a nested ACL was not being handled correctly in the
7746 outer ACL. Now, if condition failure induced by the nested "drop" causes
7747 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
7748 or "require"), the connection is dropped.
7750 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
7751 that yield "discard" can now be used with an "accept" or a "discard" verb,
7752 but an error is generated for any others (because I can't see a useful way
7753 to define what should happen).
7755 13. When an ACL is read dynamically from a file (or anywhere else), the lines
7756 are now processed in the same way as lines in the Exim configuration file.
7757 In particular, continuation lines are supported.
7759 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
7761 15. Added -ti meaning -t -i.
7763 16. Check for letters, digits, hyphens, and dots in the names of dnslist
7764 domains, and warn by logging if others are found.
7766 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
7767 returned from GIFCONF when Exim is trying to find the list of interfaces on
7768 a host. The code in os.c has been modified to copy each ifreq to an aligned
7769 structure in all cases.
7771 Also, in some cases, the returned ifreq's were being copied to a 'struct
7772 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
7773 means the last couple of bytes of an IPv6 address could be chopped if the
7774 ifreq contained only a normal sockaddr (14 bytes storage).
7776 18. Named domain lists were not supported in the hosts_treat_as_local option.
7777 An entry such as +xxxx was not recognized, and was treated as a literal
7780 19. Ensure that header lines added by a DATA ACL are included in the reject log
7781 if the ACL subsequently rejects the message.
7783 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
7784 MD5 (which is deprecated).
7786 21. When testing a filter file using -bf, Exim was writing a message when it
7787 took the sender from a "From " line in the message, but it was not doing so
7788 when it took $return_path from a Return-Path: header line. It now does.
7790 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
7791 with a valid header line field name (a series of printing characters
7792 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
7794 23. Changed "disc" in the source to "disk" to conform to the documentation and
7795 the book and for uniformity.
7797 24. Ignore Sendmail's -Ooption=value command line item.
7799 25. When execve() failed while trying to run a command in a pipe transport,
7800 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
7801 could be confused with a return value of 69 from the command itself. This
7802 has been changed to 127, the value the shell returns if it is asked to run
7803 a non-existent command. The wording for the related log line suggests a
7804 non-existent command as the problem.
7806 26. If received_header_text expands to an empty string, do not add a Received:
7807 header line to the message. (Well, it adds a token one on the spool, but
7808 marks it "old" so that it doesn't get used or transmitted.)
7810 27. Installed eximstats 1.28 (addition of -nt option).
7812 28. There was no check for failure on the call to getsockname() in the daemon
7813 code. This can fail if there is a shortage of resources on the system, with
7814 ENOMEM, for example. A temporary error is now given on failure.
7816 29. Contrary to the C standard, it seems that in some environments, the
7817 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
7818 program. Exim now does this explicitly; it affects the formatting of
7819 timestamps using strftime().
7821 30. If exiqsumm was given junk data, it threw up some uninitialized variable
7822 complaints. I've now initialized all the variables, to avoid this.
7824 32. Header lines added by a system filter were not being "seen" during
7825 transport-time rewrites.
7827 33. The info_callback() function passed to OpenSSL is set up with type void
7828 (*)(SSL *, int, int), as described somewhere. However, when calling the
7829 function (actually a macro) that sets it up, the type void(*)() is
7830 expected. I've put in a cast to prevent warnings from picky compilers.
7832 34. If a DNS black list lookup found a CNAME record, but there were no A
7833 records associated with the domain it pointed at, Exim crashed.
7835 35. If a DNS black list lookup returned more than one A record, Exim ignored
7836 all but the first. It now scans all returned addresses if a particular IP
7837 value is being sought. In this situation, the contents of the
7838 $dnslist_value variable are a list of all the addresses, separated by a
7841 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
7842 to accept a host name and all its aliases if the forward lookup for any of
7843 them yielded the IP address of the incoming connection. Now it accepts only
7844 those names whose forward lookup yields the correct IP address. Any other
7845 names are discarded. This closes a loophole whereby a rogue DNS
7846 administrator could create reverse DNS records to break through a
7847 wildcarded host restriction in an ACL.
7849 37. If a user filter or a system filter that ran in a subprocess used any of
7850 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
7851 the wrong values were passed to the pipe command ($thisaddress had the
7852 value of $0, $0 had the value of $1, etc). This bug was introduced by
7853 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
7855 38. Improved the line breaking for long SMTP error messages from ACLs.
7856 Previously, if there was no break point between 40 and 75 characters, Exim
7857 left the rest of the message alone. Two changes have been made: (a) I've
7858 reduced the minimum length to 35 characters; (b) if it can't find a break
7859 point between 35 and 75 characters, it looks ahead and uses the first one
7860 that it finds. This may give the occasional overlong line, but at least the
7861 remaining text gets split now.
7863 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
7864 file descriptors might be low, and that setting 1000 would always raise it.
7865 It turns out that in some environments, the limit is already over 1000 and
7866 that lowering it causes trouble. So now Exim takes care not to decrease it.
7868 40. When delivering a message, the value of $return_path is set to $sender_
7869 address at the start of routing (routers may change the value). By an
7870 oversight, this default was not being set up when an address was tested by
7871 -bt or -bv, which affected the outcome if any router or filter referred to
7874 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
7875 log only when "message" or "log_message" are set. However, if one of the
7876 conditions was an address verification, or a call to a nested ACL, the
7877 messages generated by the underlying test were being passed through. This
7878 no longer happens. The underlying message is available in $acl_verify_
7879 message for both "message" and "log_message" expansions, so it can be
7880 passed through if needed.
7882 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
7883 new expansion $bh_ to give the encoded byte string without charset
7884 translation. Translation happens only if iconv() is available; HAVE_ICONV
7885 indicates this at build time. HEADERS_CHARSET gives the charset to
7886 translate to; headers_charset can change it in the configuration, and
7887 "headers charset" can change it in an individual filter file.
7889 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
7890 that creates RFC 2047 encoded "words" labels them as that charset instead
7891 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
7892 expansion operator; (ii) when Exim creates a From: line for a local
7893 message; (iii) when a header line is rewritten to include a "phrase" part.
7895 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
7896 buggy, causing it to skip the first lines of messages whose message ID
7897 ended in 'D'. This would not have bitten before Exim release 4.14, because
7898 message IDs were unlikely to end in 'D' before then. The effect was to have
7899 incorrect size information for certain domains.
7901 45. #include "config.h" was missing at the start of the crypt16.c module. This
7902 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
7905 46. If there was a timeout during a "random" callout check, Exim treated it as
7906 a failure of the random address, and carried on sending RSET and the real
7907 address. If the delay was just some slowness somewhere, the response to the
7908 original RCPT would be taken as a response to RSET and so on, causing
7909 mayhem of various kinds.
7911 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
7912 when I implemented it. It didn't allow for the fact that some option values
7913 may legitimately be negative (e.g. size_addition), and it didn't even do
7914 the right test for positive values.
7916 48. Domain names in DNS records are case-independent. Exim always looks them up
7917 in lower case. Some resolvers return domain names in exactly the case they
7918 appear in the zone file, that is, they may contain uppercase letters. Not
7919 all resolvers do this - some return always lower case. Exim was treating a
7920 change of case by a resolver as a change of domain, similar to a widening
7921 of a domain abbreviation. This triggered its re-routing code and so it was
7922 trying to route what was effectively the same domain again. This normally
7923 caused routing to fail (because the router wouldn't handle the domain
7924 twice). Now Exim checks for this case specially, and just changes the
7925 casing of the domain that it ultimately uses when it transmits the message
7928 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
7931 50. If a filter generated a file delivery with a non-absolute name (possible if
7932 no home directory exists for the router), the forbid_file option was not
7935 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
7936 the existing equality matching.
7938 52. Exim was using ints instead of ino_t variables in some places where it was
7939 dealing with inode numbers.
7941 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
7942 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
7943 called TMPDIR, and if it finds it is different, it changes its value.
7945 54. The smtp_printf() function is now made available to local_scan() so
7946 additional output lines can be written before returning. There is also an
7947 smtp_fflush() function to enable the detection of a dropped connection.
7948 The variables smtp_input and smtp_batched_input are exported to
7951 55. Changed the default runtime configuration: the message "Unknown user"
7952 has been removed from the ACL, and instead placed on the localuser router,
7953 using the cannot_route_message feature. This means that any verification
7954 failures that generate their own messages won't get overridden. Similarly,
7955 the "Unrouteable address" message that was in the ACL for unverifiable
7956 relay addresses has also been removed.
7958 56. Added hosts_avoid_esmtp to the smtp transport.
7960 57. The exicyclog script was not checking for the esoteric option
7961 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
7962 will work only if exicyclog is run under the appropriate euid.
7964 58. Following a discussion on the list, the rules by which Exim recognises line
7965 endings on incoming messages have been changed. The -dropcr and drop_cr
7966 options are now no-ops, retained only for backwards compatibility. The
7967 following line terminators are recognized: LF CRLF CR. However, special
7968 processing applies to CR:
7970 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7971 nor a local message in the state where . is a terminator.
7973 (ii) If a bare CR is encountered in a header line, an extra space is added
7974 after the line terminator so as not to end the header. The reasoning
7975 behind this is that bare CRs in header lines are most likely either
7976 to be mistakes, or people trying to play silly games.
7978 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7979 was being incorrectly given as 18 bytes larger than it should have been.
7980 This is a VOB (very old bug).
7982 60. This may never have affected anything current, but just in case it has:
7983 When the local host is found other than at the start of a list of hosts,
7984 the local host, those with the same MX, and any that follow, are discarded.
7985 When the list in question was part of a longer list of hosts, the following
7986 hosts (not currently being processed) were also being discarded. This no
7987 longer happens. I'm not sure if this situation could ever has previously
7990 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7993 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7994 header. This is something that is recommended in a new Internet Draft, and
7995 is something that is documented as being done by Sendmail. There are two
7996 possible values. For messages generated by the autoreply transport, Exim
7999 Auto-Submitted: auto-replied
8001 whereas for all other generated messages (e.g. bounces) it adds
8003 Auto-Submitted: auto-generated
8005 63. The "personal" condition in filters now includes a test for the
8006 Auto-Submitted: header. If it contains the string "auto-" the message it
8007 not considered personal.
8009 64. Added rcpt_include_affixes as a generic transport option.
8011 65. Added queue_only_override (default true).
8013 66. Added the syslog_duplication option.
8015 67. If what should have been the first header line of a message consisted of
8016 a space followed by a colon, Exim was mis-interpreting it as a header line.
8017 It isn't of course - it is syntactically invalid and should therefore be
8018 treated as the start of the message body. The misbehaviour could have
8019 caused a number of strange effects, including loss of data in subsequent
8020 header lines, and spool format errors.
8022 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
8023 client host had authenticated. This control can now be exercised by an ACL
8024 for more flexibility.
8026 69. By default, callouts do not happen when testing with -bh. There is now a
8027 variant, -bhc, which does actually run the callout code, including
8028 consulting and updating the callout cache.
8030 70. Added support for saslauthd authentication, courtesy of Alexander
8033 71. If statvfs() failed on the spool or log directories while checking their
8034 size for availability, Exim confusingly gave the error "space shortage".
8035 Furthermore, in debugging mode it crashed with a floating point exception.
8036 These checks are done if check_{spool,log}_{space,inodes} are set, and when
8037 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
8038 serious problem, Exim now writes to the main and panic logs when this
8039 happens, with details of the failure. It then refuses to accept the
8040 incoming message, giving the message "spool directory problem" or "log
8041 directory problem" with a 421 code for SMTP messages.
8043 72. When Exim is about to re-exec itself, it ensures that the file descriptors
8044 0, 1, and 2 exist, because some OS complain for execs without them (see
8045 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
8046 descriptors. However, the code omitted to check that the open succeeded,
8047 causing mysterious errors if for some reason the permissions on /dev/null
8048 got screwed. Now Exim writes a message to the main and panic logs, and
8049 bombs out if it can't open /dev/null.
8051 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
8052 interact so that it is all more flexible. It is supposed to remain
8053 backwards compatible. Also added extra_local_interfaces.
8055 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
8056 to bomb out with an assertion failure - to the client this appears as a
8057 connection drop. This problem occurs in the part of the code that was taken
8058 from the Samba project. Fortunately, the assertion is in a very simple
8059 function, so I have fixed this by reproducing the function inline in the
8060 one place where it is called, and arranging for authentication to fail
8061 instead of killing the process with assert().
8063 75. The SPA client code was not working when the server requested OEM rather
8064 than Unicode encoding.
8066 76. Added code to make require_files with a specific uid setting more usable in
8067 the case where statting the file as root fails - usually a non-root-mounted
8068 NFS file system. When this happens and the failure is EACCES, Exim now
8069 forks a subprocess and does the per-uid checking as the relevant uid.
8071 77. Added process_log_path.
8073 78. If log_file_path was not explicitly set, a setting of check_log_space or
8074 check_log_inodes was ignored.
8076 79. If a space check for the spool or log partitions fails, the incident is now
8077 logged. Of course, in the latter case the data may get lost...
8079 80. Added the %p formatting code to string_format() so that it can be used to
8080 print addresses in debug_print(). Adjusted all the address printing in the
8081 debugging in store.c to use %p rather than %d.
8083 81. There was a concern that a line of code in smtp_in.c could overflow a
8084 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
8085 initially expressed, the concern was not well-founded, because trailing
8086 spaces are removed early. However, if the trailing spaces were followed by
8087 a NULL, they did not get removed, so the overflow was possible. Two fixes
8090 (a) I re-wrote the offending code in a cleaner fashion.
8091 (b) If an incoming SMTP command contains a NULL character, it is rejected
8094 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
8095 runs initgroups(), so that if the Exim user is in any additional groups,
8096 they will be used during message reception.
8102 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob