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 Avoid costly startup code when not strictly needed. This reduces time
10 for some exim process initialisations. It does mean that the logging
11 of TLS configuration problems is only done for the daemon startup.
13 JH/02 Early-pipelining support code is now included unless disabled in Makefile.
15 JH/03 DKIM verification defaults no long accept sha1 hashes, to conform to
16 RFC 8301. They can still be enabled, using the dkim_verify_hashes main
19 JH/04 Support CHUNKING from an smtp transport using a transport_filter, when
20 DKIM signing is being done. Previously a transport_filter would always
21 disable CHUNKING, falling back to traditional DATA.
23 JH/05 Regard command-line receipients as tainted.
25 JH/06 Bug 340: Remove the daemon pid file on exit, whe due to SIGTERM.
27 JH/07 Bug 2489: Fix crash in the "pam" expansion condition. It seems that the
28 PAM library frees one of the arguments given to it, despite the
29 documentation. Therefore a plain malloc must be used.
31 JH/08 Bug 2491: Use tainted buffers for the transport smtp context. Previously
32 on-stack buffers were used, resulting in a taint trap when DSN information
33 copied from a received message was written into the buffer.
35 JH/09 Bug 2493: Harden ARC verify against Outlook, whick has been seen to mix
36 the ordering of its ARC headers. This caused a crash.
38 JH/10 Bug 2492: Use tainted memory for retry record when needed. Previously when
39 a new record was being constructed with information from the peer, a trap
42 JH/11 Bug 2494: Unset the default for dmarc_tld_file. Previously a naiive
43 installation would get error messages from DMARC verify, when it hit the
44 nonexistent file indicated by the default. Distros wanting DMARC enabled
45 should both provide the file and set the option.
46 Also enforce no DMARC verification for command-line sourced messages.
48 JH/12 Fix an uninitialised flag in early-pipelining. Previously connections
49 could, depending on the platform, hang at the STARTTLS response.
51 JH/13 Bug 2498: Reset a counter used for ARC verify before handling another
52 message on a connection. Previously if one message had ARC headers and
53 the following one did not, a crash could result when adding an
54 Authentication-Results: header.
56 JH/14 Bug 2500: Rewind some of the common-coding in string handling between the
57 Exim main code and Exim-related utities. The introduction of taint
58 tracking also did many adjustments to string handling. Since then, eximon
59 frequently terminated with an assert failure.
61 JH/15 When PIPELINING, synch after every hundred or so RCPT commands sent and
62 check for 452 responses. This slightly helps the inefficieny of doing
63 a large alias-expansion into a recipient-limited target. The max_rcpt
64 transport option still applies (and at the current default, will override
65 the new feature). The check is done for either cause of synch, and forces
66 a fast-retry of all 452'd recipients using a new MAIL FROM on the same
67 connection. The new facility is not tunable at this time.
69 JH/16 Fix the variables set by the gsasl authenticator. Previously a pointer to
70 library live data was being used, so the results became garbage. Make
71 copies while it is still usable.
73 JH/17 Logging: when the deliver_time selector ise set, include the DT= field
74 on delivery deferred (==) and failed (**) lines (if a delivery was
75 attemtped). Previously it was only on completion (=>) lines.
77 JH/18 Authentication: the gsasl driver not provides the $authN variables in time
78 for the expansion of the server_scram_iter and server_scram_salt options.
80 WB/01 SPF: DNS lookups for the obsolete SPF RR type done by the libspf2 library
81 are now specifically given a NO_DATA response without hitting the system
82 resolver. The library goes on to do the now-standard TXT lookup.
83 Use of dnsdb lookups is not affected.
85 JH/19 Bug 2507: Modules: on handling a dynamic-module (lookups) open failure,
86 only retrieve the errormessage once. Previously two calls to dlerror()
87 were used, and the second one (for mainlog/paniclog) retrieved null
90 JH/20 Taint checking: disallow use of tainted data for
91 - the appendfile transport file and directory options
92 - the pipe transport command
93 - the autoreply transport file, log and once options
94 - file names used by the redirect router (including filter files)
96 - paths used by single-key lookups
97 Previously this was permitted.
99 JH/21 Bug 2501: Fix init call in the heimdal authenticator. Previously it
100 adjusted the size of a major service buffer; this failed because the
101 buffer was in use at the time. Change to a compile-time increase in the
102 buffer size, when this authenticator is compiled into exim.
104 JH/22 Taint-checking: move to safe-mode taint checking on all platforms. The
105 previous fast-mode was untenable in the face of glibs using mmap to
106 support larger malloc requests.
108 PP/01 Update the openssl_options possible values through OpenSSL 1.1.1c.
109 New values supported, if defined on system where compiled:
110 allow_no_dhe_kex, cryptopro_tlsext_bug, enable_middlebox_compat,
111 no_anti_replay, no_encrypt_then_mac, prioritize_chacha, tlsext_padding
113 JH/23 Performance improvement in the initial phase of a two-pass queue run. By
114 running a limited number of proceses in parallel, a benefit is gained. The
115 amount varies with the platform hardware and load. The use of the option
116 queue_run_in_order means we cannot do this, as ordering becomes
119 JH/24 Bug 2524: fix the cyrus_sasl auth driver gssapi usage. A previous fix
120 had introduced a string-copy (for ensuring NUL-termination) which was not
121 appropriate for that case, which can include embedded NUL bytes in the
122 block of data. Investigation showed the copy to actually be needless, the
123 data being length-specified.
125 JH/25 Fix use of concurrent TLS connections under GnuTLS. When a callout was
126 done during a receiving connection, and both used TLS, global info was
127 used rather than per-connection info for tracking the state of data
128 queued for transmission. This could result in a connection hang.
130 JH/26 Fix use of the SIZE parameter on MAIL commands, on continued connections.
131 Previously, when delivering serveral messages down a single connection
132 only the first would provide a SIZE. This was due to the size information
133 not being properly tracked.
135 JH/27 Bug 2530: When operating in a timezone with sub-minute offset, such as
136 TAI (at 37 seconds currently), pretend to be in UTC for time-related
137 expansion and logging. Previously, spurious values such as a future
138 minute could be seen.
140 JH/28 Bug 2533: Fix expansion of ${tr } item. When called in some situations
141 it could crash from a null-deref. This could also affect the
142 ${addresses: } operator and ${readsock } item.
144 JH/29 Bug 2537: Fix $mime_part_count. When a single connection had a non-mime
145 message following a mime one, the variable was not reset.
147 JH/30 When an pipelined-connect fails at the first response, assume incorrect
148 cached capability (perhaps the peer reneged?) and immediately retry in
151 JH/31 Fix spurious detection of timeout while writing to transport filter.
153 JH/32 Bug 2541: Fix segfault on bad cmdline -f (sender) argument. Previously
154 an attempt to copy the string was made before checking it.
156 JH/33 Fix the dsearch lookup to return an untainted result. Previously the
157 taint of the lookup key was maintained; we now regard the presence in the
158 filesystem as sufficient validation.
160 JH/34 Fix the readsocket expansion to not segfault when an empty "options"
161 argument is supplied.
163 JH/35 The dsearch lookup now requires that the directory is an absolute path.
164 Previously this was not checked, and nonempty relative paths made an
165 access under Exim's current working directory.
167 JH/36 Bug 2554: Fix msg:defer event for the hosts_max_try_hardlimit case.
168 Previously no event was raised.
170 JH/37 Bug 2552: Fix the check on spool space during reception to use the SIZE
171 parameter supplied by the sender MAIL FROM command. Previously it was
172 ignored, and only the check_spool_space option value for the required
175 JH/38 Fix $dkim_key_length. This should, after a DKIM verification, present
176 the size of the signing public-key. Previously it was instead giving
177 the size of the signature hash.
179 JH/39 DKIM verification: the RFC 8301 restriction on sizes of RSA keys is now
180 the default. See the (new) dkim_verify_min_keysizes option.
182 JH/40 Fix a memory-handling bug: when a connection carried multiple messages
183 and an ACL use a lookup for checking either the local_part or domain,
184 stale data could be accessed. Ensure that variable references are
185 dropped between messages.
187 JH/41 Bug 2571: Fix SPA authenticator. Running as a server, an offset supplied
188 by the client was not checked as pointing within response data before
189 being used. A malicious client could thus cause an out-of-bounds read and
190 possibly gain authentication. Fix by adding the check.
192 JH/42 Internationalisation: change the default for downconversion in the smtp
193 transport to be "if needed". Previously it was "as previously set" for
194 the message, which usually meant "if needed" for message-submission but
195 "no" for everything else. However, MTAs have been seen using SMTPUTF8
196 even when the envelope addresses did not need it, resulting in forwarding
197 failures to non-supporting MTAs. A downconvert in such cases will be
198 a no-op on the addresses, merely dropping the use of SMTPUTF8 by the
199 transport. The change does mean that addresses needing conversion will
200 be converted when previously a delivery failure would occur.
202 JH/43 Fix possible long line in DSN. Previously when a very long SMTP error
203 response was received it would be used unchecked in a fail-DSN, violating
204 standards on line-length limits. Truncate if needed.
210 JH/01 OpenSSL: With debug enabled output keying information sufficient, server
211 side, to decode a TLS 1.3 packet capture.
213 JH/02 OpenSSL: Suppress the sending of (stateful) TLS1.3 session tickets.
214 Previously the default library behaviour applied, sending two, each in
217 JH/03 Debug output for ACL now gives the config file name and line number for
220 JH/04 The default received_header_text now uses the RFC 8314 tls cipher clause.
222 JH/05 DKIM: ensure that dkim_domain elements are lowercased before use.
224 JH/06 Fix buggy handling of autoreply bounce_return_size_limit, and a possible
225 buffer overrun for (non-chunking) other transports.
227 JH/07 GnuTLS: Our use of late (post-handshake) certificate verification, under
228 TLS1.3, means that a server rejecting a client certificate is not visible
229 to the client until the first read of encrypted data (typically the
230 response to EHLO). Add detection for that case and treat it as a failed
231 TLS connection attempt, so that the normal retry-in-clear can work (if
232 suitably configured).
234 JB/01 Bug 2375: fix expansions of 822 addresses having comments in local-part
235 and/or domain. Found and fixed by Jason Betts.
237 JH/08 Add hardening against SRV & TLSA lookups the hit CNAMEs (a nonvalid
238 configuration). If a CNAME target was not a wellformed name pattern, a
241 JH/09 Logging: Fix initial listening-on line for multiple ports for an IP when
242 the OS reports them interleaved with other addresses.
244 JH/10 OpenSSL: Fix aggregation of messages. Previously, when PIPELINING was
245 used both for input and for a verify callout, both encrypted, SMTP
246 responses being sent by the server could be lost. This resulted in
247 dropped connections and sometimes bounces generated by a peer sending
250 JH/11 Harden plaintext authenticator against a badly misconfigured client-send
251 string. Previously it was possible to cause undefined behaviour in a
252 library routine (usually a crash). Found by "zerons".
254 JH/12 Bug 2384: fix "-bP smtp_receive_timeout". Previously it returned no
257 JH/13 Bug 2386: Fix builds with Dane under LibreSSL 2.9.0 onward. Some old
258 API was removed, so update to use the newer ones.
260 JH/14 Bug 1891: Close the log file if receiving a non-smtp message, without
261 any timeout set, is taking a long time. Previously we would hang on to a
262 rotated logfile "forever" if the input was arriving with long gaps
263 (a previous attempt to fix addressed lack, for a long time, of initial
266 HS/01 Bug 2390: Use message_id for tempfile creation to avoid races in a
267 shared (NFS) environment. The length of the tempfile name is now
268 4 + 16 ("hdr.$message_exim_id") which might break on file
269 systems which restrict the file name length to lower values.
272 HS/02 Bug 2390: Use message_id for tempfile creation to avoid races in a
273 shared (NFS) environment.
275 HS/03 Bug 2392: exigrep does case sensitive *option* processing (as it
276 did for all versions <4.90). Notably -M, -m, --invert, -I may be
279 JH/15 Use unsigned when creating bitmasks in macros, to avoid build errors
280 on some platforms for bit 31.
282 JH/16 GnuTLS: rework ciphersuite strings under recent library versions. Thanks
283 to changes apparently associated with TLS1.3 handling some of the APIs
284 previously used were either nonfunctional or inappropriate. Strings
285 like TLS1.3:ECDHE_SECP256R1__RSA_PSS_RSAE_SHA256__AES_256_GCM__AEAD:256
286 and TLS1.2:ECDHE_SECP256R1__RSA_SHA256__AES_128_CBC__SHA256:128 replace
287 the previous TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256 .
288 This affects log line X= elements, the $tls_{in,out}_cipher variables,
289 and the use of specific cipher names in the encrypted= ACL condition.
291 JH/17 OpenSSL: the default openssl_options now disables ssl_v3.
293 JH/18 GnuTLS: fix $tls_out_ocsp under hosts_request_ocsp. Previously the
294 verification result was not updated unless hosts_require_ocsp applied.
296 JH/19 Bug 2398: fix listing of a named-queue. Previously, even with the option
297 queue_list_requires_admin set to false, non-admin users were denied the
300 JH/20 Bug 2389: fix server advertising of usable certificates, under GnuTLS in
301 directory-of-certs mode. Previously they were advertised despite the
304 JH/21 The smtp transport option "hosts_noproxy_tls" is now unset by default.
305 A single TCP connection by a client will now hold a TLS connection open
306 for multiple message deliveries, by default. Previoud the default was to
309 JH/22 The smtp transport option "hosts_try_dane" now enables all hosts by
310 default. If built with the facility, DANE will be used. The facility
311 SUPPORT_DANE is now enabled in the prototype build Makefile "EDITME".
313 JH/23 The build default is now for TLS to be included; the SUPPORT_TLS define
314 is replaced with DISABLE_TLS. Either USE_GNUTLS or (the new) USE_OPENSSL
315 must be defined and you must still, unless you define DISABLE_TLS, manage
316 the the include-dir and library-file requirements that go with that
317 choice. Non-TLS builds are still supported.
319 JH/24 Fix duplicated logging of peer name/address, on a transport connection-
322 JH/25 The smtp transport option "hosts_try_fastopen" now enables all hosts by
323 default. If the platform supports and has the facility enabled, it will
324 be requested on all coneections.
326 JH/26 The PIPE_CONNECT facility is promoted from experimental status and is now
327 controlled by the build-time option SUPPORT_PIPE_CONNECT.
329 PP/01 Unbreak heimdal_gssapi, broken in 4.92.
331 JH/27 Bug 2404: Use the main-section configuration option "dsn_from" for
332 success-DSN messages. Previously the From: header was always the default
333 one for these; the option was ignored.
335 JH/28 Fix the timeout on smtp response to apply to the whole response.
336 Previously it was reset for every read, so a teergrubing peer sending
337 single bytes within the time limit could extend the connection for a
338 long time. Credit to Qualsys Security Advisory Team for the discovery.
340 JH/29 Fix DSN Final-Recipient: field. Previously it was the post-routing
341 delivery address, which leaked information of the results of local
342 forwarding. Change to the original envelope recipient address, per
345 JH/30 Bug 2411: Fix DSN generation when RFC 3461 failure notification is
346 requested. Previously not bounce was generated and a log entry of
347 error ignored was made.
349 JH/31 Avoid re-expansion in ${sort } expansion. (CVE-2019-13917)
351 JH/32 Introduce a general tainting mechanism for values read from the input
352 channel, and values derived from them. Refuse to expand any tainted
353 values, to catch one form of exploit.
355 JH/33 Bug 2413: Fix dkim_strict option. Previously the expansion result
356 was unused and the unexpanded text used for the test. Found and
357 fixed by Ruben Jenster.
359 JH/34 Fix crash after TLS shutdown. When the TCP/SMTP channel was left open,
360 an attempt to use a TLS library read routine dereffed a nul pointer,
363 JH/35 Bug 2409: filter out-of-spec chars from callout response before using
364 them in our smtp response.
366 JH/36 Have the general router option retry_use_local_part default to true when
367 any of the restrictive preconditions are set (to anything). Previously it
368 was only for check_local user. The change removes one item of manual
369 configuration which is required for proper retries when a remote router
370 handles a subset of addresses for a domain.
372 JH/37 Appendfile: when evaluating quota use (non-quota_size_regex) take the file
373 link count into consideration.
375 HS/04 Fix handling of very log lines in -H files. If a -<key> <value> line
376 caused the extension of big_buffer, the following lines were ignored.
378 JH/38 Bug 1395: Teach the DNS negative-cache about TTL value from the SOA in
379 accordance with RFC 2308. Previously there was no expiry, so a longlived
380 receive process (eg. due to ACL delays) versus a short SOA value could
383 HS/05 Handle trailing backslash gracefully. (CVE-2019-15846)
385 JH/39 Promote DMARC support to mainline.
387 JH/40 Bug 2452: Add a References: header to DSNs.
389 JH/41 With GnuTLS 3.6.0 (and later) do not attempt to manage Diffie-Hellman
390 parameters. The relevant library call is documented as "Deprecated: This
391 function is unnecessary and discouraged on GnuTLS 3.6.0 or later. Since
392 3.6.0, DH parameters are negotiated following RFC7919."
394 HS/06 Change the default of dnssec_request_domains to "*"
396 JH/42 Bug 2545: Fix CHUNKING for all RCPT commands rejected. Previously we
397 carried on and emitted a BDAT command, even when PIPELINING was not
400 JH/43 Bug 2465: Fix taint-handling in dsearch lookup. Previously a nontainted
401 buffer was used for the filename, resulting in a trap when tainted
402 arguments (eg. $domain) were used.
404 JH/44 With OpenSSL 1.1.1 (onwards) disable renegotiation for TLS1.2 and below;
405 recommended to avoid a possible server-load attack. The feature can be
406 re-enabled via the openssl_options main cofiguration option.
408 JH/45 local_scan API: documented the current smtp_printf() call. This changed
409 for version 4.90 - adding a "more data" boolean to the arguments.
410 Bumped the ABI version number also, this having been missed previously;
411 release versions 4.90 to 4.92.3 inclusive were effectively broken in
412 respect of usage of smtp_printf() by either local_scan code or libraries
413 accessed via the ${dlfunc } expansion item. Both will need coding
414 adjustment for any calls to smtp_printf() to match the new function
415 signature; a FALSE value for the new argument is always safe.
417 JH/46 FreeBSD: fix use of the sendfile() syscall. The shim was not updating
418 the file-offset (which the Linux syscall does, and exim expects); this
419 resulted in an indefinite loop.
421 JH/47 ARC: fix crash in signing, triggered when a configuration error failed
422 to do ARC verification. The Authentication-Results: header line added
423 by the configuration then had no ARC item.
429 JH/01 Remove code calling the customisable local_scan function, unless a new
430 definition "HAVE_LOCAL_SCAN=yes" is present in the Local/Makefile.
432 JH/02 Bug 1007: Avoid doing logging from signal-handlers, as that can result in
433 non-signal-safe functions being used.
435 JH/03 Bug 2269: When presented with a received message having a stupidly large
436 number of DKIM-Signature headers, disable DKIM verification to avoid
437 a resource-consumption attack. The limit is set at twenty.
439 JH/04 Add variables $arc_domains, $arc_oldest_pass for ARC verify. Fix the
440 report of oldest_pass in ${authres } in consequence, and separate out
441 some descriptions of reasons for verification fail.
443 JH/05 Bug 2273: Cutthrough delivery left a window where the received messsage
444 files in the spool were present and unlocked. A queue-runner could spot
445 them, resulting in a duplicate delivery. Fix that by doing the unlock
446 after the unlink. Investigation by Tim Stewart. Take the opportunity to
447 add more error-checking on spoolfile handling while that code is being
450 PP/01 Refuse to open a spool data file (*-D) if it's a symlink.
451 No known attacks, no CVE, this is defensive hardening.
453 JH/06 Bug 2275: The MIME ACL unlocked the received message files early, and
454 a queue-runner could start a delivery while other operations were ongoing.
455 Cutthrough delivery was a common victim, resulting in duplicate delivery.
456 Found and investigated by Tim Stewart. Fix by using the open message data
457 file handle rather than opening another, and not locally closing it (which
458 releases a lock) for that case, while creating the temporary .eml format
459 file for the MIME ACL. Also applies to "regex" and "spam" ACL conditions.
461 JH/07 Bug 177: Make a random-recipient callout success visible in ACL, by setting
462 $sender_verify_failure/$recipient_verify_failure to "random".
464 JH/08 When generating a selfsigned cert, use serial number 1 since zero is not
467 JH/09 Bug 2274: Fix logging of cmdline args when starting in an unlinked cwd.
468 Previously this would segfault.
470 JH/10 Fix ARC signing for case when DKIM signing failed. Previously this would
473 JH/11 Bug 2264: Exim now only follows CNAME chains one step by default. We'd
474 like zero, since the resolver should be doing this for us, But we need one
475 as a CNAME but no MX presence gets the CNAME returned; we need to check
476 that doesn't point to an MX to declare it "no MX returned" rather than
477 "error, loop". A new main option is added so the older capability of
478 following some limited number of chain links is maintained.
480 JH/12 Add client-ip info to non-pass iprev ${authres } lines.
482 JH/13 For receent Openssl versions (1.1 onward) use modern generic protocol
483 methods. These should support TLS 1.3; they arrived with TLS 1.3 and the
484 now-deprecated earlier definitions used only specified the range up to TLS
485 1.2 (in the older-version library docs).
487 JH/14 Bug 2284: Fix DKIM signing for body lines starting with a pair of dots.
489 JH/15 Rework TLS client-side context management. Stop using a global, and
490 explicitly pass a context around. This enables future use of TLS for
491 connections to service-daemons (eg. malware scanning) while a client smtp
492 connection is using TLS; with cutthrough connections this is quite likely.
494 JH/16 Fix ARC verification to do AS checks in reverse order.
496 JH/17 Support a "tls" option on the ${readsocket } expansion item.
498 JH/18 Bug 2287: Fix the protocol name (eg utf8esmtp) for multiple messages
499 using the SMTPUTF8 option on their MAIL FROM commands, in one connection.
500 Previously the "utf8" would be re-prepended for every additional message.
502 JH/19 Reject MAIL FROM commands with SMTPUTF8 when the facility was not advertised.
503 Previously thery were accepted, resulting in issues when attempting to
504 forward messages to a non-supporting MTA.
506 PP/02 Let -n work with printing macros too, not just options.
508 JH/20 Bug 2296: Fix cutthrough for >1 address redirection. Previously only
509 one parent address was copied, and bogus data was used at delivery-logging
510 time. Either a crash (after delivery) or bogus log data could result.
511 Discovery and analysis by Tim Stewart.
513 PP/03 Make ${utf8clean:} expansion operator detect incomplete final character.
514 Previously if the string ended mid-character, we did not insert the
515 promised '?' replacement.
517 PP/04 Documentation: current string operators work on bytes, not codepoints.
519 JH/21 Change as many as possible of the global flags into one-bit bitfields; these
520 should pack well giving a smaller memory footprint so better caching and
521 therefore performance. Group the declarations where this can't be done so
522 that the byte-sized flag variables are not interspersed among pointer
523 variables, giving a better chance of good packing by the compiler.
525 JH/22 Bug 1896: Fix the envelope from for DMARC forensic reports to be possibly
526 non-null, to avoid issues with sites running BATV. Previously reports were
527 sent with an empty envelope sender so looked like bounces.
529 JH/23 Bug 2318: Fix the noerror command within filters. It wasn't working.
530 The ignore_error flag wasn't being returned from the filter subprocess so
531 was not set for later routers. Investigation and fix by Matthias Kurz.
533 JH/24 Bug 2310: Raise a msg:fail:internal event for each undelivered recipient,
534 and a msg:complete for the whole, when a message is manually removed using
535 -Mrm. Developement by Matthias Kurz, hacked on by JH.
537 JH/25 Avoid fixed-size buffers for pathnames in DB access. This required using
538 a "Gnu special" function, asprintf() in the DB utility binary builds; I
539 hope that is portable enough.
541 JH/26 Bug 2311: Fix DANE-TA verification under GnuTLS. Previously it was also
542 requiring a known-CA anchor certificate; make it now rely entirely on the
543 TLSA as an anchor. Checking the name on the leaf cert against the name
544 on the A-record for the host is still done for TA (but not for EE mode).
546 JH/27 Fix logging of proxy address. Previously, a pointless "PRX=[]:0" would be
547 included in delivery lines for non-proxied connections, when compiled with
548 SUPPORT_SOCKS and running with proxy logging enabled.
550 JH/28 Bug 2314: Fire msg:fail:delivery event even when error is being ignored.
551 Developement by Matthias Kurz, tweaked by JH. While in that bit of code,
552 move the existing event to fire before the normal logging of message
553 failure so that custom logging is bracketed by normal logging.
555 JH/29 Bug 2322: A "fail" command in a non-system filter (file) now fires the
556 msg:fail:internal event. Developement by Matthias Kurz.
558 JH/30 Bug 2329: Increase buffer size used for dns lookup from 2k, which was
559 far too small for todays use of crypto signatures stored there. Go all
560 the way to the max DNS message size of 64kB, even though this might be
561 overmuch for IOT constrained device use.
563 JH/31 Fix a bad use of a copy function, which could be used to pointlessly
564 copy a string over itself. The library routine is documented as not
565 supporting overlapping copies, and on MacOS it actually raised a SIGABRT.
567 JH/32 For main options check_spool_space and check_inode_space, where the
568 platform supports 64b integers, support more than the previous 2^31 kB
569 (i.e. more than 2 TB). Accept E, P and T multipliers in addition to
570 the previous G, M, k.
572 JH/33 Bug 2338: Fix the cyrus-sasl authenticator to fill in the
573 $authenticated_fail_id variable on authentication failure. Previously
576 JH/34 Increase RSA keysize of autogen selfsign cert from 1024 to 2048. RHEL 8.0
577 OpenSSL didn't want to use such a weak key. Do for GnuTLS also, and for
578 more-modern GnuTLS move from GNUTLS_SEC_PARAM_LOW to
579 GNUTLS_SEC_PARAM_MEDIUM.
581 JH/35 OpenSSL: fail the handshake when SNI processing hits a problem, server
582 side. Previously we would continue as if no SNI had been received.
584 JH/36 Harden the handling of string-lists. When a list consisted of a sole
585 "<" character, which should be a list-separator specification, we walked
586 off past the nul-terimation.
588 JH/37 Bug 2341: Send "message delayed" warning MDNs (restricted to external
589 causes) even when the retry time is not yet met. Previously they were
590 not, meaning that when (say) an account was over-quota and temp-rejecting,
591 and multiple senders' messages were queued, only one sender would get
592 notified on each configured delay_warning cycle.
594 JH/38 Bug 2351: Log failures to extract envelope addresses from message headers.
596 JH/39 OpenSSL: clear the error stack after an SSL_accept(). With anon-auth
597 cipher-suites, an error can be left on the stack even for a succeeding
598 accept; this results in impossible error messages when a later operation
601 AM/01 Bug 2359: GnuTLS: repeat lowlevel read and write operations while they
602 return error codes indicating retry. Under TLS1.3 this becomes required.
604 JH/40 Fix the feature-cache refresh for EXPERIMENTAL_PIPE_CONNECT. Previously
605 it only wrote the new authenticators, resulting in a lack of tracking of
606 peer changes of ESMTP extensions until the next cache flush.
608 JH/41 Fix the loop reading a message header line to check for integer overflow,
609 and more-often against header_maxsize. Previously a crafted message could
610 induce a crash of the recive process; now the message is cleanly rejected.
612 JH/42 Bug 2366: Fix the behaviour of the dkim_verify_signers option. It had
613 been totally disabled for all of 4.91. Discovery and fix by "Mad Alex".
619 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
620 When redis_servers is set to a list of > 1 element, and the Redis servers
621 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
622 case of MOVED into a DEFER case instead, thus moving the query onto the
623 next server in the list. For a cluster of N elements, all N servers must
624 be defined in redis_servers.
626 GF/02 Catch and remove uninitialized value warning in exiqsumm
627 Check for existence of @ARGV before looking at $ARGV[0]
629 JH/01 Replace the store_release() internal interface with store_newblock(),
630 which internalises the check required to safely use the old one, plus
631 the allocate and data copy operations duplicated in both (!) of the
632 extant use locations.
634 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
635 modifier. This matches the restriction on the commandline.
637 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
638 Previously only the last row was returned.
640 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
641 we assumed that tags in the header were well-formed, and parsed the
642 element content after inspecting only the first char of the tag.
643 Assumptions at that stage could crash the receive process on malformed
646 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
647 While running the DKIM ACL we operate on the Permanent memory pool so that
648 variables created with "set" persist to the DATA ACL. Also (at any time)
649 DNS lookups that fail create cache records using the Permanent pool. But
650 expansions release any allocations made on the current pool - so a dnsdb
651 lookup expansion done in the DKIM ACL releases the memory used for the
652 DNS negative-cache, and bad things result. Solution is to switch to the
653 Main pool for expansions.
654 While we're in that code, add checks on the DNS cache during store_reset,
655 active in the testsuite.
656 Problem spotted, and debugging aided, by Wolfgang Breyha.
658 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
659 When none of the hosts presented to a transport match an already-open
660 connection, close it and proceed with the list. Previously we would
661 queue the message. Spotted by Lena with Yahoo, probably involving
664 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
665 Previously a spurious "250 OK id=" response was appended to the proper
668 JH/08 The "support for" informational output now, which built with Content
669 Scanning support, has a line for the malware scanner interfaces compiled
670 in. Interface can be individually included or not at build time.
672 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
673 by the template makefile "src/EDITME". The "STREAM" support for an older
674 ClamAV interface method is removed.
676 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
677 rows affected is given instead).
679 JH/11 The runtime Berkeley DB library version is now additionally output by
680 "exim -d -bV". Previously only the compile-time version was shown.
682 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
683 SMTP connection. Previously, when one had more receipients than the
684 first, an abortive onward connection was made. Move to full support for
685 multiple onward connections in sequence, handling cutthrough connection
686 for all multi-message initiating connections.
688 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
689 routers. Previously, a multi-recipient message would fail to match the
690 onward-connection opened for the first recipient, and cause its closure.
692 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
693 a timeout on read on a GnuTLS initiating connection, resulting in the
694 initiating connection being dropped. This mattered most when the callout
695 was marked defer_ok. Fix to keep the two timeout-detection methods
698 JH/15 Relax results from ACL control request to enable cutthrough, in
699 unsupported situations, from error to silently (except under debug)
700 ignoring. This covers use with PRDR, frozen messages, queue-only and
703 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
705 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
706 metadata, resulting in a crash in free().
708 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
709 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
710 Broken also in d185889f4, with init system revamp.
712 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
713 Previously we abruptly closed the connection after reading a malware-
714 found indication; now we go on to read the "scan ok" response line,
717 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
718 ACL. Previously, a crash would result.
720 JH/19 Speed up macro lookups during configuration file read, by skipping non-
721 macro text after a replacement (previously it was only once per line) and
722 by skipping builtin macros when searching for an uppercase lead character.
724 JH/20 DANE support moved from Experimental to mainline. The Makefile control
725 for the build is renamed.
727 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
728 was allocated for every new TLS startup, meaning one per message. Fix
729 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
731 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
732 reported the original. Fix to report (as far as possible) the ACL
733 result replacing the original.
735 JH/23 Fix memory leak during multi-message connections using STARTTLS under
736 OpenSSL. Certificate information is loaded for every new TLS startup,
737 and the resources needed to be freed.
739 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
741 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
744 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
745 DATA response info to the (existing) per-recipient response info for
746 the "C=" log element. It can have useful tracking info from the
747 destination system. Patch from Simon Arlott.
749 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
750 length value. Previously this would segfault.
752 HS/02 Support Avast multiline protoocol, this allows passing flags to
753 newer versions of the scanner.
755 JH/28 Ensure that variables possibly set during message acceptance are marked
756 dead before release of memory in the daemon loop. This stops complaints
757 about them when the debug_store option is enabled. Discovered specifically
758 for sender_rate_period, but applies to a whole set of variables.
759 Do the same for the queue-runner and queue-list loops, for variables set
760 from spool message files. Do the same for the SMTP per-message loop, for
761 certain variables indirectly set in ACL operations.
763 JH/29 Bug 2250: Fix a longstanding bug in heavily-pipelined SMTP input (such
764 as a multi-recipient message from a mailinglist manager). The coding had
765 an arbitrary cutoff number of characters while checking for more input;
766 enforced by writing a NUL into the buffer. This corrupted long / fast
767 input. The problem was exposed more widely when more pipelineing of SMTP
768 responses was introduced, and one Exim system was feeding another.
769 The symptom is log complaints of SMTP syntax error (NUL chars) on the
770 receiving system, and refused recipients seen by the sending system
771 (propating to people being dropped from mailing lists).
772 Discovered and pinpointed by David Carter.
774 JH/30 The (EXPERIMENTAL_DMARC) variable $dmarc_ar_header is withdrawn, being
775 replaced by the ${authresults } expansion.
777 JH/31 Bug 2257: Fix pipe transport to not use a socket-only syscall.
779 HS/03 Set a handler for SIGTERM and call exit(3) if running as PID 1. This
780 allows proper process termination in container environments.
782 JH/32 Bug 2258: Fix spool_wireformat in combination with LMTP transport.
783 Previously the "final dot" had a newline after it; ensure it is CR,LF.
785 JH/33 SPF: remove support for the "spf" ACL condition outcome values "err_temp"
786 and "err_perm", deprecated since 4.83 when the RFC-defined words
787 "temperror" and "permerror" were introduced.
789 JH/34 Re-introduce enforcement of no cutthrough delivery on transports having
790 transport-filters or DKIM-signing. The restriction was lost in the
791 consolidation of verify-callout and delivery SMTP handling.
792 Extend the restriction to also cover ARC-signing.
794 JH/35 Cutthrough: for a final-dot response timeout (and nonunderstood responses)
795 in defer=pass mode supply a 450 to the initiator. Previously the message
798 PP/02 DANE: add dane_require_tls_ciphers SMTP Transport option; if unset,
799 tls_require_ciphers is used as before.
801 HS/03 Malware Avast: Better match the Avast multiline protocol. Add
802 "pass_unscanned". Only tmpfails from the scanner are written to
803 the paniclog, as they may require admin intervention (permission
804 denied, license issues). Other scanner errors (like decompression
805 bombs) do not cause a paniclog entry.
807 JH/36 Fix reinitialisation of DKIM logging variable between messages.
808 Previously it was possible to log spurious information in receive log
811 JH/37 Bug 2255: Revert the disable of the OpenSSL session caching. This
812 triggered odd behaviour from Outlook Express clients.
814 PP/03 Add util/renew-opendmarc-tlds.sh script for safe renewal of public
817 JH/38 DKIM: accept Ed25519 pubkeys in SubjectPublicKeyInfo-wrapped form,
818 since the IETF WG has not yet settled on that versus the original
819 "bare" representation.
821 JH/39 Fix syslog logging for syslog_timestamp=no and log_selector +millisec.
822 Previously the millisecond value corrupted the output.
823 Fix also for syslog_pid=no and log_selector +pid, for which the pid
824 corrupted the output.
830 JH/01 Rework error string handling in TLS interface so that the caller in
831 more cases is responsible for logging. This permits library-sourced
832 string to be attached to addresses during delivery, and collapses
833 pairs of long lines into single ones.
835 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
836 during configuration. Wildcards are allowed and expanded.
838 JH/02 Rework error string handling in DKIM to pass more info back to callers.
839 This permits better logging.
841 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
842 do not close it down and have the child transport start it up again on
843 the passed-on TCP connection. Instead, proxy the child (and any
844 subsequent ones) for TLS via a unix-domain socket channel. Logging is
845 affected: the continued delivery log lines do not have any DNSSEC, TLS
846 Certificate or OCSP information. TLS cipher information is still logged.
848 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
849 identical IP addresses on different listening ports. Will also affect
852 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
853 add noisy ifdef guards to special-case this sillyness.
854 Patch from Bernd Kuhls.
856 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
857 than 255 are no longer allowed.
859 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
860 Disable the session-cache too, which might reduce our load. Since we
861 currrectly use a new context for every connection, both as server and
862 client, there is no benefit for these.
863 GnuTLS appears to not support tickets server-side by default (we don't
864 call gnutls_session_ticket_enable_server()) but client side is enabled
865 by default on recent versions (3.1.3 +) unless the PFS priority string
868 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
869 <https://reproducible-builds.org/specs/source-date-epoch/>.
871 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
872 the check for any unsuccessful recipients did not notice the limit, and
873 erroneously found still-pending ones.
875 JH/08 Pipeline CHUNKING command and data together, on kernels that support
876 MSG_MORE. Only in-clear (not on TLS connections).
878 JH/09 Avoid using a temporary file during transport using dkim. Unless a
879 transport-filter is involved we can buffer the headers in memory for
880 creating the signature, and read the spool data file once for the
881 signature and again for transmission.
883 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
884 4.77 as the kernel support then wasn't solid, having issues in 64bit
885 mode. Now, it's been long enough. Add support for FreeBSD also.
887 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
888 case where the routing stage had gathered several addresses to send to
889 a host before calling the transport for the first, we previously failed
890 to close down TLS in the old transport process before passing the TCP
891 connection to the new process. The new one sent a STARTTLS command
892 which naturally failed, giving a failed delivery and bloating the retry
893 database. Investigation and fix prototype from Wolfgang Breyha.
895 JH/12 Fix check on SMTP command input synchronisation. Previously there were
896 false-negatives in the check that the sender had not preempted a response
897 or prompt from Exim (running as a server), due to that code's lack of
898 awareness of the SMTP input buffering.
900 PP/04 Add commandline_checks_require_admin option.
901 Exim drops privileges sanely, various checks such as -be aren't a
902 security problem, as long as you trust local users with access to their
903 own account. When invoked by services which pass untrusted data to
904 Exim, this might be an issue. Set this option in main configuration
905 AND make fixes to the calling application, such as using `--` to stop
908 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
909 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
910 responses to those, into a single TLS record each way (this usually means
911 a single packet). As a side issue, smtp_enforce_sync now works on TLS
914 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
915 affects you only if you're dancing at the edge of the param size limits.
916 If you are, and this message makes sense to you, then: raise the
917 configured limit or use OpenSSL 1.1. Nothing we can do for older
920 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
921 cmdline element to get the documented default one. Previously it was
924 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
925 are unsuable for non-message-specific reasons - eg. connection timeout,
926 banner-time rejection.
928 JH/16 Fix logging of delivery remote port, when specified by router, under
931 PP/06 Repair manualroute's ability to take options in any order, even if one
932 is the name of a transport.
935 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
937 JH/17 Change the list-building routines interface to use the expanding-string
938 triplet model, for better allocation and copying behaviour.
940 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
941 Previously it was constructed the first time a possibly-matching string
942 was met in the configuration file input during startup; now it is done
945 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
946 compatible one, to avoid the (poorly documented) possibility of a config
947 file in the working directory redirecting the DB files, possibly correpting
948 some existing file. CVE-2017-10140 assigned for BDB.
950 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
951 cache-hot. Previously, although the result was properly cached, the
952 initial verify call returned a defer.
954 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
955 the main verify for receipient in uncached-mode.
957 JH/22 Retire historical build files to an "unsupported" subdir. These are
958 defined as "ones for which we have no current evidence of testing".
960 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
961 if present. Previously it was ignored.
963 JH/24 Start using specified-initialisers in C structure init coding. This is
964 a C99 feature (it's 2017, so now considered safe).
966 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
967 if was a fixed-sized field and bitmask ops via macros; it is now more
970 PP/07 GitHub PR 56: Apply MariaDB build fix.
971 Patch provided by Jaroslav Škarvada.
973 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
974 during Coverity cleanups [4.87 JH/47]
975 Diagnosis and fix provided by Michael Fischer v. Mollard.
977 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
978 the right size to place the terminating semicolon on its own folded
979 line, the header hash was calculated to an incorrect value thanks to
980 the (relaxed) space the fold became.
982 HS/02 Fix Bug 2130: large writes from the transport subprocess were chunked
983 and confused the parent.
985 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
986 which could crash as a result. This could lead to undeliverable messages.
988 JH/28 Logging: "next input sent too soon" now shows where input was truncated
991 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
992 matters on fast-turnover and PID-randomising systems, which were getting
993 out-of-order delivery.
995 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
996 a possibly-overlapping copy. The symptom was that "Remote host closed
997 connection in response to HELO" was logged instead of the actual 4xx
1000 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
1001 Previously only that bufferd was discarded, resulting in SYMTP command
1004 JH/32 DKIM: when a message has multiple signatures matching an identity given
1005 in dkim_verify_signers, run the dkim acl once for each. Previously only
1006 one run was done. Bug 2189.
1008 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
1009 "panic the current process" to "deliberately defer". The panic log is
1010 still written with the problem list name; the mail and reject logs now
1011 get a temp-reject line for the message that was being handled, saying
1012 something like "domains check lookup or other defer". The SMTP 451
1013 message is still "Temporary local problem".
1015 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
1016 A crafted sequence of BDAT commands could result in in-use memory beeing
1017 freed. CVE-2017-16943.
1019 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
1020 from SMTP input. Previously it was always done; now only done for DATA
1021 and not BDAT commands. CVE-2017-16944.
1023 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
1024 to the message (such as an overlong header line). Previously this was
1025 not done and we did not exit BDAT mode. Followon from the previous item
1026 though a different problem.
1032 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
1033 than -2003 did; needs libidn2 in addition to libidn.
1035 JH/02 The path option on a pipe transport is now expanded before use.
1037 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
1038 Patch provided by "Björn", documentation fix added too.
1040 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
1041 missing a wire-to-host endian conversion.
1043 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
1044 close after a BDAT command line could be taken as a following command,
1045 giving a synch failure. Fix by only checking for synch immediately
1046 before acknowledging the chunk.
1048 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
1049 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
1050 macro. Patches provided by Josh Soref.
1052 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
1053 Previously we did not; the RFC seems ambiguous and VRFY is not listed
1054 by IANA as a service extension. However, John Klensin suggests that we
1057 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
1058 the dkim code may be unix-mode line endings rather than smtp wire-format
1059 CRLF, so prepend a CR to any bare LF.
1061 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
1062 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
1064 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
1065 the first were themselves being wrongly included in the feed into dkim
1066 processing; with most chunk sizes in use this resulted in an incorrect
1067 body hash calculated value.
1069 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
1070 DKIM signature block, for verification. Although advised against by
1071 standards it is specifically not ruled illegal.
1073 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
1075 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
1076 missing a body hash (the bh= tag).
1078 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
1079 It seems that HAProxy sends the Proxy Protocol information in clear and
1080 only then does a TLS startup, so do the same.
1082 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
1083 TCP connections (such as for Spamd) unless the daemon successfully set
1084 Fast Open mode on its listening sockets. This fixes breakage seen on
1085 too-old kernels or those not configured for Fast Open, at the cost of
1086 requiring both directions being enabled for TFO, and TFO never being used
1087 by non-daemon-related Exim processes.
1089 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
1090 endings, at least on the first header line. Try to canonify any that get
1091 past that check, despite the cost.
1093 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
1094 now limited to an arbitrary five deep, while parsing addresses with the
1095 strip_excess_angle_brackets option enabled.
1097 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
1098 instead leave the unprompted TLS handshake in socket buffer for the
1099 TLS library to consume.
1101 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
1103 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
1105 JH/16 Drop variables when they go out of scope. Memory management drops a whole
1106 region in one operation, for speed, and this leaves assigned pointers
1107 dangling. Add checks run only under the testsuite which checks all
1108 variables at a store-reset and panics on a dangling pointer; add code
1109 explicitly nulling out all the variables discovered. Fixes one known
1110 bug: a transport crash, where a dangling pointer for $sending_ip_address
1111 originally assigned in a verify callout, is re-used.
1113 PP/06 Drop '.' from @INC in various Perl scripts.
1115 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
1117 PP/08 Reduce a number of compilation warnings under clang; building with
1118 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
1119 should be warning-free.
1121 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
1123 HS/01 Fix portability problems introduced by PP/08 for platforms where
1124 realloc(NULL) is not equivalent to malloc() [SunOS et al].
1126 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
1127 chunk. This allows us to accept broken chunked messages. We need a more
1128 general solution here.
1130 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
1131 already-broken messages in the queue.
1133 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
1135 JH/19 Fix reference counting bug in routing-generated-address tracking.
1141 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
1142 supports it and a size is available (ie. the sending peer gave us one).
1144 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
1145 years of being deprecated). The replacements are the ACLs
1146 acl_smtp_mime and acl_not_smtp_mime.
1148 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
1149 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
1150 or even an in-clear connection were permitted. Now, if the host lookup
1151 was dnssec and dane was requested then the host is only used if the
1152 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
1153 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
1154 if one fails this test.
1155 This means that a poorly-configured remote DNS will make it incommunicado;
1156 but it protects against a DNS-interception attack on it.
1158 JH/04 Bug 1810: make continued-use of an open smtp transport connection
1159 non-noisy when a race steals the message being considered.
1161 JH/05 If main configuration option tls_certificate is unset, generate a
1162 self-signed certificate for inbound TLS connections.
1164 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
1165 in rewrites and routers.
1167 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
1168 and logged a warning sing 4.83; now they are a configuration file error.
1170 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
1171 (lacking @domain). Apply the same qualification processing as RCPT.
1173 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
1175 JH/10 Support ${sha256:} applied to a string (as well as the previous
1178 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
1179 a cutthrough deliver is pending, as we always want to make a connection.
1180 This also avoids re-routing the message when later placing the cutthrough
1181 connection after a verify cache hit.
1182 Do not update it with the verify result either.
1184 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
1185 when routing results in more than one destination address.
1187 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
1188 signing (which inhibits the cutthrough capability). Previously only
1189 the presence of an option was tested; now an expansion evaluating as
1190 empty is permissible (obviously it should depend only on data available
1191 when the cutthrough connection is made).
1193 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
1194 the relevant preceding SMTP command did not note the pipelining mode.
1196 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
1197 Previously they were not counted.
1199 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
1200 as one having no matching records. Previously we deferred the message
1201 that needed the lookup.
1203 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
1204 distinguished as "(=".
1206 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
1207 for missing MX records. Previously it only worked for missing A records.
1209 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
1211 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
1212 after the data-go-ahead and data-ack. Patch from Jason Betts.
1214 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
1215 even for a "none" policy. Patch from Tony Meyer.
1217 JH/22 Fix continued use of a connection for further deliveries. If a port was
1218 specified by a router, it must also match for the delivery to be
1221 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
1222 When one of the recipients of a message was unsuitable for the connection
1223 (has no matching addresses), we lost track of needing to mark it
1224 deferred. As a result mail would be lost.
1226 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
1228 JH/25 Decoding ACL controls is now done using a binary search; the source code
1229 takes up less space and should be simpler to maintain. Merge the ACL
1230 condition decode tables also, with similar effect.
1232 JH/26 Fix problem with one_time used on a redirect router which returned the
1233 parent address unchanged. A retry would see the parent address marked as
1234 delivered, so not attempt the (identical) child. As a result mail would
1237 JH/27 Fix a possible security hole, wherein a process operating with the Exim
1238 UID can gain a root shell. Credit to http://www.halfdog.net/ for
1239 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
1242 JH/28 Enable {spool,log} filesystem space and inode checks as default.
1243 Main config options check_{log,spool}_{inodes,space} are now
1244 100 inodes, 10MB unless set otherwise in the configuration.
1246 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
1247 Previously it only applied to the main-section connection policy
1250 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
1252 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
1253 by me. Added RFC7919 DH primes as an alternative.
1255 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
1256 are not in the system include path.
1258 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
1259 GnuTLS, when a session startup failed (eg because the client disconnected)
1260 Exim did stdio operations after fclose. This was exposed by a recent
1261 change which nulled out the file handle after the fclose.
1263 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
1264 signed directly by the cert-signing cert, rather than an intermediate
1265 OCSP-signing cert. This is the model used by LetsEncrypt.
1267 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
1269 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
1270 an incoming connection.
1272 HS/02 Bug 1802: Do not half-close the connection after sending a request
1275 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
1276 fallback to "prime256v1".
1278 JH/34 SECURITY: Use proper copy of DATA command in error message.
1279 Could leak key material. Remotely exploitable. CVE-2016-9963.
1285 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
1286 and 3.4.4 - once the server is enabled to respond to an OCSP request
1287 it does even when not requested, resulting in a stapling non-aware
1288 client dropping the TLS connection.
1290 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
1291 support variable-length bit vectors. No functional change.
1293 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
1294 The I= interface field on outgoing lines is now after the H= remote
1295 host field, same as incoming lines. There is a separate
1296 outgoing_interface log selector which allows you to disable the
1299 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
1300 If not running log_selector +smtp_connection the mainlog would be held
1301 open indefinitely after a "too many connections" event, including to a
1302 deleted file after a log rotate. Leave the per net connection logging
1303 leaving it open for efficiency as that will be quickly detected by the
1304 check on the next write.
1306 HS/01 Bug 1671: Fix post transport crash.
1307 Processing the wait-<transport> messages could crash the delivery
1308 process if the message IDs didn't exist for some reason. When
1309 using 'split_spool_directory=yes' the construction of the spool
1310 file name failed already, exposing the same netto behaviour.
1312 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
1313 mime_regex ACL conditions.
1315 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
1316 to DSN fail messages (bounces): remote IP, remote greeting, remote response
1317 to HELO, local diagnostic string.
1319 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
1320 log line to debug. Even when configured with a tls authenticator many
1321 client connections are expected to not authenticate in this way, so
1322 an authenticate fail is not an error.
1324 HS/02 Add the Exim version string to the process info. This way exiwhat
1325 gives some more detail about the running daemon.
1327 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
1328 matter for fast-change records such as DNSBLs.
1330 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
1331 retry record, even if constant. There may be multiple transports with
1332 different interface settings and the retry behaviour needs to be kept
1335 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
1337 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
1339 JH/10 Bug 840: fix log_defer_output option of pipe transport
1341 JH/11 Bug 830: use same host for all RCPTS of a message, even under
1342 hosts_randomize. This matters a lot when combined with mua_wrapper.
1344 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
1345 ${quote_pgsql:<string>} operator.
1347 JH/13 Bug 1708: avoid misaligned access in cached lookup.
1349 JH/14 Change header file name for freeradius-client. Relevant if compiling
1350 with Radius support; from the Gentoo tree and checked under Fedora.
1352 JH/15 Bug 1712: Introduce $prdr_requested flag variable
1354 JH/16 Bug 1714: Permit an empty string as expansion result for transport
1355 option transport_filter, meaning no filtering.
1357 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
1359 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
1360 defaults to "*" (all hosts). The variable is now available when not built
1361 with TLS, default unset, mainly to enable keeping the testsuite sane.
1362 If a server certificate is not supplied (via tls_certificate) an error is
1363 logged, and clients will find TLS connections fail on startup. Presumably
1364 they will retry in-clear.
1365 Packagers of Exim are strongly encouraged to create a server certificate
1366 at installation time.
1368 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
1369 with the $config_file variable.
1371 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
1372 in transport context, after the attempt, and per-recipient. The latter type
1373 is per host attempted. The event data is the error message, and the errno
1374 information encodes the lookup type (A vs. MX) used for the (first) host,
1375 and the trailing two digits of the smtp 4xx response.
1377 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
1378 to write to mainlog (or rejectlog, paniclog) in the window between file
1379 creation and permissions/ownership being changed. Particularly affects
1380 installations where exicyclog is run as root, rather than exim user;
1381 result is that the running daemon panics and dies.
1383 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
1385 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
1386 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
1387 "pri" and "weight". Note that the previous implicit priority given by the
1388 list order is no longer honoured.
1390 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
1391 for DKIM processing.
1393 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
1394 by defining SUPPORT_SOCKS.
1396 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
1397 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
1398 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
1399 variables are renamed to proxy_{local,external}_{address,port}.
1401 JH/27 Move Internationalisation support from Experimental to mainline, enabled
1402 for a build by defining SUPPORT_I18N
1404 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
1405 of the query string, and make ${quote_redis:} do that quoting.
1407 JH/29 Move Events support from Experimental to mainline, enabled by default
1408 and removable for a build by defining DISABLE_EVENT.
1410 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
1412 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
1413 cached by the daemon.
1415 JH/32 Move Redis support from Experimental to mainline, enabled for a build
1416 by defining LOOKUP_REDIS. The libhiredis library is required.
1418 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
1419 keys are given for lookup.
1421 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
1422 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
1423 only supported when built with TLS support. The PolarSSL SHA routines
1424 are still used when the TLS library is too old for convenient support.
1426 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
1427 openssl_options), for security. OpenSSL forces this from version 1.1.0
1428 server-side so match that on older versions.
1430 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
1431 allocation for $value could be released as the expansion processing
1432 concluded, but leaving the global pointer active for it.
1434 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
1435 and to use the domains and local_parts ACL conditions.
1437 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
1438 incorrectly not doubled on cutthrough transmission, hence seen as a
1439 body-termination at the receiving system - resulting in truncated mails.
1440 Commonly the sender saw a TCP-level error, and retransmitted the message
1441 via the normal store-and-forward channel. This could result in duplicates
1442 received - but deduplicating mailstores were liable to retain only the
1443 initial truncated version.
1445 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
1447 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
1449 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
1450 we're in there, support oversigning also; bug 1309.
1452 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
1454 HS/04 Add support for keep_environment and add_environment options.
1456 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
1457 either intentional arithmetic overflow during PRNG, or testing config-
1460 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
1461 delivery resulted in actual delivery. Cancel cutthrough before DATA
1464 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
1465 rejects a recipient: pass the reject to the originator.
1467 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
1468 Many were false-positives and ignorable, but it's worth fixing the
1471 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
1472 for the new environment-manipulation done at startup. Move the routines
1473 from being local to tls.c to being global via the os.c file.
1475 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
1476 an extract embedded as result-arg for a map, the first arg for extract
1477 is unavailable so we cannot tell if this is a numbered or keyed
1478 extraction. Accept either.
1484 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
1487 JH/02 The smtp transport option "multi_domain" is now expanded.
1489 JH/03 The smtp transport now requests PRDR by default, if the server offers
1492 JH/04 Certificate name checking on server certificates, when exim is a client,
1493 is now done by default. The transport option tls_verify_cert_hostnames
1494 can be used to disable this per-host. The build option
1495 EXPERIMENTAL_CERTNAMES is withdrawn.
1497 JH/05 The value of the tls_verify_certificates smtp transport and main options
1498 default to the word "system" to access the system default CA bundle.
1499 For GnuTLS, only version 3.0.20 or later.
1501 JH/06 Verification of the server certificate for a TLS connection is now tried
1502 (but not required) by default. The verification status is now logged by
1503 default, for both outbound TLS and client-certificate supplying inbound
1506 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
1509 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
1510 Status Notification (bounce) messages are now MIME format per RFC 3464.
1511 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
1512 under the control of the dsn_advertise_hosts option, and routers may
1513 have a dsn_lasthop option.
1515 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
1516 default, modifiable by a malware= option. The list separator for
1517 the options can now be changed in the usual way. Bug 68.
1519 JH/10 The smtp_receive_timeout main option is now expanded before use.
1521 JH/11 The incoming_interface log option now also enables logging of the
1522 local interface on delivery outgoing connections.
1524 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
1525 if the interface and destination host and port all match.
1527 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
1530 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
1531 Patch from Andrew Lewis.
1533 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
1534 now supports optional time-restrictions, weighting, and priority
1535 modifiers per server. Patch originally by <rommer@active.by>.
1537 JH/16 The spamd_address main option now supports a mixed list of local
1538 and remote servers. Remote servers can be IPv6 addresses, and
1539 specify a port-range.
1541 JH/17 Bug 68: The spamd_address main option now supports an optional
1542 timeout value per server.
1544 JH/18 Bug 1581: Router and transport options headers_add/remove can
1545 now have the list separator specified.
1547 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
1550 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
1553 JH/21 Support for the A6 type of dns record is withdrawn.
1555 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
1556 rather than the verbs used.
1558 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
1559 from 255 to 1024 chars.
1561 JH/24 Verification callouts now attempt to use TLS by default.
1563 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
1564 are generic router options now. The defaults didn't change.
1566 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
1567 Original patch from Alexander Shikoff, worked over by JH.
1569 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
1570 files if ZCAT_COMMAND is not executable.
1572 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
1574 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
1576 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
1577 Normally benign, it bites when the pair was led to by a CNAME;
1578 modern usage is to not canonicalize the domain to a CNAME target
1579 (and we were inconsistent anyway for A-only vs AAAA+A).
1581 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
1583 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
1584 when evaluating $sender_host_dnssec.
1586 JH/31 Check the HELO verification lookup for DNSSEC, adding new
1587 $sender_helo_dnssec variable.
1589 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
1591 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
1593 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
1595 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
1596 documented as working, but never had. Support all but $spam_report.
1598 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
1599 added for tls authenticator.
1601 HS/03 Add perl_taintmode main config option
1607 TL/01 When running the test suite, the README says that variables such as
1608 no_msglog_check are global and can be placed anywhere in a specific
1609 test's script, however it was observed that placement needed to be near
1610 the beginning for it to behave that way. Changed the runtest perl
1611 script to read through the entire script once to detect and set these
1612 variables, reset to the beginning of the script, and then run through
1613 the script parsing/test process like normal.
1615 TL/02 The BSD's have an arc4random API. One of the functions to induce
1616 adding randomness was arc4random_stir(), but it has been removed in
1617 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
1618 function when detected.
1620 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
1621 cause callback expansion.
1623 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
1624 syntax errors in an expansion can be treated as a string instead of
1625 logging or causing an error, due to the internal use of bool_lax
1626 instead of bool when processing it.
1628 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
1629 server certificates when making smtp deliveries.
1631 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
1633 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
1635 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
1637 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
1638 Merged patch from Sebastian Wiedenroth.
1640 JH/05 Fix results-pipe from transport process. Several recipients, combined
1641 with certificate use, exposed issues where response data items split
1642 over buffer boundaries were not parsed properly. This eventually
1643 resulted in duplicates being sent. This issue only became common enough
1644 to notice due to the introduction of connection certificate information,
1645 the item size being so much larger. Found and fixed by Wolfgang Breyha.
1647 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
1648 size buffer was used, resulting in syntax errors when an expansion
1651 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1652 version 3.3.6 or later.
1654 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1655 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1656 both become "event_action", the variables become $event_name, $event_data
1657 and $event_defer_errno. There is a new variable $verify_mode, usable in
1658 routers, transports and related events. The tls:cert event is now also
1659 raised for inbound connections, if the main configuration event_action
1662 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1663 early OCSP support, but no stapling (appears to be less than 1.0.0).
1665 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1666 server certificate names available under the smtp transport option
1667 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1670 JH/10 Time-related extraction expansions from certificates now use the main
1671 option "timezone" setting for output formatting, and are consistent
1672 between OpenSSL and GnuTLS compilations. Bug 1541.
1674 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1675 encoded parameter in the incoming message. Bug 1558.
1677 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1678 include certificate info, eximon was claiming there were spoolfile
1681 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1683 JH/14 Log delivery-related information more consistently, using the sequence
1684 "H=<name> [<ip>]" wherever possible.
1686 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1687 are problematic for Debian distribution, omit them from the release
1690 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1692 JH/16 Fix string representation of time values on 64bit time_t architectures.
1695 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1696 output list separator was used.
1701 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1702 checkers that were complaining about end of non-void function with no
1705 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1706 This was a regression introduced in 4.83 by another bugfix.
1708 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1710 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1711 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1717 TF/01 Correctly close the server side of TLS when forking for delivery.
1719 When a message was received over SMTP with TLS, Exim failed to clear up
1720 the incoming connection properly after forking off the child process to
1721 deliver the message. In some situations the subsequent outgoing
1722 delivery connection happened to have the same fd number as the incoming
1723 connection previously had. Exim would try to use TLS and fail, logging
1724 a "Bad file descriptor" error.
1726 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1727 utilities have not been installed.
1729 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1730 temporary space as the ACL may create new global variables.
1732 TL/01 LDAP support uses per connection or global context settings, depending
1733 upon the detected version of the libraries at build time.
1735 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1736 to extract and use the src ip:port in logging and expansions as if it
1737 were a direct connection from the outside internet. PPv2 support was
1738 updated based on HAProxy spec change in May 2014.
1740 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1742 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1743 Properly escape header and check for NULL return.
1745 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1748 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1750 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1751 characters in header names, implemented as a verify condition.
1752 Contributed by Michael Fischer v. Mollard.
1754 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1755 results permerror and temperror. Previous values are deprecated but
1756 still accepted. In a future release, err_perm and err_temp will be
1757 completely removed, which will be a backward incompatibility if the
1758 ACL tests for either of these two old results. Patch contributed by
1759 user bes-internal on the mailing list.
1761 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1763 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1764 selectors, in both main and reject logs.
1766 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1769 JH/07 Add malware type "sock" for talking to simple daemon.
1771 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1773 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1774 routers/transports under cutthrough routing.
1776 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1777 numbers. Touch up "bool" conditional to keep the same definition.
1779 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1781 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1783 JH/12 Expand items in router/transport headers_add or headers_remove lists
1784 individually rather than the list as a whole. Bug 1452.
1786 Required for reasonable handling of multiple headers_ options when
1787 they may be empty; requires that headers_remove items with embedded
1788 colons must have them doubled (or the list-separator changed).
1790 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1791 view the policy declared in the DMARC record. Currently, $dmarc_status
1792 is a combined value of both the record presence and the result of the
1795 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1797 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1798 dnslookup router and the smtp transport (applying to the forward
1801 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1802 of ldap servers used for a specific lookup. Patch provided by Heiko
1805 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1806 New variable $lookup_dnssec_authenticated for observability.
1808 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1809 Patch submitted by Lars Timman.
1811 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1813 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1814 Requires trusted mode and valid format message id, aborts otherwise.
1815 Patch contributed by Heiko Schlichting.
1817 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1818 certextract with support for various fields. Bug 1358.
1820 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1821 is requested by default, modifiable by smtp transport option
1824 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1825 operate on certificate variables to give certificate fingerprints
1826 Also new ${sha256:cert_variable}.
1828 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1830 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1831 Christian Aistleitner.
1833 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1835 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1836 file. Patch from Wolfgang Breyha.
1838 JH/25 Expand the coverage of the delivery $host and $host_address to
1839 client authenticators run in verify callout. Bug 1476.
1841 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1842 align with daemon_smtp_ports. Bug 72.
1844 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1845 support and error reporting did not work properly.
1847 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1848 and is readable. Patch from Andrew Colin Kissa.
1850 TL/14 Enhance documentation of ${run expansion and how it parses the
1851 commandline after expansion, particularly in the case when an
1852 unquoted variable expansion results in an empty value.
1854 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1856 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1857 by renaming away from it. Addresses GH issue 15, reported by
1860 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1862 TL/15 SECURITY: prevent double expansion in math comparison functions
1863 (can expand unsanitized data). Not remotely exploitable.
1870 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1872 PP/02 Make -n do something, by making it not do something.
1873 When combined with -bP, the name of an option is not output.
1875 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1878 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1879 $sender_host_name and config options to manage this, and basic check
1882 PP/05 DSCP support for outbound connections and control modifier for inbound.
1884 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1885 (Only plugin which currently uses this is kerberos4, which nobody should
1886 be using, but we should make it available and other future plugins might
1887 conceivably use it, even though it would break NAT; stuff *should* be
1888 using channel bindings instead).
1890 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1891 name; added for Sendmail compatibility; requires admin caller.
1892 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1893 just ignore it); requires trusted caller.
1894 Also parse but ignore: -Ac -Am -X<logfile>
1897 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1899 TL/02 Add +smtp_confirmation as a default logging option.
1901 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1902 Patch by Magnus Holmgren from 2007-02-20.
1904 TL/04 Bugzilla 1281 - Spec typo.
1905 Bugzilla 1283 - Spec typo.
1906 Bugzilla 1290 - Spec grammar fixes.
1908 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1910 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1912 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1913 mailing list by by Dmitry Isaikin.
1915 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1917 JH/02 Support "G" suffix to numbers in ${if comparisons.
1919 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1921 NM/01 Bugzilla 1197 - Spec typo
1922 Bugzilla 1196 - Spec examples corrections
1924 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1926 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1927 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1930 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1931 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1933 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1934 "acl {{name}{arg}...}", and optional args on acl condition
1937 JH/05 Permit multiple router/transport headers_add/remove lines.
1939 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1941 JH/07 Avoid using a waiting database for a single-message-only transport.
1942 Performance patch from Paul Fisher. Bugzilla 1262.
1944 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1947 JH/09 Add $headers_added variable, with content from use of ACL modifier
1948 add_header (but not yet added to the message). Bugzilla 199.
1950 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1951 Pulled from Bugzilla 817 by Wolfgang Breyha.
1953 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1955 (nb: this is the same fix as in Exim 4.80.1)
1957 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1960 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1963 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1965 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1966 advertise SMTP AUTH mechanism to us, instead of a generic
1967 protocol violation error. Also, make Exim more robust to bad
1968 data from the Dovecot auth socket.
1970 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1972 When a queue runner is handling a message, Exim first routes the
1973 recipient addresses, during which it prunes them based on the retry
1974 hints database. After that it attempts to deliver the message to
1975 any remaining recipients. It then updates the hints database using
1978 So if a recipient address works intermittently, it can get repeatedly
1979 deferred at routing time. The retry hints record remains fresh so the
1980 address never reaches the final cutoff time.
1982 This is a fairly common occurrence when a user is bumping up against
1983 their storage quota. Exim had some logic in its local delivery code
1984 to deal with this. However it did not apply to per-recipient defers
1985 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1987 This change adds a proper retry rule check during routing so that the
1988 final cutoff time is checked against the message's age. We only do
1989 this check if there is an address retry record and there is not a
1990 domain retry record; this implies that previous attempts to handle
1991 the address had the retry_use_local_parts option turned on. We use
1992 this as an approximation for the destination being like a local
1993 delivery, as in LMTP.
1995 I suspect this new check makes the old local delivery cutoff check
1996 redundant, but I have not verified this so I left the code in place.
1998 TF/02 Correct gecos expansion when From: is a prefix of the username.
2000 Test 0254 submits a message to Exim with the header
2004 When I ran the test suite under the user fanf2, Exim expanded
2005 the header to contain my full name, whereas it should have added
2006 a Resent-Sender: header. It erroneously treats any prefix of the
2007 username as equal to the username.
2009 This change corrects that bug.
2011 GF/01 DCC debug and logging tidyup
2012 Error conditions log to paniclog rather than rejectlog.
2013 Debug lines prefixed by "DCC: " to remove any ambiguity.
2015 TF/03 Avoid unnecessary rebuilds of lookup-related code.
2017 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
2018 Bug spotted by Jeremy Harris; was flawed since initial commit.
2019 Would have resulted in OCSP responses post-SNI triggering an Exim
2020 NULL dereference and crash.
2022 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
2024 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
2025 Bug detection, analysis and fix by Samuel Thibault.
2026 Bugzilla 1331, Debian bug #698092.
2028 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
2030 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
2031 Server implementation by Todd Lyons, client by JH.
2032 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
2033 config variable "prdr_enable" controls whether the server
2034 advertises the facility. If the client requests PRDR a new
2035 acl_data_smtp_prdr ACL is called once for each recipient, after
2036 the body content is received and before the acl_smtp_data ACL.
2037 The client is controlled by both of: a hosts_try_prdr option
2038 on the smtp transport, and the server advertisement.
2039 Default client logging of deliveries and rejections involving
2040 PRDR are flagged with the string "PRDR".
2042 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
2043 fclose(). Diagnosis by Todd Lyons.
2045 PP/17 Update configure.default to handle IPv6 localhost better.
2046 Patch by Alain Williams (plus minor tweaks).
2049 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
2050 This is now consistent with GnuTLS, and is now documented: the
2051 previous undocumented portable approach to treating the option as
2052 unset was to force an expansion failure. That still works, and
2053 an empty string is now equivalent.
2055 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
2056 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
2057 not performing validation itself.
2059 PP/20 Added force_command boolean option to pipe transport.
2060 Patch from Nick Koston, of cPanel Inc.
2062 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
2065 TF/04 Added udpsend ACL modifier and hexquote expansion operator
2067 PP/21 Fix eximon continuous updating with timestamped log-files.
2068 Broken in a format-string cleanup in 4.80, missed when I repaired the
2069 other false fix of the same issue.
2070 Report and fix from Heiko Schlichting.
2073 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
2074 Report from Prashanth Katuri.
2076 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
2077 It's SecureTransport, so affects any MacOS clients which use the
2078 system-integrated TLS libraries, including email clients.
2080 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
2081 using a MIME ACL for non-SMTP local injection.
2082 Report and assistance in diagnosis by Warren Baker.
2084 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
2086 JH/16 Fix comparisons for 64b. Bugzilla 1385.
2088 TL/09 Add expansion variable $authenticated_fail_id to keep track of
2089 last id that failed so it may be referenced in subsequent ACL's.
2091 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
2094 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
2095 ldap library initialization, allowing self-signed CA's to be
2096 used. Also properly sets require_cert option later in code by
2097 using NULL (global ldap config) instead of ldap handle (per
2098 session). Bug diagnosis and testing by alxgomz.
2100 TL/12 Enhanced documentation in the ratelimit.pl script provided in
2101 the src/util/ subdirectory.
2103 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
2104 renamed to Transport Post Delivery Action by Jeremy Harris, as
2107 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
2108 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
2109 redis_servers = needs to be configured which will be used by the redis
2110 lookup. Patch from Warren Baker, of The Packet Hub.
2112 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
2114 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
2115 hostname or reverse DNS when processing a host list. Used suggestions
2116 from multiple comments on this bug.
2118 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
2120 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
2121 Missed a few lines, added it to make the runtest require no keyboard
2124 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
2125 contains upper case chars. Make router use caseful_local_part.
2127 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
2128 support when GnuTLS has been built with p11-kit.
2134 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
2136 This, or similar/improved, will also be change PP/11 of 4.82.
2142 PP/01 Handle short writes when writing local log-files.
2143 In practice, only affects FreeBSD (8 onwards).
2144 Bugzilla 1053, with thanks to Dmitry Isaikin.
2146 NM/01 Bugzilla 949 - Documentation tweak
2148 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
2151 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
2153 PP/02 Implemented gsasl authenticator.
2155 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
2157 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
2158 `pkg-config foo` for cflags/libs.
2160 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
2161 with rest of GSASL and with heimdal_gssapi.
2163 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
2164 `pkg-config foo` for cflags/libs for the TLS implementation.
2166 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
2167 properties get this fed in as external SSF. A number of robustness
2168 and debugging improvements to the cyrus_sasl authenticator.
2170 PP/08 cyrus_sasl server now expands the server_realm option.
2172 PP/09 Bugzilla 1214 - Log authentication information in reject log.
2173 Patch by Jeremy Harris.
2175 PP/10 Added dbmjz lookup type.
2177 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
2179 PP/12 MAIL args handles TAB as well as SP, for better interop with
2180 non-compliant senders.
2181 Analysis and variant patch by Todd Lyons.
2183 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
2184 Bug report from Lars Müller <lars@samba.org> (via SUSE),
2185 Patch from Dirk Mueller <dmueller@suse.com>
2187 PP/13 tls_peerdn now print-escaped for spool files.
2188 Observed some $tls_peerdn in wild which contained \n, which resulted
2189 in spool file corruption.
2191 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
2192 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
2193 or write after TLS renegotiation, which otherwise led to messages
2196 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
2197 as a tracking header (ie: a signed header comes before the signature).
2198 Patch from Wolfgang Breyha.
2200 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
2201 comma-sep list; embedded commas doubled.
2203 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
2205 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
2207 Report and patch from Dmitry Banschikov.
2209 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
2210 Removed SSL_clear() after SSL_new() which led to protocol negotiation
2211 failures. We appear to now support TLS1.1+ with Exim.
2213 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
2214 lets Exim select keys and certificates based upon TLS SNI from client.
2215 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
2216 before an outbound SMTP session. New log_selector, +tls_sni.
2218 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
2219 NULL dereference. Report and patch from Alun Jones.
2221 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
2222 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
2223 Not seeing resolver debug output on NetBSD, but suspect this is a
2224 resolver implementation change.
2226 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
2227 Left warnings. Added "eximon gdb" invocation mode.
2229 PP/21 Defaulting "accept_8bitmime" to true, not false.
2231 PP/22 Added -bw for inetd wait mode support.
2233 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
2234 locate the relevant includes and libraries. Made this the default.
2236 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
2237 Bugzilla 1246, report and most of solution from Tomasz Kusy.
2239 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
2240 This may cause build issues on older platforms.
2242 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
2243 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
2244 gnutls_require_mac & gnutls_require_protocols (no longer supported).
2245 Added SNI support via GnuTLS too.
2246 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
2248 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
2250 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
2251 Applied second patch from Janne, implementing suggestion to default
2252 multiple-strings-in-record handling to match SPF spec.
2254 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
2256 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
2257 read-only, out of scope).
2258 Patch from Wolfgang Breyha, report from Stuart Northfield.
2260 PP/29 Fix three issues highlighted by clang analyser static analysis.
2261 Only crash-plausible issue would require the Cambridge-specific
2262 iplookup router and a misconfiguration.
2263 Report from Marcin Mirosław.
2265 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
2267 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
2268 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
2269 As part of this, removing so much warning spew let me fix some minor
2270 real issues in debug logging.
2272 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
2273 assignment on my part. Fixed.
2275 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
2276 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
2277 Janne Snabb (who went above and beyond: thank you).
2279 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
2280 string otherwise requires a connection and a bunch more work and it's
2281 relatively easy to get wrong. Should also expose TLS library linkage
2284 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
2285 64-bit ${eval} (JH/03).
2287 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
2288 GNU libc to support some of the 64-bit stuff, should not lead to
2289 conflicts. Defined before os.h is pulled in, so if a given platform
2290 needs to override this, it can.
2292 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
2293 protection layer was required, which is not implemented.
2294 Bugzilla 1254, patch from Wolfgang Breyha.
2296 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
2297 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
2298 tls_dhparam take prime identifiers. Also unbreak combination of
2299 OpenSSL+DH_params+TLSSNI.
2301 PP/39 Disable SSLv2 by default in OpenSSL support.
2307 PP/01 Solaris build fix for Oracle's LDAP libraries.
2308 Bugzilla 1109, patch from Stephen Usher.
2310 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
2312 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
2315 TF/02 Fix a couple more cases where we did not log the error message
2316 when unlink() failed. See also change 4.74-TF/03.
2318 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
2319 lock up or crash if it happened to be inside a call to libc when it
2320 got a SIGUSR1 from exiwhat.
2322 The SIGUSR1 handler appends the current process status to the process
2323 log which is later printed by exiwhat. It used to use the general
2324 purpose logging code to do this, but several functions it calls are
2325 not safe for signals.
2327 The new output code in the SIGUSR1 handler is specific to the process
2328 log, and simple enough that it's easy to inspect for signal safety.
2329 Removing some special cases also simplifies the general logging code.
2330 Removing the spurious timestamps from the process log simplifies
2333 TF/04 Improved ratelimit ACL condition.
2335 The /noupdate option has been deprecated in favour of /readonly which
2336 has clearer semantics. The /leaky, /strict, and /readonly update modes
2337 are mutually exclusive. The update mode is no longer included in the
2338 database key; it just determines when the database is updated. (This
2339 means that when you upgrade Exim will forget old rate measurements.)
2341 Exim now checks that the per_* options are used with an update mode that
2342 makes sense for the current ACL. For example, when Exim is processing a
2343 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
2344 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
2345 must specify per_mail/readonly. If you omit the update mode it defaults to
2346 /leaky where that makes sense (as before) or /readonly where required.
2348 The /noupdate option is now undocumented but still supported for
2349 backwards compatibility. It is equivalent to /readonly except that in
2350 ACLs where /readonly is required you may specify /leaky/noupdate or
2351 /strict/noupdate which are treated the same as /readonly.
2353 A useful new feature is the /count= option. This is a generalization
2354 of the per_byte option, so that you can measure the throughput of other
2355 aggregate values. For example, the per_byte option is now equivalent
2356 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
2358 The per_rcpt option has been generalized using the /count= mechanism
2359 (though it's more complicated than the per_byte equivalence). When it is
2360 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
2361 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
2362 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
2363 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
2364 Note that using per_rcpt with a non-readonly update mode in more than
2365 one ACL will cause the recipients to be double-counted. (The per_mail
2366 and per_byte options don't have this problem.)
2368 The handling of very low rates has changed slightly. If the computed rate
2369 is less than the event's count (usually one) then this event is the first
2370 after a long gap. In this case the rate is set to the same as this event's
2371 count, so that the first message of a spam run is counted properly.
2373 The major new feature is a mechanism for counting the rate of unique
2374 events. The new per_addr option counts the number of different
2375 recipients that someone has sent messages to in the last time period. It
2376 behaves like per_rcpt if all the recipient addresses are different, but
2377 duplicate recipient addresses do not increase the measured rate. Like
2378 the /count= option this is a general mechanism, so the per_addr option
2379 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
2380 example, measure the rate that a client uses different sender addresses
2381 with the options per_mail/unique=$sender_address. There are further
2382 details in the main documentation.
2384 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
2386 TF/06 Removed a few PCRE remnants.
2388 TF/07 Automatically extract Exim's version number from tags in the git
2389 repository when doing development or release builds.
2391 PP/02 Raise smtp_cmd_buffer_size to 16kB.
2392 Bugzilla 879. Patch from Paul Fisher.
2394 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
2395 Heavily based on revision 40f9a89a from Simon Arlott's tree.
2398 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
2400 PP/05 Variable $av_failed, true if the AV scanner deferred.
2401 Bugzilla 1078. Patch from John Horne.
2403 PP/06 Stop make process more reliably on build failure.
2404 Bugzilla 1087. Patch from Heiko Schlittermann.
2406 PP/07 Make maildir_use_size_file an _expandable_ boolean.
2407 Bugzilla 1089. Patch from Heiko Schlittermann.
2409 PP/08 Handle ${run} returning more data than OS pipe buffer size.
2410 Bugzilla 1131. Patch from Holger Weiß.
2412 PP/09 Handle IPv6 addresses with SPF.
2413 Bugzilla 860. Patch from Wolfgang Breyha.
2415 PP/10 GnuTLS: support TLS 1.2 & 1.1.
2417 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
2420 PP/11 match_* no longer expand right-hand-side by default.
2421 New compile-time build option, EXPAND_LISTMATCH_RHS.
2422 New expansion conditions, "inlist", "inlisti".
2424 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
2426 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
2428 PP/14 fix log_write() format string regression from TF/03.
2429 Bugzilla 1152. Patch from Dmitry Isaikin.
2435 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
2437 PP/02 Harmonised TLS library version reporting; only show if debugging.
2438 Layout now matches that introduced for other libraries in 4.74 PP/03.
2440 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
2442 PP/04 New "dns_use_edns0" global option.
2444 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
2447 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
2448 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
2450 TK/01 Updated PolarSSL code to 0.14.2.
2451 Bugzilla 1097. Patch from Andreas Metzler.
2453 PP/07 Catch divide-by-zero in ${eval:...}.
2454 Fixes bugzilla 1102.
2456 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
2459 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
2460 format-string attack -- SECURITY: remote arbitrary code execution.
2462 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
2463 time unintentionally subject to list matching rules, letting the header
2464 cause arbitrary Exim lookups (of items which can occur in lists, *not*
2465 arbitrary string expansion). This allowed for information disclosure.
2467 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
2468 INT_MIN/-1 -- value coerced to INT_MAX.
2474 NM/01 Workaround for PCRE version dependency in version reporting
2477 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
2478 This fixes portability to compilers other than gcc, notably
2479 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
2481 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
2482 makefiles for portability to HP-UX and POSIX correctness.
2484 PP/01 Permit LOOKUP_foo enabling on the make command-line.
2485 Also via indented variable definition in the Makefile.
2486 (Debugging by Oliver Heesakkers).
2488 PP/02 Restore caching of spamd results with expanded spamd_address.
2489 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
2491 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
2492 Improves build reliability. Fix from: Frank Elsner
2494 NM/02 Fix wide character breakage in the rfc2047 coding
2495 Fixes bug 1064. Patch from Andrey N. Oktyabrski
2497 NM/03 Allow underscore in dnslist lookups
2498 Fixes bug 1026. Patch from Graeme Fowler
2500 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
2501 Code patches from Adam Ciarcinski of NetBSD.
2503 NM/04 Fixed exiqgrep to cope with mailq missing size issue
2506 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
2507 is logged, to avoid truncation. Patch from John Horne.
2509 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
2510 Patch from Jakob Hirsch.
2512 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
2513 SQL string expansion failure details.
2514 Patch from Andrey Oktyabrski.
2516 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
2517 Patch from Simon Arlott.
2519 PP/09 New lookups functionality failed to compile on old gcc which rejects
2520 extern declarations in function scope.
2521 Patch from Oliver Fleischmann
2523 PP/10 Use sig_atomic_t for flags set from signal handlers.
2524 Check getgroups() return and improve debugging.
2525 Fixed developed for diagnosis in bug 927 (which turned out to be
2528 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
2529 Patch from Mark Zealey.
2531 PP/12 Bugzilla 1056: Improved spamd server selection.
2532 Patch from Mark Zealey.
2534 PP/13 Bugzilla 1086: Deal with maildir quota file races.
2535 Based on patch from Heiko Schlittermann.
2537 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
2538 Patch from Uwe Doering, sign-off by Michael Haardt.
2540 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
2541 variable declaration deep within a block. Bug and patch from
2544 PP/15 lookups-Makefile IRIX compatibility coercion.
2546 PP/16 Make DISABLE_DKIM build knob functional.
2548 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
2549 Patch by Simon Arlott
2551 TF/03 Fix valgrind.h portability to C89 compilers that do not support
2552 variable argument macros. Our copy now differs from upstream.
2558 TF/01 Failure to get a lock on a hints database can have serious
2559 consequences so log it to the panic log.
2561 TF/02 Log LMTP confirmation messages in the same way as SMTP,
2562 controlled using the smtp_confirmation log selector.
2564 TF/03 Include the error message when we fail to unlink a spool file.
2566 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
2567 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
2568 for maintaining out-of-tree patches for some time.
2570 PP/01 Bugzilla 139: Documentation and portability issues.
2571 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
2572 Handle per-OS dynamic-module compilation flags.
2574 PP/02 Let /dev/null have normal permissions.
2575 The 4.73 fixes were a little too stringent and complained about the
2576 permissions on /dev/null. Exempt it from some checks.
2577 Reported by Andreas M. Kirchwitz.
2579 PP/03 Report version information for many libraries, including
2580 Exim version information for dynamically loaded libraries. Created
2581 version.h, now support a version extension string for distributors
2582 who patch heavily. Dynamic module ABI change.
2584 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
2585 privilege escalation vulnerability whereby the Exim run-time user
2586 can cause root to append content of the attacker's choosing to
2589 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
2592 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
2593 If dropping privileges for untrusted macros, we disabled normal logging
2594 on the basis that it would fail; for the Exim run-time user, this is not
2595 the case, and it resulted in successful deliveries going unlogged.
2596 Fixed. Reported by Andreas Metzler.
2602 PP/01 Date: & Message-Id: revert to normally being appended to a message,
2603 only prepend for the Resent-* case. Fixes regression introduced in
2604 Exim 4.70 by NM/22 for Bugzilla 607.
2606 PP/02 Include check_rfc2047_length in configure.default because we're seeing
2607 increasing numbers of administrators be bitten by this.
2609 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
2611 PP/03 Bugzilla 994: added openssl_options main configuration option.
2613 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
2615 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
2617 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
2619 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
2620 without a peer certificate, leading to a segfault because of an
2621 assumption that peers always have certificates. Be a little more
2622 paranoid. Problem reported by Martin Tscholak.
2624 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
2625 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
2626 NB: ClamAV planning to remove STREAM in "middle of 2010".
2627 CL also introduces -bmalware, various -d+acl logging additions and
2628 more caution in buffer sizes.
2630 PP/09 Implemented reverse_ip expansion operator.
2632 PP/10 Bugzilla 937: provide a "debug" ACL control.
2634 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
2636 PP/12 Bugzilla 973: Implement --version.
2638 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
2640 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
2642 PP/15 Bugzilla 816: support multiple condition rules on Routers.
2644 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
2645 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
2646 ignore trailing whitespace.
2648 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2650 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2653 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2654 Notification from Dr Andrew Aitchison.
2656 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2657 ExtendedDetectionInfo response format.
2658 Notification from John Horne.
2660 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2663 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2664 XSL and documented dependency on system catalogs, with examples of how
2667 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2670 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2671 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2672 configuration file which is writeable by the Exim user or group.
2674 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2675 of configuration files to cover files specified with the -C option if
2676 they are going to be used with root privileges, not just the default
2679 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2680 option (effectively making it always true).
2682 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2683 files to be used while preserving root privileges.
2685 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2686 that rogue child processes cannot use them.
2688 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2689 run-time user, instead of root.
2691 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2692 Exim run-time user without dropping privileges.
2694 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2695 result string, instead of calling string_vformat() twice with the same
2698 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2699 for other users. Others should always drop root privileges if they use
2700 -C on the command line, even for a whitelisted configure file.
2702 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2704 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2710 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2711 $data_path, and $header_path variables; fixed documentation bugs and
2714 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2715 exipick to access non-standard spools, including the "frozen" queue
2718 NM/01 Bugzilla 965: Support mysql stored procedures.
2719 Patch from Alain Williams
2721 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2723 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2724 Patch from Andreas Metzler
2726 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2727 Patch from Kirill Miazine
2729 NM/05 Bugzilla 671: Added umask to procmail example.
2731 JJ/03 installed exipick 20100323.0, fixing doc bug
2733 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2734 directory. Notification and patch from Dan Rosenberg.
2736 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2738 TK/02 Improve log output when DKIM signing operation fails.
2740 MH/01 Treat the transport option dkim_domain as a colon separated
2741 list, not as a single string, and sign the message with each element,
2742 omitting multiple occurences of the same signer.
2744 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2745 Bugzilla 985, 986. Patch by Simon Arlott
2747 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2748 Patch by Simon Arlott
2750 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2751 MBX locking. Notification from Dan Rosenberg.
2757 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2759 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2761 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2763 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2765 NM/04 Bugzilla 915: Flag broken perl installation during build.
2771 TK/01 Added patch by Johannes Berg that expands the main option
2772 "spamd_address" if it starts with a dollar sign.
2774 TK/02 Write list of recipients to X-Envelope-Sender header when building
2775 the mbox-format spool file for content scanning (suggested by Jakob
2778 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2779 (http://www.dcc-servers.net/) support via dccifd. Activated by
2780 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2782 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2783 by Mark Daniel Reidel <mr@df.eu>.
2785 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2786 When building exim an external PCRE library is now needed -
2787 PCRE is a system library on the majority of modern systems.
2788 See entry on PCRE_LIBS in EDITME file.
2790 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2791 conversation. Added nologin parameter to request.
2792 Patch contributed by Kirill Miazine.
2794 TF/01 Do not log submission mode rewrites if they do not change the address.
2796 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2798 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2799 log files in place. Contributed by Roberto Lima.
2801 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2803 TF/03 Bugzilla 615: When checking the local_parts router precondition
2804 after a local_part_suffix or local_part_prefix option, Exim now
2805 does not use the address's named list lookup cache, since this
2806 contains cached lookups for the whole local part.
2808 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2809 Robert Millan. Documentation is in experimental-spec.txt.
2811 TF/04 Bugzilla 668: Fix parallel build (make -j).
2813 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2815 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2816 Patch provided by Jan Srzednicki.
2818 TF/05 Leading white space used to be stripped from $spam_report which
2819 wrecked the formatting. Now it is preserved.
2821 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2822 that they are available at delivery time.
2824 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2826 TF/08 TLS error reporting now respects the incoming_interface and
2827 incoming_port log selectors.
2829 TF/09 Produce a more useful error message if an SMTP transport's hosts
2830 setting expands to an empty string.
2832 NM/06 Bugzilla 744: EXPN did not work under TLS.
2833 Patch provided by Phil Pennock.
2835 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2836 Patch provided by Richard Godbee.
2838 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2839 acl_smtp_notquit, added index entry.
2841 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2842 Patch provided by Eugene Bujak.
2844 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2845 accept(). Patch provided by Maxim Dounin.
2847 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2848 Patch provided by Phil Pennock.
2850 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2852 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2853 Patch provided by Brad "anomie" Jorsch.
2855 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2856 Patch provided by Dean Brooks.
2858 TK/05 Add native DKIM support (does not depend on external libraries).
2860 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2861 Patch provided by Graeme Fowler.
2863 NM/16 Bugzilla 851: Documentation example syntax fix.
2865 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2867 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2870 NM/19 Bugzilla 745: TLS version reporting.
2871 Patch provided by Phil Pennock.
2873 NM/20 Bugzilla 167: bool: condition support.
2874 Patch provided by Phil Pennock.
2876 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2877 clients. Patch provided by Phil Pennock.
2879 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2880 Patch provided by Brad "anomie" Jorsch.
2882 NM/23 Bugzilla 687: Fix misparses in eximstats.
2883 Patch provided by Heiko Schlittermann.
2885 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2886 Patch provided by Heiko Schlittermann.
2888 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2889 plus update to original patch.
2891 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2893 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2894 Patch provided by David Brownlee.
2896 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2898 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2900 NM/30 Bugzilla 888: TLS documentation bugfixes.
2902 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2904 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2905 Unlike the original bugzilla I have changed all shell scripts in src tree.
2907 NM/33 Bugzilla 898: Transport filter timeout fix.
2908 Patch by Todd Rinaldo.
2910 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2911 Patch by Serge Demonchaux.
2913 NM/35 Bugzilla 39: Base64 decode bug fixes.
2914 Patch by Jakob Hirsch.
2916 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2918 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2920 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2922 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2928 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2929 ALT-N's libdkim that I have put here:
2930 http://duncanthrax.net/exim-experimental/
2932 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2933 were called 'true' and it seems that C99 defines that as a reserved
2934 keyword to be used with 'bool' variable types. That means you could
2935 not include C99-style headers which use bools without triggering
2936 build errors in sieve.c.
2938 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2939 as mailq or other aliases. Changed the --help handling significantly
2940 to do whats expected. exim_usage() emits usage/help information.
2942 SC/01 Added the -bylocaldomain option to eximstats.
2944 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2946 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2948 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2954 PH/01 Another patch from the Sieve maintainer.
2956 PH/02 When an IPv6 address is converted to a string for single-key lookup
2957 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2958 dots are used instead of colons so that keys in lsearch files need not
2959 contain colons. This was done some time before quoting was made available
2960 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2961 (notated using the quote facility) so as to distinguish them from IPv4
2962 keys. This meant that lookups for IP addresses in host lists did not work
2963 for iplsearch lookups.
2965 This has been fixed by arranging for IPv6 addresses to be expressed with
2966 colons if the lookup type is iplsearch. This is not incompatible, because
2967 previously such lookups could never work.
2969 The situation is now rather anomalous, since one *can* have colons in
2970 ordinary lsearch keys. However, making the change in all cases is
2971 incompatible and would probably break a number of configurations.
2973 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2976 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2977 conversion specification without a maximum field width, thereby enabling
2978 a rogue spamd server to cause a buffer overflow. While nobody in their
2979 right mind would setup Exim to query an untrusted spamd server, an
2980 attacker that gains access to a server running spamd could potentially
2981 exploit this vulnerability to run arbitrary code as the Exim user.
2983 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2984 $primary_hostname instead of what libspf2 thinks the hosts name is.
2986 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2987 a directory entry by the name of the lookup key. Previously, if a
2988 symlink pointed to a non-existing file or a file in a directory that
2989 Exim lacked permissions to read, a lookup for a key matching that
2990 symlink would fail. Now it is enough that a matching directory entry
2991 exists, symlink or not. (Bugzilla 503.)
2993 PH/03 The body_linecount and body_zerocount variables are now exported in the
2996 PH/04 Added the $dnslist_matched variable.
2998 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2999 This means they are set thereafter only if the connection becomes
3002 PH/06 Added the client_condition to authenticators so that some can be skipped
3003 by clients under certain conditions.
3005 PH/07 The error message for a badly-placed control=no_multiline_responses left
3006 "_responses" off the end of the name.
3008 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
3010 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
3011 (without spaces) instead of just copying the configuration text.
3013 PH/10 Added the /noupdate option to the ratelimit ACL condition.
3015 PH/11 Added $max_received_linelength.
3017 PH/12 Added +ignore_defer and +include_defer to host lists.
3019 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
3020 way in which PCRE > 7.0 is built.
3022 PH/14 Implemented queue_only_load_latch.
3024 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
3025 MAIL command. The effect was to mangle the value on 64-bit systems.
3027 PH/16 Another patch from the Sieve maintainer.
3029 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
3031 PH/18 If a system quota error occurred while trying to create the file for
3032 a maildir delivery, the message "Mailbox is full" was not appended to the
3033 bounce if the delivery eventually timed out. Change 4.67/27 below applied
3034 only to a quota excession during the actual writing of the file.
3036 PH/19 It seems that peer DN values may contain newlines (and other non-printing
3037 characters?) which causes problems in log lines. The DN values are now
3038 passed through string_printing() before being added to log lines.
3040 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
3041 and InterBase are left for another time.)
3043 PH/21 Added message_body_newlines option.
3045 PH/22 Guard against possible overflow in moan_check_errorcopy().
3047 PH/23 POSIX allows open() to be a macro; guard against that.
3049 PH/24 If the recipient of an error message contained an @ in the local part
3050 (suitably quoted, of course), incorrect values were put in $domain and
3051 $local_part during the evaluation of errors_copy.
3057 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
3058 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
3061 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
3062 issue a MAIL command.
3064 PH/02 In an ACL statement such as
3066 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
3068 if a client was not listed at all, or was listed with a value other than
3069 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
3070 the condition was not true (as it should be), so access was not denied.
3071 The bug was that the ! inversion was incorrectly passed on to the second
3072 item. This has been fixed.
3074 PH/03 Added additional dnslists conditions == and =& which are different from
3075 = and & when the dns lookup returns more than one IP address.
3077 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
3078 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
3080 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
3081 FSYNC, which compiles an option called disable_fsync that allows for
3082 bypassing fsync(). The documentation is heavily laced with warnings.
3084 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
3086 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
3087 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
3088 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
3089 including adding "make clean"; (3) Added -fPIC when compiling the test
3090 dynamically loaded module, to get rid of a warning.
3092 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
3093 message fails, move_frozen_messages = true and ignore_bounce_errors_after
3094 = 0s. The bug is otherwise harmless.
3096 PH/07 There was a bug in the dovecot authenticator such that the value of
3097 $auth1 could be overwritten, and so not correctly preserved, after a
3098 successful authentication. This usually meant that the value preserved by
3099 the server_setid option was incorrect.
3101 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
3103 PH/09 Installed PCRE release 7.0.
3105 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
3106 run for batched SMTP input. It is now run at the start of every message
3107 in the batch. While fixing this I discovered that the process information
3108 (output by running exiwhat) was not always getting set for -bs and -bS
3109 input. This is fixed, and it now also says "batched" for BSMTP.
3111 PH/11 Added control=no_pipelining.
3113 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
3114 patch, slightly modified), and move the expansion of helo_data till after
3115 the connection is made in the smtp transport (so it can use these
3118 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
3120 PH/14 Added log_selector = +pid.
3122 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
3124 PH/16 Add ${if forany and ${if forall.
3126 PH/17 Added dsn_from option to vary the From: line in DSNs.
3128 PH/18 Flush SMTP output before performing a callout, unless control =
3129 no_callout_flush is set.
3131 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
3132 was true (the default) a successful delivery failed to delete the retry
3133 item, thus causing premature timeout of the address. The bug is now
3136 PH/20 Added hosts_avoid_pipelining to the smtp transport.
3138 PH/21 Long custom messages for fakedefer and fakereject are now split up
3139 into multiline responses in the same way that messages for "deny" and
3140 other ACL rejections are.
3142 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
3143 with slight modification.
3145 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
3146 draft, changing the syntax and factoring some duplicate code".
3148 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
3149 for deliveries of the second and subsequent messages over the same SMTP
3152 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
3153 ${reduce, with only minor "tidies".
3155 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
3157 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
3158 expansion side effects.
3160 PH/27 When a message times out after an over-quota error from an Exim-imposed
3161 quota, the bounce message says "mailbox is full". This message was not
3162 being given when it was a system quota that was exceeded. It now should
3165 MH/03 Made $recipients available in local_scan(). local_scan() already has
3166 better access to the recipient list through recipients_list[], but
3167 $recipients can be useful in postmaster-provided expansion strings.
3169 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
3170 in the case of a MAIL command with additional options following the
3171 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
3172 were accidentally chopped off.
3174 PH/29 SMTP synchronization checks are implemented when a command is read -
3175 there is a check that no more input is waiting when there shouldn't be
3176 any. However, for some commands, a delay in an ACL can mean that it is
3177 some time before the response is written. In this time, more input might
3178 arrive, invalidly. So now there are extra checks after an ACL has run for
3179 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
3180 pipelining has not been advertised.
3182 PH/30 MH's patch to allow iscntrl() characters to be list separators.
3184 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
3185 returned in the SMTP response when smtp_return_error_details was false.
3186 This has been fixed.
3188 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
3189 instead of the C I/O that was originally supplied, because problems were
3190 reported on Solaris.
3192 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
3193 Exim which did not show up earlier: it was assuming that a call to
3194 SSL_CTX_set_info_callback() might give an error value. In fact, there is
3195 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
3196 was a macro that became an assignment, so it seemed to work. This has
3197 changed to a proper function call with a void return, hence the compile
3198 error. Exim's code has been fixed.
3200 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
3203 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
3205 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
3207 PH/37 If a message is not accepted after it has had an id assigned (e.g.
3208 because it turns out to be too big or there is a timeout) there is no
3209 "Completed" line in the log. When some messages of this type were
3210 selected by exigrep, they were listed as "not completed". Others were
3211 picked up by some special patterns. I have improved the selection
3212 criteria to be more general.
3214 PH/38 The host_find_failed option in the manualroute router can now be set
3215 to "ignore", to completely ignore a host whose IP address cannot be
3216 found. If all hosts are ignored, the behaviour is controlled by the new
3217 host_all_ignored option.
3219 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
3220 homing or because of multiple MX records with /mx) generated more than
3221 one IP address, and the following item turned out to be the local host,
3222 all the secondary addresses of the first item were incorrectly removed
3223 from the list, along with the local host and any following hosts (which
3224 is what is supposed to happen).
3226 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
3227 whoever called Exim into the -H file. In the case of the daemon it was
3228 behaving confusingly. When first started, it used values for whoever
3229 started the daemon, but after a SIGHUP it used the Exim user (because it
3230 calls itself on a restart). I have changed the code so that it now always
3233 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
3234 message are rejected with the same error (e.g. no authentication or bad
3235 sender address), and a DATA command is nevertheless sent (as can happen
3236 with PIPELINING or a stupid MUA), the error message that was given to the
3237 RCPT commands is included in the rejection of the DATA command. This is
3238 intended to be helpful for MUAs that show only the final error to their
3241 PH/42 Another patch from the Sieve maintainer.
3243 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
3244 Eximstats - Fixed some broken HTML links and added missing column headers
3246 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
3247 columns for Rejects, Temp Rejects, Ham, and Spam rows.
3249 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
3251 PH/43 Yet another patch from the Sieve maintainer.
3253 PH/44 I found a way to check for a TCP/IP connection going away before sending
3254 the response to the final '.' that terminates a message, but only in the
3255 case where the client has not sent further data following the '.'
3256 (unfortunately, this is allowed). However, in many cases there won't be
3257 any further data because there won't be any more messages to send. A call
3258 to select() can be used: if it shows that the input is "ready", there is
3259 either input waiting, or the socket has been closed. An attempt to read
3260 the next input character can distinguish the two cases. Previously, Exim
3261 would have sent an OK response which the client would never have see.
3262 This could lead to message repetition. This fix should cure that, at
3263 least in a lot of common cases.
3265 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
3266 advertised in response to EHLO.
3272 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
3273 fixed by 4.65/MH/01 (is this a record?) are fixed:
3275 (i) An empty string was always treated as zero by the numeric comparison
3276 operators. This behaviour has been restored.
3278 (ii) It is documented that the numeric comparison operators always treat
3279 their arguments as decimal numbers. This was broken in that numbers
3280 starting with 0 were being interpreted as octal.
3282 While fixing these problems I realized that there was another issue that
3283 hadn't been noticed. Values of message_size_limit (both the global option
3284 and the transport option) were treated as octal if they started with 0.
3285 The documentation was vague. These values are now always treated as
3286 decimal, and I will make that clear in the documentation.
3292 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
3293 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
3296 MH/01 Don't check that the operands of numeric comparison operators are
3297 integers when their expansion is in "skipping" mode (fixes bug
3298 introduced by 4.64-PH/07).
3300 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
3301 child addresses, Exim now panics and dies. Previously, because the count
3302 is held in a short int, deliveries were likely to be lost. As such a
3303 large number of recipients for a single message is ridiculous
3304 (performance will be very, very poor), I have chosen to impose a limit
3305 rather than extend the field.
3311 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
3312 leftover -K file (the existence of which was triggered by #402).
3313 While we were at it, introduced process PID as part of the -K
3314 filename. This should rule out race conditions when creating
3317 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
3318 processing considerably. Previous code took too long for large mails,
3319 triggering a timeout which in turn triggers #401.
3321 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
3322 in the DK code in transports.c. sendfile() is not really portable,
3323 hence the _LINUX specificness.
3325 TF/01 In the add_headers option to the mail command in an Exim filter,
3326 there was a bug that Exim would claim a syntax error in any
3327 header after the first one which had an odd number of characters
3330 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
3331 callout verification, Exim cached a "reject" for the entire domain. This
3332 is correct for most verifications, but it is not correct for a recipient
3333 verification with use_sender or use_postmaster set, because in that case
3334 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
3335 case of MAIL FROM:<> rejection from other early rejections (e.g.
3336 rejection of HELO). When verifying a recipient using a non-null MAIL
3337 address, the cache is ignored if it shows MAIL FROM:<> rejection.
3338 Whatever the result of the callout, the value of the domain cache is
3339 left unchanged (for any other kind of callout, getting as far as trying
3340 RCPT means that the domain itself is ok).
3342 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
3345 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
3346 manifest itself as EPIPE rather than ECONNECT. When tidying away a
3347 session, the daemon ignores ECONNECT errors and logs others; it now
3348 ignores EPIPE as well.
3350 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
3351 (quoted-printable decoding).
3353 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
3354 later the small subsequent patch to fix an introduced bug.
3356 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
3358 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
3360 PH/08 An error is now given if message_size_limit is specified negative.
3362 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
3363 to be given (somewhat) arbitrary names.
3365 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
3368 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
3369 miscellaneous code fixes
3371 PH/10 Added the log_reject_target ACL modifier to specify where to log
3374 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
3375 hostname. This is wrong, because it relates to the incoming message (and
3376 probably the interface on which it is arriving) and not to the outgoing
3377 callout (which could be using a different interface). This has been
3378 changed to use the value of the helo_data option from the smtp transport
3379 instead - this is what is used when a message is actually being sent. If
3380 there is no remote transport (possible with a router that sets up host
3381 addresses), $smtp_active_hostname is used.
3383 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
3384 tweaks were necessary in order to get it to work (see also 21 below):
3385 (a) The code assumed that strncpy() returns a negative number on buffer
3386 overflow, which isn't the case. Replaced with Exim's string_format()
3388 (b) There were several signed/unsigned issues. I just did the minimum
3389 hacking in of casts. There is scope for a larger refactoring.
3390 (c) The code used strcasecmp() which is not a standard C function.
3391 Replaced with Exim's strcmpic() function.
3392 (d) The code set only $1; it now sets $auth1 as well.
3393 (e) A simple test gave the error "authentication client didn't specify
3394 service in request". It would seem that Dovecot has changed its
3395 interface. Fortunately there's a specification; I followed it and
3396 changed what the client sends and it appears to be working now.
3398 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
3401 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
3402 address A is aliased to B and C, where B exists and C does not. Without
3403 -v the output is "A verified" because verification stops after a
3404 successful redirection if more than one address is generated. However,
3405 with -v the child addresses are also verified. Exim was outputting "A
3406 failed to verify" and then showing the successful verification for C,
3407 with its parentage. It now outputs "B failed to verify", showing B's
3408 parentage before showing the successful verification of C.
3410 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
3411 look up a TXT record in a specific list after matching in a combined
3414 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
3415 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
3416 they consult the DNS. I had assumed they would set it the way they
3417 wanted; and indeed my experiments on Linux seem to show that in some
3418 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
3419 To be on the safe side, however, I have now made the interface to
3420 host_find_byname() similar to host_find_bydns(), with an argument
3421 containing the DNS resolver options. The host_find_byname() function now
3422 sets these options at its start, just as host_find_bydns() does. The smtp
3423 transport options dns_qualify_single and dns_search_parents are passed to
3424 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
3425 of host_find_byname() use the default settings of RES_DEFNAMES
3426 (qualify_single) but not RES_DNSRCH (search_parents).
3428 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
3429 spool_read_header() do less string testing, by means of a preliminary
3430 switch on the second character of optional "-foo" lines. (This is
3431 overdue, caused by the large number of possibilities that now exist.
3432 Originally there were few.) While I was there, I also converted the
3433 str(n)cmp tests so they don't re-test the leading "-" and the first
3434 character, in the hope this might squeeze out yet more improvement.
3436 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
3437 flag allowing group syntax was set by the header_syntax check but not
3438 turned off, possible causing trouble later; (2) The flag was not being
3439 set at all for the header_verify test, causing "group"-style headers to
3440 be rejected. I have now set it in this case, and also caused header_
3441 verify to ignore an empty address taken from a group. While doing this, I
3442 came across some other cases where the code for allowing group syntax
3443 while scanning a header line wasn't quite right (mostly, not resetting
3444 the flag correctly in the right place). These bugs could have caused
3445 trouble for malformed header lines. I hope it is now all correct.
3447 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
3448 with the "reply" argument non-NULL. The code, however (which originally
3449 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
3450 but it didn't always do it. This confused somebody who was copying the
3451 code for some other use. I have removed all the tests.
3453 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
3454 feature that was used to support insecure browsers during the U.S. crypto
3455 embargo. It requires special client support, and Exim is probably the
3456 only MTA that supported it -- and would never use it because real RSA is
3457 always available. This code has been removed, because it had the bad
3458 effect of slowing Exim down by computing (never used) parameters for the
3459 RSA_EXPORT functionality.
3461 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
3462 authenticator to fail if there's a tab character in the incoming data
3463 (there should never be unless someone is messing about, as it's supposed
3464 to be base64-encoded). Also added, on Timo's advice, the "secured" option
3465 if the connection is using TLS or if the remote IP is the same as the
3466 local IP, and the "valid-client-cert option" if a client certificate has
3469 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
3470 authenticators. This can be used for authorization after authentication
3471 succeeds. (In the case of plaintext, it servers for both authentication
3474 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
3475 if any retry times were supplied.
3477 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
3478 connection, where there is no client IP address to check. In this
3479 situation, the verify now always succeeds.
3481 PH/25 Applied John Jetmore's -Mset patch.
3483 PH/26 Added -bem to be like -Mset, but loading a message from a file.
3485 PH/27 In a string expansion for a processed (not raw) header when multiple
3486 headers of the same name were present, leading whitespace was being
3487 removed from all of them, but trailing whitespace was being removed only
3488 from the last one. Now trailing whitespace is removed from each header
3489 before concatenation. Completely empty headers in a concatenation (as
3490 before) are ignored.
3492 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
3493 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
3495 PH/29 [Removed. This was a change that I later backed out, and forgot to
3496 correct the ChangeLog entry (that I had efficiently created) before
3497 committing the later change.]
3499 PH/30 Exim was sometimes attempting to deliver messages that had suffered
3500 address errors (4xx response to RCPT) over the same connection as other
3501 messages routed to the same hosts. Such deliveries are always "forced",
3502 so retry times are not inspected. This resulted in far too many retries
3503 for the affected addresses. The effect occurred only when there were more
3504 hosts than the hosts_max_try setting in the smtp transport when it had
3505 the 4xx errors. Those hosts that it had tried were not added to the list
3506 of hosts for which the message was waiting, so if all were tried, there
3507 was no problem. Two fixes have been applied:
3509 (i) If there are any address or message errors in an SMTP delivery, none
3510 of the hosts (tried or untried) are now added to the list of hosts
3511 for which the message is waiting, so the message should not be a
3512 candidate for sending over the same connection that was used for a
3513 successful delivery of some other message. This seems entirely
3514 reasonable: after all the message is NOT "waiting for some host".
3515 This is so "obvious" that I'm not sure why it wasn't done
3516 previously. Hope I haven't missed anything, but it can't do any
3517 harm, as the worst effect is to miss an optimization.
3519 (ii) If, despite (i), such a delivery is accidentally attempted, the
3520 routing retry time is respected, so at least it doesn't keep
3521 hammering the server.
3523 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
3524 in ${readsocket because some servers need this prod.
3526 PH/32 Added some extra debug output when updating a wait-xxx database.
3528 PH/33 The hint "could be header name not terminated by colon", which has been
3529 given for certain expansion errors for a long time, was not being given
3530 for the ${if def:h_colon_omitted{... case.
3532 PH/34 The spec says: "With one important exception, whenever a domain list is
3533 being scanned, $domain contains the subject domain." There was at least
3534 one case where this was not true.
3536 PH/35 The error "getsockname() failed: connection reset by peer" was being
3537 written to the panic log as well as the main log, but it isn't really
3538 panic-worthy as it just means the connection died rather early on. I have
3539 removed the panic log writing for the ECONNRESET error when getsockname()
3542 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
3543 runs only) independently of the message's sender address. This meant
3544 that, if the 4xx error was in fact related to the sender, a different
3545 message to the same recipient with a different sender could confuse
3546 things. In particular, this can happen when sending to a greylisting
3547 server, but other circumstances could also provoke similar problems.
3548 I have changed the default so that the retry time for these errors is now
3549 based a combination of the sender and recipient addresses. This change
3550 can be overridden by setting address_retry_include_sender=false in the
3553 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
3554 remote server are returned as part of bounce messages. This was not
3555 happening for LMTP over a pipe (the lmtp transport), but now it is the
3556 same for both kinds of LMTP.
3558 PH/38 Despite being documented as not happening, Exim was rewriting addresses
3559 in header lines that were in fact CNAMEs. This is no longer the case.
3561 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
3562 and queue runs started by the daemon processed all messages. This has
3563 been fixed so that -R and -S can now usefully be given with -q<time>.
3565 PH/40 Import PCRE release 6.7 (fixes some bugs).
3567 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
3569 PH/42 Give an error if -q is specified more than once.
3571 PH/43 Renamed the variables $interface_address and $interface_port as
3572 $received_ip_address and $received_port, to make it clear that these
3573 values apply to message reception, and not to the outgoing interface when
3574 a message is delivered. (The old names remain recognized, of course.)
3576 PH/44 There was no timeout on the connect() call when using a Unix domain
3577 socket in the ${readsocket expansion. There now is.
3579 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
3580 be meaningful with "accept".
3582 SC/01 Eximstats V1.43
3583 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
3585 SC/02 Eximstats V1.44
3586 Use a glob alias rather than an array ref in the generated
3587 parser. This improves both readability and performance.
3589 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
3590 Collect SpamAssassin and rejection statistics.
3591 Don't display local sender or destination tables unless
3592 there is data to show.
3593 Added average volumes into the top table text output.
3595 SC/04 Eximstats V1.46
3596 Collect data on the number of addresses (recipients)
3597 as well as the number of messages.
3599 SC/05 Eximstats V1.47
3600 Added 'Message too big' to the list of mail rejection
3601 reasons (thanks to Marco Gaiarin).
3603 SC/06 Eximstats V1.48
3604 Mainlog lines which have GMT offsets and are too short to
3605 have a flag are now skipped.
3607 SC/07 Eximstats V1.49 (Alain Williams)
3608 Added the -emptyok flag.
3610 SC/08 Eximstats V1.50
3611 Fixes for obtaining the IP address from reject messages.
3613 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
3614 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
3615 whitespace changes from 4.64-PH/27
3617 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
3620 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
3621 are found, allow negative numbers in numeric criteria)
3623 JJ/06 exipick.20061117.2, added new $message_body_missing variable
3625 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
3626 to match changes made in 4.64-PH/43
3628 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
3630 PH/47 Put in an explicit test for a DNS lookup of an address record where the
3631 "domain" is actually an IP address, and force a failure. This locks out
3632 those revolvers/nameservers that support "A-for-A" lookups, in
3633 contravention of the specifications.
3635 PH/48 When a host name was looked up from an IP address, and the subsequent
3636 forward lookup of the name timed out, the host name was left in
3637 $sender_host_name, contrary to the specification.
3639 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
3640 restricted to single-key lookups, Exim was not diagnosing an error if
3641 * or *@ was used with a query-style lookup.
3643 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
3645 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
3646 long ago, but noone interested enough thought of it. Let's just say that
3647 the "1.1" means that there are some new functions that weren't there at
3648 some point in the past.
3650 PH/51 Error processing for expansion failure of helo_data from an smtp
3651 transport during callout processing was broken.
3653 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3654 tested/used via the -bh/-bhc/-bs options.
3656 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3657 bug, fixed in subsequent PCRE releases).
3659 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3660 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3662 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3668 SC/01 Use a glob alias rather than an array ref in eximstats generated
3669 parser. This improves both readability and performance.
3671 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3672 Don't display local sender or destination tables in eximstats unless
3673 there is data to show.
3674 Added average volumes into the eximstats top table text output.
3676 SC/03 Collect data on the number of addresses (recipients) as well
3677 as the number of messages in eximstats.
3679 TF/01 Correct an error in the documentation for the redirect router. Exim
3680 does (usually) call initgroups() when daemonizing.
3682 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3683 with consistent privilege compared to when running as a daemon.
3685 TF/03 Note in the spec that $authenticated_id is not set for local
3686 submissions from trusted users.
3688 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3689 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3691 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3692 by adding some example configuration directives to the default
3693 configuration file. A little bit of work is required to uncomment the
3694 directives and define how usernames and passwords are checked, but
3695 there is now a framework to start from.
3697 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3698 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3699 without this. I don't know how relevant this is to other LDAP libraries.
3701 PH/02 Add the verb name to the "unknown ACL verb" error.
3703 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3705 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3707 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3708 directory not expanded when it should be if an expanded home directory
3709 was set for the address (which is overridden by the transport).
3711 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3714 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3715 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3716 because it is too late at that time, and has no effect.
3718 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3719 security issue with \' (bugzilla #107). I could not use the
3720 PQescapeStringConn() function, because it needs a PGconn value as one of
3723 PH/08 When testing addresses using -bt, indicate those final addresses that
3724 are duplicates that would not cause an additional delivery. At least one
3725 person was confused, thinking that -bt output corresponded to deliveries.
3726 (Suppressing duplicates isn't a good idea as you lose the information
3727 about possibly different redirections that led to the duplicates.)
3729 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3730 systems where poll() doesn't work, in particular OS X.
3732 PH/10 Added more information to debugging output for retry time not reached.
3734 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3735 operations in malware.c.
3737 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3740 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3741 syslog_duplication set false, log lines that would normally be written
3742 both the the main log and to the reject log were not written to syslog at
3745 PH/14 In the default configuration, change the use of "message" in ACL warn
3746 statements to "add_header".
3748 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3749 not followed by a command (e.g. "seen endif").
3751 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3752 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3755 PH/17 Added extra conditions to the default value of delay_warning_condition
3759 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3760 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3761 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3764 The Auto-Submitted: and various List- headers are standardised, whereas I
3765 don't think Precedence: ever was.
3767 PH/18 Refactored debugging code in route_finduser() to show more information,
3768 in particular, the error code if getpwnam() issues one.
3770 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3771 This is apparently needed in addition to the PH/07 change above to avoid
3772 any possible encoding problems.
3774 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3775 but not after initializing Perl.
3777 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3778 output them only if debugging. By default they are written stderr,
3779 apparently, which is not desirable.
3781 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3784 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3787 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3789 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3790 authenticated or an ident call has been made. Suppress the default
3791 values for $authenticated_id and $authenticated_sender (but permit -oMai
3792 and -oMas) when testing with -bh.
3794 PH/24 Re-jigged the order of the tests in the default configuration so that the
3795 tests for valid domains and recipients precede the DNS black list and CSA
3796 tests, on the grounds that those ones are more expensive.
3798 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3799 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3800 This bug exists in every version of Exim that I still have, right back to
3803 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3804 However, an attempt to turn on case-sensitivity in a regex key by
3805 including (?-i) didn't work because the subject string was already
3806 lowercased, and the effects were non-intuitive. It turns out that a
3807 one-line patch can be used to allow (?-i) to work as expected.
3813 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3814 other effects) broke the use of negated acl sub-conditions.
3816 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3819 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3820 "Deny" causes Exim to reject the incoming connection with a 554 error.
3821 Unfortunately, if there is a major crisis, such as a disk failure,
3822 tcp-wrappers gives "deny", whereas what one would like would be some
3823 kind of temporary error. A kludge has been added to help with this.
3824 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3825 554 error is used if errno is still zero or contains ENOENT (which occurs
3826 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3829 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3831 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3832 errors. Otherwise a message that provokes a temporary error (when other
3833 messages do not) can cause a whole host to time out.
3835 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3836 addresses were routed directly to files or pipes from a redirect router.
3837 File deliveries just didn't batch; pipe deliveries might have suffered
3840 PH/06 A failure to get a lock for a hints database would erroneously always say
3841 "Failed to get write lock", even when it was really a read lock.
3843 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3844 of 0600. This has been changed to use the value of the lockfile_mode
3845 option (which defaults to 0600).
3847 PH/08 Applied small patch from the Sieve maintainer.
3849 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3850 folder from quota calculations, a direct delivery into this folder messed
3851 up the contents of the maildirsize file. This was because the regex was
3852 used only to exclude .Trash (or whatever) when the size of the mailbox
3853 was calculated. There was no check that a delivery was happening into an
3854 excluded directory. This bug has been fixed by ignoring all quota
3855 processing for deliveries into excluded directories.
3857 PH/10 Added the maildirfolder_create_regex option to appendfile.
3863 PH/01 The code for finding all the local interface addresses on a FreeBSD
3864 system running IPv6 was broken. This may well have applied to all BSD
3865 systems, as well as to others that have similar system calls. The broken
3866 code found IPv4 interfaces correctly, but gave incorrect values for the
3867 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3868 that it would not match correctly against @[] and not recognize the IPv6
3871 PH/02 The ipliteral router was not recognizing addresses of the form user@
3872 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3874 PH/03 Added disable_ipv6.
3876 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3877 decoded form, because it is most often used to construct To: headers
3878 lines in autoreplies, and the decoded form may well be syntactically
3879 invalid. However, $reply_address has leading white space removed, and all
3880 newlines turned into spaces so that the autoreply transport does not
3883 PH/05 If group was specified without a user on a router, and no group or user
3884 was specified on a transport, the group from the router was ignored.
3886 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3887 for visible compile-time settings that can be used to change these
3888 numbers, for those that want even more. Backwards compatibility with old
3889 spool files has been maintained. However, going back to a previous Exim
3890 release will lost any variables that are in spool files.
3892 PH/07 Two small changes when running in the test harness: increase delay when
3893 passing a TCP/IP connection to a new process, in case the original
3894 process has to generate a bounce, and remove special handling of
3895 127.0.0.2 (sic), which is no longer necessary.
3897 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3898 be the same on different OS.
3900 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3903 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3904 whether --show-vars was specified or not
3906 JJ/02 exipick: Added support for new ACL variable spool format introduced
3909 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3910 syntactically invalid From: or Reply-to: line, and a filter used this to
3911 generate an autoreply, and therefore failed to obtain an address for the
3912 autoreply, Exim could try to deliver to a non-existent relative file
3913 name, causing unrelated and misleading errors. What now happens is that
3914 it logs this as a hard delivery error, but does not attempt to create a
3917 PH/11 The exinext utility has a -C option for testing purposes, but although
3918 the given file was scanned by exinext itself; it wasn't being passed on
3919 when Exim was called.
3921 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3922 an end-of-file indication when reading a command response.
3924 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3925 compiled. In many other places in Exim, IPv6 addresses are always
3926 recognized, so I have changed this. It also means that IPv4 domain
3927 literals of the form [IPV4:n.n.n.n] are now always recognized.
3929 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3930 used if the router is not running as root, for example, when verifying at
3931 ACL time, or when using -bh. The debugging output from this situation was
3932 non-existent - all you got was a failure to exec. I have made two
3935 (a) Failures to set uid/gid, the current directory, or a process leader
3936 in a subprocess such as that created by queryprogram now generate
3937 suitable debugging output when -d is set.
3939 (b) The queryprogram router detects when it is not running as root,
3940 outputs suitable debugging information if -d is set, and then runs
3941 the subprocess without attempting to change uid/gid.
3943 PH/15 Minor change to Makefile for building test_host (undocumented testing
3946 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3947 additional section of a DNS packet that returns MX or SRV records.
3948 Instead, it always explicitly searches for A/AAAA records. This avoids
3949 major problems that occur when a DNS server includes only records of one
3950 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3951 fixed another bug: if SRV records were looked up and the corresponding
3952 address records were *not* found in the additional section, the port
3953 values from the SRV records were lost.
3955 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3956 using the correct key (the original address) when searching the retry
3957 rules in order to find which one to use for generating the retry hint.
3959 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3960 adding the default one. Similarly, if it contains a Reply-To: header, the
3961 errors_reply_to option, if set, is not used.
3963 PH/19 When calculating a retry time, Exim used to measure the "time since
3964 failure" by looking at the "first failed" field in the retry record. Now
3965 it does not use this if it is later than than the arrival time of the
3966 message. Instead it uses the arrival time. This makes for better
3967 behaviour in cases where some deliveries succeed, thus re-setting the
3968 "first failed" field. An example is a quota failure for a huge message
3969 when small messages continue to be delivered. Without this change, the
3970 "time since failure" will always be short, possible causing more frequent
3971 delivery attempts for the huge message than are intended.
3972 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3974 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3975 $1, $2, $3) because the numerical variables can be reset during some
3976 expansion items (e.g. "match"), thereby losing the authentication data.
3978 PH/21 Make -bV show the size of off_t variables so that the test suite can
3979 decide whether to run tests for quotas > 2G.
3981 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3982 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3983 filecount value is greater than 2G or if a quota value is greater than 2G
3984 on a system where the size of off_t is not greater than 4, a panic error
3987 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3988 never match. The debug and -bh output now contains an explicit error
3989 message indicating a malformed IPv4 address or mask.
3991 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3992 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3993 PH/23 above applies.
3995 PH/25 Do not write to syslog when running in the test harness. The only
3996 occasion when this arises is a failure to open the main or panic logs
3997 (for which there is an explicit test).
3999 PH/26 Added the /no_tell option to "control=freeze".
4001 PH/27 If a host name lookup failed very early in a connection, for example, if
4002 the IP address matched host_lookup and the reverse lookup yielded a name
4003 that did not have a forward lookup, an error message of the form "no IP
4004 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
4005 could be logged. Now it outputs the IP address instead of "NULL".
4007 PH/28 An enabling patch from MH: add new function child_open_exim2() which
4008 allows the sender and the authenticated sender to be set when
4009 submitting a message from within Exim. Since child_open_exim() is
4010 documented for local_scan(), the new function should be too.
4012 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
4013 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
4014 results in an empty string is now treated as unset.
4016 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
4018 PH/31 Added sender_verify_fail logging option.
4020 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
4021 needed by RFC 822 but not by RFC 2822 was commented out. I have now
4022 tidied the source and removed it altogether.
4024 PH/33 When a queue run was abandoned because the load average was too high, a
4025 log line was always written; now it is written only if the queue_run log
4026 selector is set. In addition, the log line for abandonment now contains
4027 information about the queue run such as the pid. This is always present
4028 in "start" and "stop" lines but was omitted from the "abandon" line.
4030 PH/34 Omit spaces between a header name and the colon in the error message that
4031 is given when verify = headers_syntax fails (if there are lots of them,
4032 the message gets confusing).
4034 PH/35 Change the default for dns_check_names_pattern to allow slashes within
4035 names, as there are now some PTR records that contain slashes. This check
4036 is only to protect against broken name servers that fall over on strange
4037 characters, so the fact that it applies to all lookups doesn't matter.
4039 PH/36 Now that the new test suite is complete, we can remove some of the
4040 special code in Exim that was needed for the old test suite. For example,
4041 sorting DNS records because real resolvers return them in an arbitrary
4042 order. The new test suite's fake resolver always returns records in the
4045 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
4046 bounces) except when queue_only is set, to avoid logging races between
4047 the different processes.
4049 PH/38 Panic-die if .include specifies a non-absolute path.
4051 PH/39 A tweak to the "H" retry rule from its user.
4053 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
4054 a label. They prevented compilation on older perls.
4056 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
4057 a warning to be raised on newish perls.
4059 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
4060 on queue. Changes to match documented behaviour of showing count of
4061 messages matching specified criteria.
4063 PH/40 Changed the default ident timeout from 30s to 5s.
4065 PH/41 Added support for the use of login_cap features, on those BSD systems
4066 that have them, for controlling the resources used by pipe deliveries.
4068 PH/42 The content-scanning code uses fopen() to create files in which to put
4069 message data. Previously it was not paying any attention to the mode of
4070 the files. Exim runs with umask(0) because the rest of the code creates
4071 files with open(), and sets the required mode explicitly. Thus, these
4072 files were ending up world-writeable. This was not a big issue, because,
4073 being within the spool directory, they were not world-accessible. I have
4074 created a function called modefopen, which takes an additional mode
4075 argument. It sets umask(777), creates the file, chmods it to the required
4076 mode, then resets the umask. All the relevant calls to fopen() in the
4077 content scanning code have been changed to use this function.
4079 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
4080 to 24 hours. This avoids potential overflow problems when processing G
4081 and H retry rules. I suspect nobody ever tinkers with this value.
4083 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
4085 PH/45 When the plaintext authenticator is running as a client, the server's
4086 challenges are checked to ensure they are valid base64 strings. By
4087 default, the authentication attempt is cancelled if an invalid string is
4088 received. Setting client_ignore_invalid_base64 true ignores these errors.
4089 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
4090 they are received. Thus, the responses can be made to depend on the
4091 challenges. If an invalid string is ignored, an empty string is placed in
4094 PH/46 Messages that are created by the autoreply transport now contains a
4095 References: header, in accordance with RFCs 2822 and 3834.
4097 PH/47 Added authenticated_sender_force to the smtp transport.
4099 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
4101 PH/49 Installed latest patch from the Sieve maintainer.
4103 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
4104 was also set, the appendfile transport was unnecessarily scanning a
4105 directory of message files (e.g. for maildir delivery) to find the count
4106 of files (along with the size), even though it did not need this
4107 information. It now does the scan only if it needs to find either the
4108 size of the count of files.
4110 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
4112 PH/52 Two bugs concerned with error handling when the smtp transport is
4115 (i) Exim was not creating retry information for temporary errors given
4116 for individual recipients after the DATA command when the smtp transport
4117 was used in LMTP mode. This meant that they could be retried too
4118 frequently, and not timed out correctly.
4120 (ii) Exim was setting the flag that allows error details to be returned
4121 for LMTP errors on RCPT commands, but not for LMTP errors for individual
4122 recipients that were returned after the DATA command.
4124 PH/53 This is related to PH/52, but is more general: for any failing address,
4125 when detailed error information was permitted to be returned to the
4126 sender, but the error was temporary, then after the final timeout, only
4127 "retry timeout exceeded" was returned. Now it returns the full error as
4128 well as "retry timeout exceeded".
4130 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
4131 do this, and (what is worse) MTAs that accept it.
4133 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
4134 will now be deprecated.
4136 PH/56 New os.c-cygwin from the Cygwin maintainer.
4138 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
4139 formats (previously only available in exim formats via -bpr, -bpru,
4140 and -bpra. Now also available in native and exiqgrep formats)
4142 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
4143 with very large, slow to parse queues
4145 JJ/08 exipick: added ! as generic prefix to negate any criteria format
4147 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
4149 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
4150 responses to authentication challenges, though it was showing the
4151 challenges; (ii) I've removed the CR characters from the debug output for
4154 PH/58 Allow for the insertion of a newline as well as a space when a string
4155 is turned into more than one encoded-word during RFC 2047 encoding. The
4156 Sieve code now uses this.
4158 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
4159 data_4xx, lost_connection, tls_required.
4161 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
4162 message was being sent as an SMTP response.
4164 PH/61 Add -l and -k options to exicyclog.
4166 PH/62 When verifying, if an address was redirected to one new address, so that
4167 verification continued, and the new address failed or deferred after
4168 having set something in $address_data, the value of $address_data was not
4169 passed back to the ACL. This was different to the case when no
4170 redirection occurred. The value is now passed back in both cases.
4172 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
4173 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
4174 use login_cap.h, so on its own it isn't the distinguishing feature. The
4175 new name refers directly to the setclassresources() function.
4177 PH/65 Added configuration files for NetBSD3.
4179 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
4181 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
4182 is preferred over IPv4.
4184 PH/68 The bounce_return_message and bounce_return_body options were not being
4185 honoured for bounces generated during the reception of non-SMTP messages.
4186 In particular, this applied to messages rejected by the ACL. This bug has
4187 been fixed. However, if bounce_return_message is true and bounce_return_
4188 body is false, the headers that are returned for a non-SMTP message
4189 include only those that have been read before the error was detected.
4190 (In the case of an ACL rejection, they have all been read.)
4192 PH/69 The HTML version of the specification is now built in a directory called
4193 spec_html instead of spec.html, because the latter looks like a path with
4194 a MIME-type, and this confuses some software.
4196 PH/70 Catch two compiler warnings in sieve.c.
4198 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
4199 function verify_get_ident() calls ip_connect() to connect a socket, but
4200 if the "connect()" function timed out, ip_connect() used to close the
4201 socket. However, verify_get_ident() also closes the socket later, and in
4202 between Exim writes to the log, which may get opened at this point. When
4203 the socket was closed in ip_connect(), the log could get the same file
4204 descriptor number as the socket. This naturally causes chaos. The fix is
4205 not to close the socket in ip_connect(); the socket should be closed by
4206 the function that creates it. There was only one place in the code where
4207 this was missing, in the iplookup router, which I don't think anybody now
4208 uses, but I've fixed it anyway.
4210 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
4211 well as to direct DNS lookups. Otherwise the handling of names in host
4212 lists is inconsistent and therefore confusing.
4218 PH/01 Two changes to the default runtime configuration:
4220 (1) Move the checks for relay_from_hosts and authenticated clients from
4221 after to before the (commented out) DNS black list checks.
4223 (2) Add control=submission to the relay_from_hosts and authenticated
4224 clients checks, on the grounds that messages accepted by these
4225 statements are most likely to be submissions.
4227 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
4229 (1) Generate an error if the third argument for the ${prvs expansion is
4232 (2) Treat a missing third argument of ${prvscheck as if it were an empty
4235 (3) Reset the variables that are obtained from the first argument of
4236 ${prvscheck and used in the second argument before leaving the code,
4237 because their memory is reclaimed, so using them afterwards may do
4240 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
4241 one (it's much easier than Tom thought :-).
4243 (5) Because of (4), we can now allow for the use of $prvscheck_result
4244 inside the third argument.
4246 PH/03 For some reason, the default setting of PATH when running a command from
4247 a pipe transport was just "/usr/bin". I have changed it to
4250 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
4251 anything to be listed in the output from -bV.
4253 PH/05 When a filter generated an autoreply, the entire To: header line was
4254 quoted in the delivery log line, like this:
4256 => >A.N.Other <ano@some.domain> <original@ddress> ...
4258 This has been changed so that it extracts the operative address. There
4259 may be more than one such address. If so, they are comma-separated, like
4262 => >ano@some.domain,ona@other.domain <original@ddress> ...
4264 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
4265 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
4266 not being looked up in the rDNS to get a host name, Exim was showing the
4267 IP address twice in Received: lines, even though the IP addresses were
4268 identical. For example:
4270 Received: from [1.2.3.4] (helo=[1.2.3.4])
4272 However, if the real host name was known, it was omitting the HELO data
4273 if it matched the actual IP address. This has been tidied up so that it
4274 doesn't show the same IP address twice.
4276 PH/07 When both +timestamp and +memory debugging was on, the value given by
4277 $tod_xxx expansions could be wrong, because the tod_stamp() function was
4278 called by the debug printing, thereby overwriting the timestamp buffer.
4279 Debugging no longer uses the tod_stamp() function when +timestamp is set.
4281 PH/08 When the original message was included in an autoreply transport, it
4282 always said "this is a copy of the message, including all the headers",
4283 even if body_only or headers_only was set. It now gives an appropriate
4286 PH/09 Applied a patch from the Sieve maintainer which:
4288 o fixes some comments
4289 o adds the (disabled) notify extension core
4290 o adds some debug output for the result of if/elsif tests
4291 o points to the current vacation draft in the documentation
4292 and documents the missing references header update
4296 o fixes a bug in processing the envelope test (when testing
4297 multiple envelope elements, the last element determined the
4300 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
4301 Electronic Mail") by including:
4303 Auto-submitted: auto-generated
4305 in the messages that it generates (bounce messages and others, such as
4306 warnings). In the case of bounce messages for non-SMTP messages, there was
4307 also a typo: it was using "Auto_submitted" (underscore instead of
4308 hyphen). Since every message generated by Exim is necessarily in response
4309 to another message, thes have all been changed to:
4311 Auto-Submitted: auto-replied
4313 in accordance with these statements in the RFC:
4315 The auto-replied keyword:
4317 - SHOULD be used on messages sent in direct response to another
4318 message by an automatic process,
4320 - MUST NOT be used on manually-generated messages,
4322 - MAY be used on Delivery Status Notifications (DSNs) and Message
4323 Disposition Notifications (MDNs),
4325 - MUST NOT be used on messages generated by automatic or periodic
4326 processes, except for messages which are automatic responses to
4329 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
4330 to the default Received: header definition.
4332 PH/12 Added log selector acl_warn_skipped (default on).
4334 PH/13 After a successful wildlsearch lookup, discard the values of numeric
4335 variables because (a) they are in the wrong storage pool and (b) even if
4336 they were copied, it wouldn't work properly because of the caching.
4338 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
4339 checking when decoding. Apparently there are clients that generate
4340 overlong encoded strings. Why am I not surprised?
4342 PH/15 If the first argument of "${if match_address" was not empty, but did not
4343 contain an "@" character, Exim crashed. Now it writes a panic log message
4344 and treats the condition as false.
4346 PH/16 In autoreply, treat an empty string for "once" the same as unset.
4348 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
4349 extension "envelope-auth". The code is finished and in agreement with
4350 other implementations, but there is no documentation so far and in fact,
4351 nobody wrote the draft yet. This extension is currently #undef'ed, thus
4352 not changing the active code.
4354 Print executed "if" and "elsif" statements when debugging is used. This
4355 helps a great deal to understand what a filter does.
4357 Document more things not specified clearly in RFC3028. I had all this
4358 sorted out, when out of a sudden new issues came to my mind. Oops."
4360 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
4363 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
4364 canonical form (as documented). However, after a host name lookup from
4365 the IP address, check_host() was doing a simple string comparison with
4366 addresses acquired from the DNS when checking that the found name did
4367 have the original IP as one of its addresses. Since any found IPv6
4368 addresses are likely to be in abbreviated form, the comparison could
4369 fail. Luckily, there already exists a function for doing the comparison
4370 by converting both addresses to binary, so now that is used instead of
4371 the text comparison.
4373 PH/20 There was another similar case to PH/19, when a complete host name was
4374 given in a host list; looking up its IP address could give an abbreviated
4375 form, whereas the current host's name might or might not be abbreviated.
4376 The same fix has been applied.
4382 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
4383 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
4386 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
4387 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
4389 PH/03 Typo: missing ".o" in src/pcre/Makefile.
4391 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
4392 header line, restrict the check to what is listed in RFCs 2369 and 2929.
4393 Also, for "Auto-Submitted", treat anything other than "no" as
4394 non-personal, in accordance with RFC 3834. (Previously it treated
4395 anything starting "auto-" as non-personal.)
4397 TF/01 The control=submission/name=... option had a problem with syntax
4398 errors if the name included a slash character. The /name= option
4399 now slurps the rest of the string, so it can include any characters
4400 but it must come last in the list of options (after /sender_retain
4403 PH/05 Some modifications to the interface to the fake nameserver for the new
4411 TK/01 Added the "success_on_redirect" address verification option. See
4412 NewStuff for rationale and an example.
4414 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
4416 PH/02 Patch to exigrep to allow it to work on syslog lines.
4418 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
4419 fread() to skip over the body file's header line, because in Cygwin the
4420 header line is locked and is inaccessible.
4422 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
4423 co-exist for some time) to make it clear that it is the Exim ID that is
4424 referenced, not the Message-ID: header line.
4426 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
4427 string_format() function, because snprintf() does not exist on all
4430 PH/06 The use of forbid_filter_existstest now also locks out the use of the
4431 ${stat: expansion item.
4433 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
4434 protocol synchronization error", to keep the pedants happy.
4436 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
4437 well as for IRIX systems, when gcc is being used. See the host.c source
4440 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
4442 PH/10 Named domain lists were not working if used in a queue_smtp_domains
4445 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
4446 transport and to the smtp transport in LMTP mode.
4448 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
4450 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
4451 run a filter in a subprocess. This could lead to confusion in subsequent
4452 lookups in the parent process. There should also be a search_tidyup() at
4453 the end of the subprocess.
4455 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
4456 only if the host matched helo_try_verify_hosts, which caused the
4457 verification to occur when the EHLO/HELO command was issued. The ACL just
4458 tested the remembered result. Now, if a previous verification attempt has
4459 not happened, "verify = helo" does it there and then.
4461 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
4463 TK/03 Fix log output including CR from clamd.
4465 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
4466 exist provoked a memory error which could cause a segfault.
4468 PH/15 Installed PCRE 6.2
4470 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
4472 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
4473 of the problem. Specifically, suggested +O2 rather than +O1 for the
4476 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
4478 PH/20 If a delivery was routed to a non-standard port by means of an SRV
4479 record, the port was not correctly logged when the outgoing_port log
4480 selector was set (it logged the transort's default port).
4482 PH/21 Added support for host-specific ports to manualroute, queryprogram,
4483 fallback_hosts, and "hosts" in the smtp transport.
4485 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
4486 host errors such as "Connection refused".
4488 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
4489 authentication with radiusclient 0.4.9:
4491 - Error returned from rc_read_config was caught wrongly
4492 - Username/password not passed on to radius server due to wrong length.
4494 The presumption is that some radiusclient API changes for 4.51/PH/17
4495 were not taken care of correctly. The code is still untested by me (my
4496 Linux distribution still has 0.3.2 of radiusclient), but it was
4497 contributed by a Radius user.
4499 PH/24 When doing a callout, the value of $domain wasn't set correctly when
4500 expanding the "port" option of the smtp transport.
4502 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
4503 while reading a MIME header. Thanks to Tom Hughes for a patch.
4505 PH/24 Include config.h inside local_scan.h so that configuration settings are
4508 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
4509 that in an ACL for RCPT (for example), you can examine exactly what was
4512 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
4513 commands, but it was not correctly comparing the address with the actual
4514 client host address. Thus, it would show the EHLO address in Received:
4515 header lines when this was not necessary.
4517 PH/27 Added the % operator to ${eval:}.
4519 PH/28 Exim tries to create and chdir to its spool directory when it starts;
4520 it should be ignoring failures (because with -C, for example, it has lost
4521 privilege). It wasn't ignoring creation failures other than "already
4524 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
4527 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
4528 because an input line was too long, either on its own, or by virtue of
4529 too many continuations, the temporary file was not being removed, and the
4530 return code was incorrect.
4532 PH/31 Missing "BOOL" in function definition in filtertest.c.
4534 PH/32 Applied Sieve patches from the maintainer.
4536 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
4538 PH/33 Added "verify = not_blind".
4540 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
4541 Local/Makefile (with some defaults set). These are used in built scripts
4542 such as exicyclog, but they have never been used in the exim_install
4543 script (though there are many overriding facilities there). I have
4544 arranged that the exim_install script now takes note of these two
4547 PH/35 Installed configuration files for Dragonfly.
4549 PH/36 When a locally submitted message by a trusted user did not contain a
4550 From: header, and the sender address was obtained from -f or from an SMTP
4551 MAIL command, and the trusted user did not use -F to supply a sender
4552 name, $originator_name was incorrectly used when constructing a From:
4553 header. Furthermore, $originator_name was used for submission mode
4554 messages from external hosts without From: headers in a similar way,
4555 which is clearly wrong.
4557 PH/37 Added control=suppress_local_fixups.
4559 PH/38 When log_selector = +received_sender was set, and the addition of the
4560 sender made the log line's construction buffer exactly full, or one byte
4561 less than full, an overflow happened when the terminating "\n" was
4564 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
4565 when the result of a list match is failure because a DNS lookup failed.
4567 PH/40 RM_COMMAND is now used in the building process.
4569 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
4570 the "build-* directories that it finds.
4572 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
4573 address in a domain literal was a prefix of an interface address.
4575 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
4576 when verifying a sender address, unless rewrite_headers is false.
4578 PH/44 Wrote a long comment about why errors_to addresses are verified as
4579 recipients, not senders.
4581 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
4582 the ratelimit ACL was added.
4584 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
4586 PH/46 Added extra information about PostgreSQL errors to the error string.
4588 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
4589 avoiding the need to install special zones in a real server. This is
4590 backwards compatible; if it can't find the fake resolver, it drops back.
4591 Thus, both old and new test suites can be run.
4593 TF/02 Added util/ratelimit.pl
4595 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
4596 clock is set back in time.
4598 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
4599 Candler <B.Candler@pobox.com>.
4601 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
4602 OK for addresses that are the result of redirections.
4604 PH/48 A number of further additions for the benefit of the new test suite,
4605 including a fake gethostbyname() that interfaces to the fake DNS resolver
4608 TF/06 The fix for widen_domains has also been applied to qualify_single and
4609 search_parents which are the other dnslookup options that can cause
4612 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
4615 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
4617 TF/07 Exim produced the error message "an SRV record indicated no SMTP
4618 service" if it encountered an MX record with an empty target hostname.
4619 The message is now "an MX or SRV record indicated no SMTP service".
4621 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
4622 if the DNS of the sending site is misconfigured. This is quite a
4623 common situation. This change restores the behaviour of treating a
4624 helo verification defer as a failure.
4626 PH/51 If self=fail was set on a router, the bounce message did not include the
4627 actual error message.
4633 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
4635 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
4636 command itself ended in error, the underlying message about the transport
4637 filter timeout was being overwritten with the pipe command error. Now the
4638 underlying error message should be appended to the second error message.
4640 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
4642 PH/02 Reduce the amount of output that "make" produces by default. Full output
4643 can still be requested.
4645 PH/03 The warning log line about a condition test deferring for a "warn" verb
4646 was being output only once per connection, rather than after each
4647 occurrence (because it was using the same function as for successful
4648 "warn" verbs). This seems wrong, so I have changed it.
4650 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4651 it should not have, which might have caused a crash in the right
4652 circumstances, but probably never did.
4654 PH/04 Installed a modified version of Tony Finch's patch to make submission
4655 mode fix the return path as well as the Sender: header line, and to
4656 add a /name= option so that you can make the user's friendly name appear
4659 TF/03 Added the control = fakedefer ACL modifier.
4661 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4662 Mark Lowes for thorough testing.
4664 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4666 TK/03 Merged latest SRS patch from Miles Wilton.
4668 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4669 with the definition in sysexits.h (which is #included earlier).
4670 Fortunately, Exim does not actually use EX_OK. The code used to try to
4671 preserve the sysexits.h value, by assuming that macro definitions were
4672 scanned for macro replacements. I have been disabused of this notion,
4673 so now the code just undefines EX_OK before #including unistd.h.
4675 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4676 in the smtp transport. When a block could not be written in a single
4677 write() function, the timeout was being re-applied to each part-write.
4678 This seems wrong - if the receiver was accepting one byte at a time it
4679 would take for ever. The timeout is now adjusted when this happens. It
4680 doesn't have to be particularly precise.
4682 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4683 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4685 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4686 without a domain if the check to <postmaster@domain> fails.
4688 SC/01 Eximstats: added -xls and the ability to specify output files
4689 (patch written by Frank Heydlauf).
4691 SC/02 Eximstats: use FileHandles for outputting results.
4693 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4695 SC/04 Eximstats: fixed display of large numbers with -nvr option
4697 SC/05 Eximstats: fixed merging of reports with empty tables.
4699 SC/06 Eximstats: added the -include_original_destination flag
4701 SC/07 Eximstats: removed tabs and trailing whitespace.
4703 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4705 TK/06 MBOX spool code: Add real "From " MBOX separator line
4706 so the .eml file is really in mbox format (even though
4707 most programs do not really care). Patch from Alex Miller.
4709 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4710 The latter is generated from $received_to and is only set if the
4711 message has one envelope recipient. SA can use these headers,
4712 obviously out-of-the-box. Patch from Alex Miller.
4714 PH/08 The ${def test on a variable was returning false if the variable's
4715 value was "0", contrary to what the specification has always said!
4716 The result should be true unless the variable is empty.
4718 PH/09 The syntax error of a character other than { following "${if
4719 def:variable_name" (after optional whitespace) was not being diagnosed.
4720 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4721 accidental colon was present, for example, could give incorrect results.
4723 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4724 result is used (not including appendfile, where other changes are about
4727 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4728 This involved changing a lot of size variables from int to off_t. It
4729 should work with maildirs and everything.
4731 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4732 spamd dying while we are connected to it.
4734 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4737 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4738 function for BDB 4.3.
4740 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4742 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4743 the sources for PCRE have been re-arranged and more files are now
4746 PH/15 The code I had for printing potentially long long variables in PH/11
4747 above was not the best (it lost precision). The length of off_t variables
4748 is now inspected at build time, and an appropriate printing format (%ld
4749 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4750 to be "long long int" or "long int". This is needed for the internal
4751 formatting function string_vformat().
4753 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4754 the configuration file to be ":syslog", then the script "guesses" where
4755 the logs files are, rather than using the compiled in default. In our
4756 case the guess is not the same as the compiled default, so the script
4757 suddenly stopped working when I started to use syslog. The patch checks
4758 to see if log_file_path is "". If so, it attempts to read it from exim
4759 with no configuration file to get the compiled in version, before it
4760 falls back to the previous guessing code."
4762 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4763 implementing BATV in an Exim configuration. See NewStuff for the gory
4766 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4767 Makefile that are specific to HP-UX.
4769 PH/18 If the "use_postmaster" option was set for a recipient callout together
4770 with the "random" option, the postmaster address was used as the MAIL
4771 FROM address for the random test, but not for the subsequent recipient
4772 test. It is now used for both.
4774 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4775 patch removes a few documentation additions to RFC 3028, because the
4776 latest draft now contains them. It adds the new en;ascii-case comparator
4777 and a new error check for 8bit text in MIME parts. Comparator and
4778 require names are now matched exactly. I enabled the subaddress
4779 extension, but it is not well tested yet (read: it works for me)."
4781 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4782 rework some of the code of TK/09 above to avoid the hardwired use of
4783 "%lld" and "long long". Replaced the call to snprintf() with a call to
4786 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4787 records point to non-existent hosts", "retry timeout exceeded", and
4788 "retry time not reached for any host after a long failure period".
4790 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4791 experimental DomainKeys support:
4793 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4794 (2) On an error such as an illegally used "control", the wrong name for
4795 the control was given.
4797 These problems did NOT occur unless DomainKeys support was compiled.
4799 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4801 PH/24 Added ${if match_ip condition.
4803 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4804 databases so that it will be absolutely obvious if a crash occurs in the
4805 DB library. This is a regular occurrence (often caused by mis-matched
4808 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4809 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4810 picked up on a user's system that detects such things. There doesn't seem
4811 to be a gcc warning option for this - only an attribute that has to be
4812 put on the function's prototype. It seems that in Fedora Core 4 they have
4813 set this on a number of new functions. No doubt there will be more in due
4816 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4817 specify a transport. However, if an address that was verified by such a
4818 router was the subject of a callout, Exim crashed because it tried to
4819 read the rcpt_include_affixes from the non-existent transport. Now it
4820 just assumes that the setting of that option is false. This bug was
4821 introduced by 4.51/PH/31.
4823 PH/28 Changed -d+all to exclude +memory, because that information is very
4824 rarely of interest, but it makes the output a lot bigger. People tend to
4825 do -d+all out of habit.
4827 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4828 code in os-type was giving problems when libc.so lives in lib64, like on
4831 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4832 aren't the modern standard, and it seems that some systems' include files
4833 don't always have them. Exim was already checking for some of the newer
4834 ones like T_AAAA, and defining it itself. I've added checks for all the
4835 record types that Exim uses.
4837 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4838 not automatically generating a new one, as it is supposed to. This
4839 prevented TLS from working. If the file did exist, but contained invalid
4840 data, a new version was generated, as expected. It was only the case of a
4841 non-existent file that was broken.
4843 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4844 with a change in libdomainkeys > 0.64.
4846 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4847 from DNS. If the selector record carries the flag, it now has
4848 precedence over the domain-wide flag.
4850 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4852 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4853 the use of a port name that isn't defined in /etc/services) occurred, the
4854 message was deferred as in a normal delivery, and thus remained on the
4855 spool, instead of being failed because of the mua_wrapper setting. This
4856 is now fixed, and I tidied up some of the mua_wrapper messages at the
4859 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4860 the messages in a hash of arrays rather than using individual hashes.
4861 This is a bit cleaner and results in dramatic memory savings, albeit
4862 at a slight CPU cost.
4864 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4865 as requested by Marc Sherman.
4867 SC/10 Eximstats: added histograms for user specified patterns as requested
4870 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4872 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4873 fopen() in the content-scanning modules that did not already have it.
4879 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4880 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4882 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4884 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4886 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4887 to test Sieve filters that use "vacation".
4889 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4890 that changes the way the GnuTLS parameters are stored in the cache file.
4891 The new format can be generated externally. For backward compatibility,
4892 if the data in the cache doesn't make sense, Exim assumes it has read an
4893 old-format file, and it generates new data and writes a new file. This
4894 means that you can't go back to an older release without removing the
4897 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4898 work if there are any delivery delays because "one_time" forces the
4899 parent to be marked "delivered", so its unseen clone is never tried
4900 again. For this reason, Exim now forbids the simultaneous setting of
4903 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4904 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4905 entry if you want to know the details. The fix, however, neglected to
4906 consider the case when local delivery batching is involved. The test for
4907 "previously delivered" was not happening when checking to see if an
4908 address could be batched with a previous (undelivered) one; under
4909 certain circumstances this could lead to multiple deliveries to the same
4912 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4913 in its include files, and this causes problems building Exim.
4915 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4916 header_syntax) but Exim was just ignoring anything given after a slash.
4917 In particular, this caused confusion with an attempt to use "verify =
4918 reverse_host_lookup/defer_ok". An error is now given when options are
4919 supplied for verify items that do not have them. (Maybe reverse_host_
4920 lookup should have a defer_ok option, but that's a different point.)
4922 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4923 defined by RFC 821) to 2048, because there were problems with some AUTH
4924 commands, and RFC 1869 says the size should be increased for extended
4925 SMTP commands that take arguments.
4927 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4930 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4931 "unknown" error; now it says that the functionality isn't in the binary.
4933 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4934 an address' error message when a string expansion fails (syntax or
4935 whatever). Otherwise the password may appear in the log. Following change
4936 PH/42 below, there is no longer a chance of it appearing in a bounce
4939 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4941 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4942 of its addresses was ever tried. (Bugzilla bug #2.)
4944 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4945 the result incorrectly in the debug output. (It correctly added a newline
4946 to what was transported.)
4948 TF/01 Added $received_time.
4950 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4951 commented out examples of how to interface to a virus scanner and to
4952 SpamAssassin. Also added commented examples of av_scanner and
4953 spamd_address settings.
4955 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4956 and controls are allowed in which ACLs. There were a couple of minor
4957 errors. Some of the entries in the conditions table (which is a table of
4958 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4959 negation of where the condition IS allowed.
4961 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4963 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4964 header file does not have a version number, so I've had to invent a new
4965 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4966 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4967 radiusclient), but it was contributed by a Radius user.
4969 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4970 files or directories, for OpenSSL.
4972 PH/19 When an Exim process that is running as root has to create an Exim log
4973 file, it does so in a subprocess that runs as exim:exim so as to get the
4974 ownership right at creation (otherwise, other Exim processes might see
4975 the file with the wrong ownership). There was no test for failure of this
4976 fork() call, which would lead to the process getting stuck as it waited
4977 for a non-existent subprocess. Forks do occasionally fail when resources
4978 run out. I reviewed all the other calls to fork(); they all seem to check
4981 PH/20 When checking for unexpected SMTP input at connect time (before writing
4982 the banner), Exim was not dealing correctly with a non-positive return
4983 from the read() function. If the client had disconnected by this time,
4984 the result was a log entry for a synchronization error with an empty
4985 string after "input=" when read() returned zero. If read() returned -1
4986 (an event I could not check), uninitialized data bytes were printed.
4987 There were reports of junk text (parts of files, etc) appearing after
4990 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4992 PH/22 Added support for macro redefinition, and (re)definition in between
4993 driver and ACL definitions.
4995 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4996 forgetting to use the resulting value; it was using the unexpanded value.
4998 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4999 hadn't been configured. The fix is from Juergen Kreileder, who
5000 understands it better than I do:
5002 "Here's what I see happening with three configured cyrus_sasl
5003 authenticators configured (plain, login, cram-md5):
5005 On startup auth_cyrus_sasl_init() gets called for each of these.
5006 This means three calls to sasl_listmech() without a specified mech_list.
5007 => SASL tests which mechs of all available mechs actually work
5008 => three warnings about OTP not working
5009 => the returned list contains: plain, login, cram-md5, digest-md5, ...
5011 With the patch, sasl_listmech() also gets called three times. But now
5012 SASL's mech_list option is set to the server_mech specified in the the
5013 authenticator. Or in other words, the answer from sasl_listmech()
5014 gets limited to just the mech you're testing for (which is different
5016 => the return list contains just 'plain' or 'login', 'cram-md5' or
5017 nothing depending on the value of ob->server_mech.
5019 I've just tested the patch: Authentication still works fine,
5020 unavailable mechs specified in the exim configuration are still
5021 caught, and the auth.log warnings about OTP are gone."
5023 PH/25 When debugging is enabled, the contents of the command line are added
5024 to the debugging output, even when log_selector=+arguments is not
5027 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
5028 answer is "GNU", and only if the return is "GNU/something" is the answer
5031 PH/27 $acl_verify_message is now set immediately after the failure of a
5032 verification in an ACL, and so is available in subsequent modifiers. In
5033 particular, the message can be preserved by coding like this:
5035 warn !verify = sender
5036 set acl_m0 = $acl_verify_message
5038 Previously, $acl_verify_message was set only while expanding "message"
5039 and "log_message" when a very denied access.
5041 PH/28 Modified OS/os.c-Linux with
5043 -#ifndef OS_LOAD_AVERAGE
5044 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
5046 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
5047 nomenclature these days.)
5049 PH/29 Installed patch from the Sieve maintainer that adds the options
5050 sieve_useraddress and sieve_subaddress to the redirect router.
5052 PH/30 In these circumstances:
5053 . Two addresses routed to the same list of hosts;
5054 . First host does not offer TLS;
5055 . First host accepts first address;
5056 . First host gives temporary error to second address;
5057 . Second host offers TLS and a TLS session is established;
5058 . Second host accepts second address.
5059 Exim incorrectly logged both deliveries with the TLS parameters (cipher
5060 and peerdn, if requested) that were in fact used only for the second
5063 PH/31 When doing a callout as part of verifying an address, Exim was not paying
5064 attention to any local part prefix or suffix that was matched by the
5065 router that accepted the address. It now behaves in the same way as it
5066 does for delivery: the affixes are removed from the local part unless
5067 rcpt_include_affixes is set on the transport.
5069 PH/32 Add the sender address, as F=<...>, to the log line when logging a
5070 timeout during the DATA phase of an incoming message.
5072 PH/33 Sieve envelope tests were broken for match types other than :is. I have
5073 applied a patch sanctioned by the Sieve maintainer.
5075 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
5076 the uid or gid is negative. A case of a negative gid caused this to be
5077 noticed. The fix allows for either to be negative.
5079 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
5080 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
5083 PH/36 The change PH/12 above was broken. Fixed it.
5085 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
5086 the grounds that routing the same address twice would always produce the
5087 same answer. This might have been true once, but it is certainly no
5088 longer true now. Routing a child address may depend on the previous
5089 routing that produced that child. Some complicated redirection strategies
5090 went wrong when messages had multiple recipients, and made Exim's
5091 behaviour dependent on the order in which the addresses were given.
5093 I have moved the duplicate checking until after the routing is complete.
5094 Exim scans the addresses that are assigned to local and remote
5095 transports, and removes any duplicates. This means that more work will be
5096 done, as duplicates will always all be routed, but duplicates are
5097 presumably rare, so I don't expect this is of any significance.
5099 For deliveries to pipes, files, and autoreplies, the duplicate checking
5100 still happens during the routing process, since they are not going to be
5103 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
5104 It corrects a timeout issue with spamd. This is Ian's comment: "The
5105 background is that sometimes spamd either never reads data from a
5106 connection it has accepted, or it never writes response data. The exiscan
5107 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
5108 blindly assumes that writes won't block so it may never time out."
5110 PH/39 Allow G after quota size as well as K and M.
5112 PH/40 The value set for $authenticated_id in an authenticator may not contain
5113 binary zeroes or newlines because the value is written to log lines and
5114 to spool files. There was no check on this. Now the value is run through
5115 the string_printing() function so that such characters are converted to
5116 printable escape sequences.
5118 PH/41 $message_linecount is a new variable that contains the total number of
5119 lines in the message. Compare $body_linecount, which is the count for the
5122 PH/42 Exim no longer gives details of delivery errors for specific addresses in
5123 bounce and delay warning messages, except in certain special cases, which
5126 (a) An SMTP error message from a remote host;
5127 (b) A message specified in a :fail: redirection;
5128 (c) A message specified in a "fail" command in a system filter;
5129 (d) A message specified in a FAIL return from the queryprogram router;
5130 (e) A message specified by the cannot_route_message router option.
5132 In these cases only, Exim does include the error details in bounce and
5133 warning messages. There are also a few cases where bland messages such
5134 as "unrouteable address" or "local delivery error" are given.
5136 PH/43 $value is now also set for the "else" part of a ${run expansion.
5138 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
5139 being worked on, but at least Exim now implements the latest version to
5142 PH/45 In a pipe transport, although a timeout while waiting for the pipe
5143 process to complete was treated as a delivery failure, a timeout while
5144 writing the message to the pipe was logged, but erroneously treated as a
5145 successful delivery. Such timeouts include transport filter timeouts. For
5146 consistency with the overall process timeout, these timeouts are now
5147 treated as errors, giving rise to delivery failures by default. However,
5148 there is now a new Boolean option for the pipe transport called
5149 timeout_defer, which, if set TRUE, converts the failures into defers for
5150 both kinds of timeout. A transport filter timeout is now identified in
5153 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
5154 systems where "make" and "gmake" are different, calling "gmake" at top
5155 level broke things. I've arranged for the value of $(MAKE) to be passed
5156 from the Makefile to this script so that it can call the same version of
5160 A note about Exim versions 4.44 and 4.50
5161 ----------------------------------------
5163 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
5164 changes of various kinds. As a consequence, a big documentation update was
5165 needed. This delayed the release for rather longer than seemed good, especially
5166 in the light of a couple of (minor) security issues. Therefore, the changes
5167 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
5168 release. So 4.44 and 4.50 are in effect two different branches that both start
5171 I have left the 4.50 change log unchanged; it contains all the changes since
5172 4.43. The change log for 4.44 is below; many of its items are identical to
5173 those for 4.50. This seems to be the most sensible way to preserve the
5174 historical information.
5180 1. Minor wording change to the doc/README.SIEVE file.
5182 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
5183 computation of the current number of files was incorrect.
5185 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5186 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5189 4. Give more explanation in the error message when the command for a transport
5190 filter fails to execute.
5192 5. There are several places where Exim runs a non-Exim command in a
5193 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5194 was being done only for the command run by the queryprogram router. It is
5195 now done for all such subprocesses. The other cases are: ${run, transport
5196 filters, and the commands run by the lmtp and pipe transports.
5198 6. Added CONFIGURE_GROUP build-time option.
5200 7. Some older OS have a limit of 256 on the maximum number of file
5201 descriptors. Exim was using setrlimit() to set 1000 as a large value
5202 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5203 systems. I've change it so that if it can't get 1000, it tries for 256.
5205 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5206 was an oversight, and furthermore, ever since the addition of extra
5207 controls (e.g. 4.43/32), the checks on when to allow different forms of
5208 "control" were broken. There should now be diagnostics for all cases when a
5209 control that does not make sense is encountered.
5211 9. Added the /retain_sender option to "control=submission".
5213 10. $recipients is now available in the predata ACL (oversight).
5215 11. Tidy the search cache before the fork to do a delivery from a message
5216 received from the command line. Otherwise the child will trigger a lookup
5217 failure and thereby defer the delivery if it tries to use (for example) a
5218 cached ldap connection that the parent has called unbind on.
5220 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5221 of $address_data from the recipient verification was clobbered by the
5222 sender verification.
5224 13. The value of address_data from a sender verification is now available in
5225 $sender_address_data in subsequent conditions in the ACL statement.
5227 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
5229 15. Added a new option "connect=<time>" to callout options, to set a different
5232 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5233 was its contents. (It was OK if the option was not defined at all.)
5235 17. A "Completed" log line is now written for messages that are removed from
5236 the spool by the -Mrm option.
5238 18. New variables $sender_verify_failure and $recipient_verify_failure contain
5239 information about exactly what failed.
5241 19. Added -dd to debug only the daemon process.
5243 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
5244 handles timeouts, both on the server side and network timeouts. Renamed the
5245 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
5247 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
5248 It is now set to "smtps".
5250 22. $host_address is now set to the target address during the checking of
5251 ignore_target_hosts.
5253 23. When checking ignore_target_hosts for an ipliteral router, no host name was
5254 being passed; this would have caused $sender_host_name to have been used if
5255 matching the list had actually called for a host name (not very likely,
5256 since this list is usually IP addresses). A host name is now passed as
5259 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5260 code that specifies a non-restarting handler (typically sigaction() in
5261 modern systems) in an attempt to fix a rare and obscure crash bug.
5263 25. Narrowed the window for a race in the daemon that could cause it to ignore
5264 SIGCHLD signals. This is not a major problem, because they are used only to
5265 wake it up if nothing else does.
5267 26. A malformed maildirsize file could cause Exim to calculate negative values
5268 for the mailbox size or file count. Odd effects could occur as a result.
5269 The maildirsize information is now recalculated if the size or filecount
5272 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5273 support for a long time. Removed HAVE_SYS_VFS_H.
5275 28. Installed the latest version of exipick from John Jetmore.
5277 29. In an address list, if the pattern was not a regular expression, an empty
5278 subject address (from a bounce message) matched only if the pattern was an
5279 empty string. Non-empty patterns were not even tested. This was the wrong
5280 because it is perfectly reasonable to use an empty address as part of a
5281 database query. An empty address is now tested by patterns that are
5282 lookups. However, all the other forms of pattern expect the subject to
5283 contain a local part and a domain, and therefore, for them, an empty
5284 address still always fails if the pattern is not itself empty.
5286 30. Exim went into a mad DNS loop when attempting to do a callout where the
5287 host was specified on an smtp transport, and looking it up yielded more
5288 than one IP address.
5290 31. Re-factored the code for checking spool and log partition space into a
5291 function that finds that data and another that does the check. The former
5292 is then used to implement four new variables: $spool_space, $log_space,
5293 $spool_inodes, and $log_inodes.
5295 32. The RFC2047 encoding function was originally intended for short strings
5296 such as real names; it was not keeping to the 75-character limit for
5297 encoded words that the RFC imposes. It now respects the limit, and
5298 generates multiple encoded words if necessary. To be on the safe side, I
5299 have increased the buffer size for the ${rfc2047: expansion operator from
5302 33. It is now permitted to omit both strings after an "if" condition; if the
5303 condition is true, the result is "true". As before, when the second string
5304 is omitted, a false condition yields an empty string. This makes it less
5305 cumbersome to write custom ACL and router conditions.
5307 34. Failure to deliver a bounce message always caused it to be frozen, even if
5308 there was an errors_to setting on the router. The errors_to setting is now
5311 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5312 canonical form (fully expanded) before being placed in
5313 $sender_host_address.
5315 36. The table in the code that translates DNS record types into text (T_A to
5316 "A" for instance) was missing entries for NS and CNAME. It is just possible
5317 that this could have caused confusion if both these types were looked up
5318 for the same domain, because the text type is used as part of Exim's
5319 per-process caching. But the chance of anyone hitting this buglet seems
5322 37. The dnsdb lookup has been extended in a number of ways.
5324 (1) There is a new type, "zns", which walks up the domain tree until it
5325 finds some nameserver records. It should be used with care.
5327 (2) There is a new type, "mxh", which is like "mx" except that it returns
5328 just the host names, not the priorities.
5330 (3) It is now possible to give a list of domains (or IP addresses) to be
5331 looked up. The behaviour when one of the lookups defers can be
5332 controlled by a keyword.
5334 (4) It is now possible to specify the separator character for use when
5335 multiple records are returned.
5337 38. The dnslists ACL condition has been extended: it is now possible to supply
5338 a list of IP addresses and/or domains to be looked up in a particular DNS
5341 39. Added log_selector=+queue_time_overall.
5343 40. When running the queue in the test harness, wait just a tad after forking a
5344 delivery process, to get repeatability of debugging output.
5346 41. Include certificate and key file names in error message when GnuTLS fails
5347 to set them up, because the GnuTLS error message doesn't include the name
5348 of the failing file when there is a problem reading it.
5350 42. Allow both -bf and -bF in the same test run.
5352 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
5354 44. The "Exiscan patch" is now merged into the mainline Exim source.
5356 45. Sometimes the final signoff response after QUIT could fail to get
5357 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5358 < 0 before doing a fflush(). This bug looks as though it goes back to the
5359 introduction of TLS in release 3.20, but "sometimes" must have been rare
5360 because the tests only now provoked it.
5362 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
5363 (this can affect the format of dates).
5365 47. exim_tidydb, when checking for the continued existence of a message for
5366 which it has found a message-specific retry record, was not finding
5367 messages that were in split spool directories. Consequently, it was
5368 deleting retry records that should have stayed in existence.
5370 48. Steve fixed some bugs in eximstats.
5372 49. The SPA authentication driver was not abandoning authentication and moving
5373 on to the next authenticator when an expansion was forced to fail,
5374 contradicting the general specification for all authenticators. Instead it
5375 was generating a temporary error. It now behaves as specified.
5377 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
5378 (the order specifies the preference for clients). The order is now AES256,
5379 AES128, 3DES, ARCFOUR128.
5381 51. Small patch to Sieve code - explicitly set From: when generating an
5384 52. Exim crashed if a remote delivery caused a very long error message to be
5385 recorded - for instance if somebody sent an entire SpamAssassin report back
5386 as a large number of 550 error lines. This bug was coincidentally fixed by
5387 increasing the size of one of Exim's internal buffers (big_buffer) that
5388 happened as part of the Exiscan merge. However, to be on the safe side, I
5389 have made the code more robust (and fixed the comments that describe what
5392 53. Now that there can be additional text after "Completed" in log lines (if
5393 the queue_time_overall log selector is set), a one-byte patch to exigrep
5394 was needed to allow it to recognize "Completed" as not the last thing in
5397 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
5398 patch that reportedly fixes this has been added. I am not expert enough to
5399 create a test for it. This is what the patch creator wrote:
5401 "I found a little strange behaviour of ldap code when working with
5402 Windows 2003 AD Domain, where users was placed in more than one
5403 Organization Units. When I tried to give exim partial DN, the exit code
5404 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
5405 But simultaneously result of request was absolutely normal ldap result,
5406 so I produce this patch..."
5408 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
5409 REFERENCE, so I have modified the code to exclude the patch when that macro
5412 55. Some experimental protocols are using DNS PTR records for new purposes. The
5413 keys for these records are domain names, not reversed IP addresses. The
5414 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5415 leaves it alone. Component reversal etc. now happens only for IP addresses.
5418 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
5420 57. Double the size of the debug message buffer (to 2048) so that more of very
5421 long debug lines gets shown.
5423 58. The exicyclog utility now does better if the number of log files to keep
5424 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
5426 59. Two changes related to the smtp_active_hostname option:
5428 (1) $smtp_active_hostname is now available as a variable.
5429 (2) The default for smtp_banner uses $smtp_active_hostname instead
5430 of $primary_hostname.
5432 60. The host_aton() function is supposed to be passed a string that is known
5433 to be a valid IP address. However, in the case of IPv6 addresses, it was
5434 not checking this. This is a hostage to fortune. Exim now panics and dies
5435 if the condition is not met. A case was found where this could be provoked
5436 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5437 components; fortuitously, this particular loophole had already been fixed
5438 by change 4.50/55 above.
5440 If there are any other similar loopholes, the new check in host_aton()
5441 itself should stop them being exploited. The report I received stated that
5442 data on the command line could provoke the exploit when Exim was running as
5443 exim, but did not say which command line option was involved. All I could
5444 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5445 running as the user.
5448 61. There was a buffer overflow vulnerability in the SPA authentication code
5449 (which came originally from the Samba project). I have added a test to the
5450 spa_base64_to_bits() function which I hope fixes it.
5453 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
5454 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
5456 63. The daemon start-up calls getloadavg() while still root for those OS that
5457 need the first call to be done as root, but it missed one case: when
5458 deliver_queue_load_max is set with deliver_drop_privilege. This is
5459 necessary for the benefit of the queue runner, because there is no re-exec
5460 when deliver_drop_privilege is set.
5462 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
5463 This has been fixed.
5465 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5466 was in use, was not putting the data itself into the right store pool;
5467 consequently, it could be overwritten for a subsequent message in the same
5468 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5471 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
5473 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
5474 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
5475 the function were treating the return as a boolean value, which happened to
5476 work because 0=false and not-0=true, but is not correct code.
5478 68. The host_aton() function was not handling scoped IPv6 addresses (those
5479 with, for example, "%eth0" on the end) correctly.
5481 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
5482 negated items (that is, ~something) in unsigned ints. Some compilers
5483 apparently mutter when there is no cast.
5485 70. If an address verification called from an ACL failed, and did not produce a
5486 user-specific message (i.e. there was only a "system" message), nothing was
5487 put in $acl_verify_message. In this situation, it now puts the system
5490 71. Change 4.23/11 added synchronization checking at the start of an SMTP
5491 session; change 4.31/43 added the unwanted input to the log line - except
5492 that it did not do this in the start of session case. It now does.
5494 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
5495 This is wrong and can cause the other end to generate a synchronization
5496 error if it is another Exim or anything else that does the synchronization
5497 check. A QUIT command is no longer sent after a timeout.
5499 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
5500 during host lookups.
5502 74. The defer_ok option of callout verification was not working if it was used
5503 when verifying addresses in header lines, that is, for this case:
5505 verify = header_sender/callout=defer_ok
5507 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
5508 those file descriptors could be used for SMTP connections. If anything
5509 wrote to stderr (the example that came up was "warn" in embedded Perl), it
5510 could be sent to the SMTP client, causing chaos. The daemon now opens
5511 stdin, stdout, and stderr to /dev/null when it puts itself into the
5514 76. Arrange for output from Perl's "warn" command to be written to Exim's main
5515 log by default. The user can override this with suitable Perl magic.
5517 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
5518 the log message when discard triggers, was not working for the DATA ACL or
5519 for the non-SMTP ACL.
5521 78. Error message wording change in sieve.c.
5523 79. If smtp_accept_max_per_host was set, the number of connections could be
5524 restricted to fewer than expected, because the daemon was trying to set up
5525 a new connection before checking whether the processes handling previous
5526 connections had finished. The check for completed processes is now done
5527 earlier. On busy systems, this bug wouldn't be noticed because something
5528 else would have woken the daemon, and it would have reaped the completed
5531 80. If a message was submitted locally by a user whose login name contained one
5532 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
5533 It caused a spool format error. I have fixed the spool reading code. A
5534 related problem was that the "from" clause in the Received: line became
5535 illegal because of the space(s). It is now covered by ${quote_local_part.
5537 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
5540 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
5542 83. Patch from Sieve maintainer to fix unterminated string problem in
5543 "vacation" handling.
5545 84. Some minor changes to the Linux configuration files to help with other
5546 OS variants using glibc.
5548 85. One more patch for Sieve to update vacation handling to latest spec.
5551 ----------------------------------------------------
5552 See the note above about the 4.44 and 4.50 releases.
5553 ----------------------------------------------------
5559 1. Change 4.43/35 introduced a bug that caused file counts to be
5560 incorrectly computed when quota_filecount was set in an appendfile
5563 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5564 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5567 3. Give more explanation in the error message when the command for a transport
5568 filter fails to execute.
5570 4. There are several places where Exim runs a non-Exim command in a
5571 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5572 was being done only for the command run by the queryprogram router. It is
5573 now done for all such subprocesses. The other cases are: ${run, transport
5574 filters, and the commands run by the lmtp and pipe transports.
5576 5. Some older OS have a limit of 256 on the maximum number of file
5577 descriptors. Exim was using setrlimit() to set 1000 as a large value
5578 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5579 systems. I've change it so that if it can't get 1000, it tries for 256.
5581 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5582 was an oversight, and furthermore, ever since the addition of extra
5583 controls (e.g. 4.43/32), the checks on when to allow different forms of
5584 "control" were broken. There should now be diagnostics for all cases when a
5585 control that does not make sense is encountered.
5587 7. $recipients is now available in the predata ACL (oversight).
5589 8. Tidy the search cache before the fork to do a delivery from a message
5590 received from the command line. Otherwise the child will trigger a lookup
5591 failure and thereby defer the delivery if it tries to use (for example) a
5592 cached ldap connection that the parent has called unbind on.
5594 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5595 of $address_data from the recipient verification was clobbered by the
5596 sender verification.
5598 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5599 was its contents. (It was OK if the option was not defined at all.)
5601 11. A "Completed" log line is now written for messages that are removed from
5602 the spool by the -Mrm option.
5604 12. $host_address is now set to the target address during the checking of
5605 ignore_target_hosts.
5607 13. When checking ignore_target_hosts for an ipliteral router, no host name was
5608 being passed; this would have caused $sender_host_name to have been used if
5609 matching the list had actually called for a host name (not very likely,
5610 since this list is usually IP addresses). A host name is now passed as
5613 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5614 code that specifies a non-restarting handler (typically sigaction() in
5615 modern systems) in an attempt to fix a rare and obscure crash bug.
5617 15. Narrowed the window for a race in the daemon that could cause it to ignore
5618 SIGCHLD signals. This is not a major problem, because they are used only to
5619 wake it up if nothing else does.
5621 16. A malformed maildirsize file could cause Exim to calculate negative values
5622 for the mailbox size or file count. Odd effects could occur as a result.
5623 The maildirsize information is now recalculated if the size or filecount
5626 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5627 support for a long time. Removed HAVE_SYS_VFS_H.
5629 18. Updated exipick to current release from John Jetmore.
5631 19. Allow an empty sender to be matched against a lookup in an address list.
5632 Previously the only cases considered were a regular expression, or an
5635 20. Exim went into a mad DNS lookup loop when doing a callout where the
5636 host was specified on the transport, if the DNS lookup yielded more than
5639 21. The RFC2047 encoding function was originally intended for short strings
5640 such as real names; it was not keeping to the 75-character limit for
5641 encoded words that the RFC imposes. It now respects the limit, and
5642 generates multiple encoded words if necessary. To be on the safe side, I
5643 have increased the buffer size for the ${rfc2047: expansion operator from
5646 22. Failure to deliver a bounce message always caused it to be frozen, even if
5647 there was an errors_to setting on the router. The errors_to setting is now
5650 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5651 canonical form (fully expanded) before being placed in
5652 $sender_host_address.
5654 24. Updated eximstats to version 1.33
5656 25. Include certificate and key file names in error message when GnuTLS fails
5657 to set them up, because the GnuTLS error message doesn't include the name
5658 of the failing file when there is a problem reading it.
5660 26. Expand error message when OpenSSL has problems setting up cert/key files.
5663 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5664 (this can affect the format of dates).
5666 28. exim_tidydb, when checking for the continued existence of a message for
5667 which it has found a message-specific retry record, was not finding
5668 messages that were in split spool directories. Consequently, it was
5669 deleting retry records that should have stayed in existence.
5671 29. eximstats updated to version 1.35
5672 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5673 1.35 - bugfix such that pie charts by volume are generated correctly
5675 30. The SPA authentication driver was not abandoning authentication and moving
5676 on to the next authenticator when an expansion was forced to fail,
5677 contradicting the general specification for all authenticators. Instead it
5678 was generating a temporary error. It now behaves as specified.
5680 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5681 (the order specifies the preference for clients). The order is now AES256,
5682 AES128, 3DES, ARCFOUR128.
5684 31. Small patch to Sieve code - explicitly set From: when generating an
5687 32. Exim crashed if a remote delivery caused a very long error message to be
5688 recorded - for instance if somebody sent an entire SpamAssassin report back
5689 as a large number of 550 error lines. This bug was coincidentally fixed by
5690 increasing the size of one of Exim's internal buffers (big_buffer) that
5691 happened as part of the Exiscan merge. However, to be on the safe side, I
5692 have made the code more robust (and fixed the comments that describe what
5695 33. Some experimental protocols are using DNS PTR records for new purposes. The
5696 keys for these records are domain names, not reversed IP addresses. The
5697 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5698 leaves it alone. Component reversal etc. now happens only for IP addresses.
5701 34. The host_aton() function is supposed to be passed a string that is known
5702 to be a valid IP address. However, in the case of IPv6 addresses, it was
5703 not checking this. This is a hostage to fortune. Exim now panics and dies
5704 if the condition is not met. A case was found where this could be provoked
5705 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5706 components; fortuitously, this particular loophole had already been fixed
5707 by change 4.50/55 or 4.44/33 above.
5709 If there are any other similar loopholes, the new check in host_aton()
5710 itself should stop them being exploited. The report I received stated that
5711 data on the command line could provoke the exploit when Exim was running as
5712 exim, but did not say which command line option was involved. All I could
5713 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5714 running as the user.
5717 35. There was a buffer overflow vulnerability in the SPA authentication code
5718 (which came originally from the Samba project). I have added a test to the
5719 spa_base64_to_bits() function which I hope fixes it.
5722 36. The daemon start-up calls getloadavg() while still root for those OS that
5723 need the first call to be done as root, but it missed one case: when
5724 deliver_queue_load_max is set with deliver_drop_privilege. This is
5725 necessary for the benefit of the queue runner, because there is no re-exec
5726 when deliver_drop_privilege is set.
5728 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5729 was in use, was not putting the data itself into the right store pool;
5730 consequently, it could be overwritten for a subsequent message in the same
5731 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5734 38. Sometimes the final signoff response after QUIT could fail to get
5735 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5736 < 0 before doing a fflush(). This bug looks as though it goes back to the
5737 introduction of TLS in release 3.20, but "sometimes" must have been rare
5738 because the tests only now provoked it.
5744 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5745 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5746 Now it returns an integer. A number of calls were still expecting a T/F
5747 return. Fortuitously, in all cases, the tests worked in OK situations,
5748 which is the norm. However, things would have gone wrong on any write
5749 failures on the smtp file descriptor. This function is used when sending
5750 messages over SMTP and also when doing verify callouts.
5752 2. When Exim is called to do synchronous delivery of a locally submitted
5753 message (the -odf or -odi options), it no longer closes stderr before doing
5756 3. Implemented the mua_wrapper option.
5758 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5760 5. Implemented the functions header_remove(), header_testname(),
5761 header_add_at_position(), and receive_remove_recipient(), and exported them
5764 6. If an ACL "warn" statement specified the addition of headers, Exim already
5765 inserted X-ACL-Warn: at the start if there was no header name. However, it
5766 was not making this test for the second and subsequent header lines if
5767 there were newlines in the string. This meant that an invalid header could
5768 be inserted if Exim was badly configured.
5770 7. Allow an ACL "warn" statement to add header lines at the start or after all
5771 the Received: headers, as well as at the end.
5773 8. Added the rcpt_4xx retry error code.
5775 9. Added postmaster_mailfrom=xxx to callout verification option.
5777 10. Added mailfrom=xxxx to the callout verification option, for verify=
5780 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5781 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5783 12. Inserted some casts to stop certain compilers warning when using pointer
5784 differences as field lengths or precisions in printf-type calls (mostly
5785 affecting debugging statements).
5787 13. Added optional readline() support for -be (dynamically loaded).
5789 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5790 same clock tick as a message's arrival, so that its received time was the
5791 same as the "first fail" time on the retry record, and that message
5792 remained on the queue past the ultimate address timeout, every queue runner
5793 would try a delivery (because it was past the ultimate address timeout) but
5794 after another failure, the ultimate address timeout, which should have then
5795 bounced the address, did not kick in. This was a "< instead of <=" error;
5796 in most cases the first failure would have been in the next clock tick
5797 after the received time, and all would be well.
5799 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5800 being recognized when the domain list was tested by the match_domain
5801 condition in an expansion string.
5803 16. Added the ${str2b64: operator.
5805 17. Exim was always calling setrlimit() to set a large limit for the number of
5806 processes, without checking whether the existing limit was already
5807 adequate. (It did check for the limit on file descriptors.) Furthermore,
5808 errors from getrlimit() and setrlimit() were being ignored. Now they are
5809 logged to the main and panic logs, but Exim does carry on, to try to do its
5810 job under whatever limits there are.
5812 18. Imported PCRE 5.0.
5814 19. Trivial typo in log message " temporarily refused connection" (the leading
5817 20. If the log selector return_path_on_delivery was set and an address was
5818 redirected to /dev/null, the delivery process crashed because it assumed
5819 that a return path would always be set for a "successful" delivery. In this
5820 case, the whole delivery is bypassed as an optimization, and therefore no
5823 21. Internal re-arrangement: the function for sending a challenge and reading
5824 a response while authentication was assuming a zero-terminated challenge
5825 string. It's now changed to take a pointer and a length, to allow for
5826 binary data in such strings.
5828 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5830 23. Exim was not respecting finduser_retries when seeking the login of the
5831 uid under which it was called; it was always trying 10 times. (The default
5832 setting of finduser_retries is zero.) Also, it was sleeping after the final
5833 failure, which is pointless.
5835 24. Implemented tls_on_connect_ports.
5837 25. Implemented acl_smtp_predata.
5839 26. If the domain in control=submission is set empty, Exim assumes that the
5840 authenticated id is a complete email address when it generates From: or
5841 Sender: header lines.
5843 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5844 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5845 chown and chgrp in /bin and hostname in /usr/bin.
5847 28. Exim was keeping the "process log" file open after each use, just as it
5848 does for the main log. This opens the possibility of it remaining open for
5849 long periods when the USR1 signal hits a daemon. Occasional processlog
5850 errors were reported, that could have been caused by this. Anyway, it seems
5851 much more sensible not to leave this file open at all, so that is what now
5854 29. The long-running daemon process does not normally write to the log once it
5855 has entered its main loop, and it closes the log before doing so. This is
5856 so that log files can straightforwardly be renamed and moved. However,
5857 there are a couple of unusual error situations where the daemon does write
5858 log entries, and I had neglected to close the log afterwards.
5860 30. The text of an SMTP error response that was received during a remote
5861 delivery was being truncated at 512 bytes. This is too short for some of
5862 the long messages that one sometimes sees. I've increased the limit to
5865 31. It is now possible to make retry rules that apply only when a message has a
5866 specific sender, in particular, an empty sender.
5868 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5869 it possible to be selective about when SMTP synchronization is enforced.
5871 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5873 32. Implemented hosts_connection_nolog.
5875 33. Added an ACL for QUIT.
5877 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5880 35. Added mailbox_size and mailbox_filecount to appendfile.
5882 36. Added control = no_multiline_responses to ACLs.
5884 37. There was a bug in the logic of the code that waits for the clock to tick
5885 in the case where the clock went backwards by a substantial amount such
5886 that the microsecond fraction of "now" was more than the microsecond
5887 fraction of "then" (but the whole seconds number was less).
5889 38. Added support for the libradius Radius client library this is found on
5890 FreeBSD (previously only the radiusclient library was supported).
5896 1. When certain lookups returned multiple values in the form name=value, the
5897 quoting of the values was not always being done properly. Specifically:
5898 (a) If the value started with a double quote, but contained no whitespace,
5900 (b) If the value contained whitespace other than a space character (i.e.
5901 tabs or newlines or carriage returns) it was not quoted.
5902 This fix has been applied to the mysql and pgsql lookups by writing a
5903 separate quoting function and calling it from the lookup code. The fix
5904 should probably also be applied to nisplus, ibase and oracle lookups, but
5905 since I cannot test any of those, I have not disturbed their existing code.
5907 2. A hit in the callout cache for a specific address caused a log line with no
5908 reason for rejecting RCPT. Now it says "Previous (cached) callout
5909 verification failure".
5911 3. There was an off-by-one bug in the queryprogram router. An over-long
5912 return line was truncated at 256 instead of 255 characters, thereby
5913 overflowing its buffer with the terminating zero. As well as fixing this, I
5914 have increased the buffer size to 1024 (and made a note to document this).
5916 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5917 when Exim is waiting for an SMTP response from a remote server, Exim
5918 restarts its select() call on the socket, thereby resetting its timeout.
5919 This is not a problem when such interrupts are rare. Somebody set up a cron
5920 job to run exiwhat every 2 minutes, which is less than the normal select()
5921 timeout (5 or 10 minutes). This meant that the select() timeout never
5922 kicked in because it was always reset. I have fixed this by comparing the
5923 time when an interrupt arrives with the time at the start of the first call
5924 to select(). If more time than the timeout has elapsed, the interrupt is
5925 treated as a timeout.
5927 5. Some internal re-factoring in preparation for the addition of Sieve
5928 extensions (by MH). In particular, the "personal" test is moved to a
5929 separate function, and given an option for scanning Cc: and Bcc: (which is
5930 not set for Exim filters).
5932 6. When Exim created an email address using the login of the caller as the
5933 local part (e.g. when creating a From: or Sender: header line), it was not
5934 quoting the local part when it contained special characters such as @.
5936 7. Installed new OpenBSD configuration files.
5938 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5939 try to make them clearer.
5941 9. Callout options, other than the timeout value, were being ignored when
5942 verifying sender addresses in header lines. For example, when using
5944 verify = header_sender/callout=no_cache
5946 the cache was (incorrectly) being used.
5948 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5949 only the Cygwin environment.
5951 11. When return_path_on_delivery was set as a log selector, if different remote
5952 addresses in the same message used different return paths and parallel
5953 remote delivery occurred, the wrong values would sometimes be logged.
5954 (Whenever a remote delivery process finished, the return path value from
5955 the most recently started remote delivery process was logged.)
5957 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5958 lines when AUTH and/or TLS are in use. This is the "received protocol"
5959 field. Exim used to use "asmtp" for authenticated SMTP, without any
5960 indication (in the protocol name) for TLS use. Now it follows the RFC and
5961 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5962 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5963 names appear in log lines as well as in Received: header lines.
5965 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5966 extensions, and comparison tests, and to fix some bugs.
5968 14. Changes to the "personal" filter test:
5970 (1) The test was buggy in that it was just doing the equivalent of
5971 "contains" tests on header lines. For example, if a user's address was
5972 anne@some.where, the "personal" test would incorrectly be true for
5974 To: susanne@some.where
5976 This test is now done by extracting each address from the header in turn,
5977 and checking the entire address. Other tests that are part of "personal"
5978 are now done using regular expressions (for example, to check local parts
5979 of addresses in From: header lines).
5981 (2) The list of non-personal local parts in From: addresses has been
5982 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5983 taken from the Sieve specification recommendations.
5985 (3) If the message contains any header line starting with "List-" it is
5986 treated as non-personal.
5988 (4) The test for "circular" in the Subject: header line has been removed
5989 because it now seems ill-conceived.
5991 15. Minor typos in src/EDITME comments corrected.
5993 16. Installed latest exipick from John Jetmore.
5995 17. If headers_add on a router specified a text string that was too long for
5996 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5997 of string_sprintf() is now avoided.
5999 18. $message_body_size was not set (it was always zero) when running the DATA
6000 ACL and the local_scan() function.
6002 19. For the "mail" command in an Exim filter, no default was being set for
6003 the once_repeat time, causing a random time value to be used if "once" was
6004 specified. (If the value happened to be <= 0, no repeat happened.) The
6005 default is now 0s, meaning "never repeat". The "vacation" command was OK
6006 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
6007 (I found it when inspecting the code).
6009 20. There is now an overall timeout for performing a callout verification. It
6010 defaults to 4 times the callout timeout, which applies to individual SMTP
6011 commands during the callout. The overall timeout applies when there is more
6012 than one host that can be tried. The timeout is checked before trying the
6013 next host. This prevents very long delays if there are a large number of
6014 hosts and all are timing out (e.g. when the network connections are timing
6015 out). The value of the overall timeout can be changed by specifying an
6016 additional sub-option for "callout", called "maxwait". For example:
6018 verify = sender/callout=5s,maxwait=20s
6020 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
6021 to the end before writing, but this should make it even safer).
6023 22. Exim was forgetting that it had advertised PIPELINING for the second and
6024 subsequent messages on an SMTP connection. It was also not resetting its
6025 memory on STARTTLS and an internal HELO.
6027 23. When Exim logs an SMTP synchronization error within a session, it now
6028 records whether PIPELINING has been advertised or not.
6030 24. Added 3 instances of "(long int)" casts to time_t variables that were being
6031 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
6032 rather than long int.
6034 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
6036 26. Added the never_mail option to autoreply.
6042 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
6043 crash if the getsockname() call failed; this can happen if a connection is
6044 closed very soon after it is established. The problem was simply in the
6045 order in which certain operations were done, causing Exim to try to write
6046 to the SMTP stream before it had set up the file descriptor. The bug has
6047 been fixed by making things happen in the correct order.
6053 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
6054 before the connection was closed, thus losing the rejection response.
6056 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
6057 some early Solaris releases, but causes trouble in current releases where
6058 socklen_t is defined.
6060 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
6063 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
6066 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
6067 that releasing the top of it at the end releases what was used for sub-
6068 expansions (unless the block got too big). However, discard this block if
6069 the first thing is a variable or header, so that we can use its block when
6070 it is dynamic (useful for very large $message_headers, for example).
6072 6. Lookups now cache *every* query, not just the most recent. A new, separate
6073 store pool is used for this. It can be recovered when all lookup caches are
6074 flushed. Lookups now release memory at the end of their result strings.
6075 This has involved some general refactoring of the lookup sources.
6077 7. Some code has been added to the store_xxx() functions to reduce the amount
6078 of flapping under certain conditions.
6080 8. log_incoming_interface used to affect only the <= reception log lines. Now
6081 it causes the local interface and port to be added to several more SMTP log
6082 lines, for example "SMTP connection from", and rejection lines.
6084 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
6086 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
6088 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
6089 could be overwritten at the end of the current message (or the start of a
6090 new message if it was set in a HELO ACL). The value is now preserved for
6091 the duration of the SMTP connection.
6093 12. If a transport had a headers_rewrite setting, and a matching header line
6094 contained an unqualified address, that address was qualified, even if it
6095 did not match any rewriting rules. The underlying bug was that the values
6096 of the flags that permit the existence of unqualified sender and recipient
6097 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
6098 non-local messages, and by -bnq for local messages) were not being
6099 preserved with the message after it was received.
6101 13. When Exim was logging an SMTP synchronization error, it could sometimes log
6102 "next input=" as part of the text comprising the host identity instead of
6103 the correct text. The code was using the same buffer for two different
6104 strings. However, depending on which order the printing function evaluated
6105 its arguments, the bug did not always show up. Under Linux, for example, my
6106 test suite worked just fine.
6108 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
6109 doesn't work with some older versions of Perl. It has been changed to "my",
6110 which in any case is probably the better facility to use.
6112 15. A really picky compiler found some instances of statements for creating
6113 error messages that either had too many or two few arguments for the format
6116 16. The size of the buffer for calls to the DNS resolver has been increased
6117 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
6118 for addresses that have a lot of PTR records. This alleviates a problem; it
6119 does not fully solve it.
6121 17. A dnsdb lookup for PTR records that receives more data than will fit in the
6122 buffer now truncates the list and logs the incident, which is the same
6123 action as happens when Exim is looking up a host name and its aliases.
6124 Previously in this situation something unpredictable would happen;
6125 sometimes it was "internal error: store_reset failed".
6127 18. If a server dropped the connection unexpectedly when an Exim client was
6128 using GnuTLS and trying to read a response, the client delivery process
6129 crashed while trying to generate an error log message.
6131 19. If a "warn" verb in an ACL added multiple headers to a message in a single
6132 string, for example:
6134 warn message = H1: something\nH2: something
6136 the text was added as a single header line from Exim's point of view
6137 though it ended up OK in the delivered message. However, searching for the
6138 second and subsequent header lines using $h_h2: did not work. This has been
6139 fixed. Similarly, if a system filter added multiple headers in this way,
6140 the routers could not see them.
6142 20. Expanded the error message when iplsearch is called with an invalid key to
6143 suggest using net-iplsearch in a host list.
6145 21. When running tests using -bh, any delays imposed by "delay" modifiers in
6146 ACLs are no longer actually imposed (and a message to that effect is
6149 22. If a "gecos" field in a passwd entry contained escaped characters, in
6150 particular, if it contained a \" sequence, Exim got it wrong when building
6151 a From: or a Sender: header from that name. A second bug also caused
6152 incorrect handling when an unquoted " was present following a character
6153 that needed quoting.
6155 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
6156 was not being matched caselessly.
6158 24. Arranged for all hyphens in the exim.8 source to be escaped with
6161 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
6162 properly. Recipient callout cache records were still being keyed to include
6163 the sender, even when use_sender was set false. This led to far more
6164 callouts that were necessary. The sender is no longer included in the key
6165 when use_sender is false.
6167 26. Added "control = submission" modifier to ACLs.
6169 27. Added the ${base62d: operator to decode base 62 numbers.
6171 28. dnsdb lookups can now access SRV records.
6173 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
6174 the configuration file.
6176 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
6177 (-v) mode. This makes the output for a verbose queue run more intelligible.
6179 31. Added a use_postmaster feature to recipient callouts.
6181 32. Added the $body_zerocount variable, containing the number of binary zero
6182 bytes in the message body.
6184 33. The time of last modification of the "new" subdirectory is now used as the
6185 "mailbox time last read" when there is a quota error for a maildir
6188 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
6190 35. Added +ignore_unknown as a special item in host lists.
6192 36. Code for decoding IPv6 addresses in host lists is now included, even if
6193 IPv6 support is not being compiled. This fixes a bug in which an IPv6
6194 address was recognized as an IP address, but was then not correctly decoded
6195 into binary, causing unexpected and incorrect effects when compared with
6202 1. Very minor rewording of debugging text in manualroute to say "list of
6203 hosts" instead of "hostlist".
6205 2. If verify=header_syntax was set, and a header line with an unqualified
6206 address (no domain) and a large number of spaces between the end of the
6207 name and the colon was received, the reception process suffered a buffer
6208 overflow, and (when I tested it) crashed. This was caused by some obsolete
6209 code that should have been removed. The fix is to remove it!
6211 3. When running in the test harness, delay a bit after writing a bounce
6212 message to get a bit more predictability in the log output.
6214 4. Added a call to search_tidyup() just before forking a reception process. In
6215 theory, someone could use a lookup in the expansion of smtp_accept_max_
6216 per_host which, without the tidyup, could leave open a database connection.
6218 5. Added the variables $recipient_data and $sender_data which get set from a
6219 lookup success in an ACL "recipients" or "senders" condition, or a router
6220 "senders" option, similar to $domain_data and $local_part_data.
6222 6. Moved the writing of debug_print from before to after the "senders" test
6225 7. Change 4.31/66 (moving the time when the Received: is generated) caused
6226 problems for message scanning, either using a data ACL, or using
6227 local_scan() because the Received: header was not generated till after they
6228 were called (in order to set the time as the time of reception completion).
6229 I have revised the way this works. The header is now generated after the
6230 body is received, but before the ACL or local_scan() are called. After they
6231 are run, the timestamp in the header is updated.
6237 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
6238 before starting a queue runner without re-exec. This happened only when
6239 deliver_drop_privilege was set or when the Exim user was set to root. The
6240 effect of the bug was that timeouts during subsequent deliveries caused
6241 crashes instead of being properly handled. The handler is now left at its
6242 default (and expected) setting.
6244 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
6245 message, again when deliver_drop_privilege is set or Exim is run as root.
6246 The bug described in (1) was not present in this case, but the tidying up
6247 of the other signals was missing. I have made the two cases consistent.
6249 3. The ignore_target_hosts setting on a manualroute router was being ignored
6250 for hosts that were looked up using the /MX notation.
6252 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
6255 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
6256 operated on the sender address. After changing the $sender_address to <>
6257 for the sender address verify, Exim was re-instated it as the original
6258 (before rewriting) address, but remembering that it had rewritten it, so it
6259 wasn't rewriting it again. This bug also had the effect of breaking the
6260 sender address verification caching when the sender address was rewritten.
6262 6. The ignore_target_hosts option was being ignored by the ipliteral router.
6263 This has been changed so that if the ip literal address matches
6264 ignore_target_hosts, the router declines.
6266 7. Added expansion conditions match_domain, match_address, and match_local_
6267 part (NOT match_host).
6269 8. The placeholder for the Received: header didn't have a length field set.
6271 9. Added code to Exim itself and to exim_lock to test for a specific race
6272 condition that could lead to file corruption when using MBX delivery. The
6273 issue is with the lockfile that is created in /tmp. If this file is removed
6274 after a process has opened it but before that process has acquired a lock,
6275 there is the potential for a second process to recreate the file and also
6276 acquire a lock. This could lead to two Exim processes writing to the file
6277 at the same time. The added code performs the same test as UW imapd; it
6278 checks after acquiring the lock that its file descriptor still refers to
6279 the same named file.
6281 10. The buffer for building added header lines was of fixed size, 8192 bytes.
6282 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
6285 11. Added the smtp_active_hostname option. If used, this will typically be made
6286 to depend on the incoming interface address. Because $interface_address is
6287 not set up until the daemon has forked a reception process, error responses
6288 that can happen earlier (such as "too many connections") no longer contain
6291 12. If an expansion in a condition on a "warn" statement fails because a lookup
6292 defers, the "warn" statement is abandoned, and the next ACL statement is
6293 processed. Previously this caused the whole ACL to be aborted.
6295 13. Added the iplsearch lookup type.
6297 14. Added ident_timeout as a log selector.
6299 15. Added tls_certificate_verified as a log selector.
6301 16. Added a global option tls_require_ciphers (compare the smtp transport
6302 option of the same name). This controls incoming TLS connections.
6304 17. I finally figured out how to make tls_require_ciphers do a similar thing
6305 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
6306 before starting the TLS session.
6308 18. Tabs are now shown as \t in -bP output.
6310 19. If the log selector return_path_on_delivery was set, Exim crashed when
6311 bouncing a message because it had too many Received: header lines.
6313 20. If two routers both had headers_remove settings, and the first one included
6314 a superfluous trailing colon, the final name in the first list and the
6315 first name in the second list were incorrectly joined into one item (with a
6316 colon in the middle).
6322 1. Added -C and -D options to the exinext utility, mainly to make it easier
6323 to include in the automated testing, but these could be helpful when
6324 multiple configurations are in use.
6326 2. The exinext utility was not formatting the output nicely when there was
6327 an alternate port involved in the retry record key, nor when there was a
6328 message id as well (for retries that were specific to a specific message
6329 and a specific host). It was also confused by IPv6 addresses, because of
6330 the additional colons they contain. I have fixed the IPv4 problem, and
6331 patched it up to do a reasonable job for IPv6.
6333 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
6334 delivery, the log line now contains "pipelined" if PIPELINING was used.
6336 4. An SMTP transport process used to panic and die if the bind() call to set
6337 an explicit outgoing interface failed. This has been changed; it is now
6338 treated in the same way as a connect() failure.
6340 5. A reference to $sender_host_name in the part of a conditional expansion
6341 that was being skipped was still causing a DNS lookup. This no longer
6344 6. The def: expansion condition was not recognizing references to header lines
6345 that used bh_ and bheader_.
6347 7. Added the _cache feature to named lists.
6349 8. The code for checking quota_filecount in the appendfile transport was
6350 allowing one more file than it should have been.
6352 9. For compatibility with Sendmail, the command line option
6360 and sets the incoming protocol and host name (for trusted callers). The
6361 host name and its colon can be omitted when only the protocol is to be set.
6362 Note the Exim already has two private options, -pd and -ps, that refer to
6363 embedded Perl. It is therefore impossible to set a protocol value of "d" or
6364 "s", but I don't think that's a major issue.
6366 10. A number of refactoring changes to the code, none of which should affect
6369 (a) The number of logging options was getting close to filling up the
6370 32-bit word that was used as a bit map. I have split them into two classes:
6371 those that are passed in the argument to log_write(), and those that are
6372 only ever tested independently outside of that function. These are now in
6373 separate 32-bit words, so there is plenty of room for expansion again.
6374 There is no change in the user interface or the logging behaviour.
6376 (b) When building, for example, log lines, the code previously used a
6377 macro that called string_cat() twice, in order to add two strings. This is
6378 not really sufficiently general. Furthermore, there was one instance where
6379 it was actually wrong because one of the argument was used twice, and in
6380 one call a function was used. (As it happened, calling the function twice
6381 did not affect the overall behaviour.) The macro has been replaced by a
6382 function that can join an arbitrary number of extra strings onto a growing
6385 (c) The code for expansion conditions now uses a table and a binary chop
6386 instead of a serial search (which was left over from when there were very
6387 few conditions). Also, it now recognizes conditions like "pam" even when
6388 the relevant support is not compiled in: a suitably worded error message is
6389 given if an attempt is made to use such a condition.
6391 11. Added ${time_interval:xxxxx}.
6393 12. A bug was causing one of the ddress fields not to be passed back correctly
6394 from remote delivery subprocesses. The field in question was not being
6395 subsequently used, so this caused to problems in practice.
6397 13. Added new log selectors queue_time and deliver_time.
6399 14. Might have fixed a bug in maildirsizefile handling that threw up
6400 "unexpected character" debug warnings, and recalculated the data
6401 unnecessarily. In any case, I expanded the warning message to give more
6404 15. Added the message "Restricted characters in address" to the statements in
6405 the default ACL that block characters like @ and % in local parts.
6407 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
6408 Three changes have been made:
6410 (a) There was a serious bug; a negative response to MAIL caused the whole
6411 recipient domain to be cached as invalid, thereby blocking all messages
6412 to all local parts at the same domain, from all senders. This bug has
6413 been fixed. The domain is no longer cached after a negative response to
6414 MAIL if the sender used is not empty.
6416 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
6419 (c) A new callout option, "use_sender" has been added for people who want
6420 the modified behaviour.
6426 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
6429 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
6430 indeed breaks things for older releases.
6432 3. Added additional logging to the case where there is a problem reading data
6433 from a filter that is running in a subprocess using a pipe, in order to
6434 try to track down a specific problem.
6436 4. Testing facility fudge: when running in the test harness and attempting
6437 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
6438 getting "No route to host". Convert this to a timeout.
6440 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
6443 6. Some OS don't have socklen_t but use size_t instead. This affects the
6444 fifth argument of getsockopt() amongst other things. This is now
6445 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
6446 can be set for individual OS. I have set it for SunOS5, OSF1, and
6447 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
6448 some earlier ones do not.
6450 7. Change 4.30/15 was not doing the test caselessly.
6452 8. The standard form for an IPv6 address literal was being rejected by address
6453 parsing in, for example, MAIL and RCPT commands. An example of this kind of
6454 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6455 this, as well as the form without the "IPv6" on the front (but only when
6456 address literals are enabled, of course).
6458 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
6460 10. Exim crashed if a message with an empty sender address specified by -f
6461 encountered a router with an errors_to setting. This could be provoked only
6462 by a command such as
6466 where an empty string was supplied; "<>" did not hit this bug.
6468 11. Installed PCRE release 4.5.
6470 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
6471 remained set. It is now erased.
6473 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
6474 times from message ids (which are base 36 rather than the normal 62).
6476 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
6477 were being counted as actual protocol errors, and logged if the log
6478 selector +smtp_protocol_error was set. One cannot be perfect in this test,
6479 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
6480 and DATA following a set of rejected RCPTs do not count as protocol errors.
6481 In other words, Exim assumes they were pipelined, though this may not
6482 actually be the case. Of course, in all cases the client gets an
6483 appropriate error code.
6485 15. If a lookup fails in an ACL condition, a message about the failure may
6486 be available; it is used if testing the ACL cannot continue, because most
6487 such messages specify what the cause of the deferral is. However, some
6488 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
6489 that caused an old message to be retained and used if a later statement
6490 caused a defer, replacing the real cause of the deferral.
6492 16. If an IP address had so many PTR records that the DNS lookup buffer
6493 was not large enough to hold them, Exim could crash while trying to process
6494 the truncated data. It now detects and logs this case.
6496 17. Further to 4.21/58, another change has been made: if (and only if) the
6497 first line of a message (the first header line) ends with CRLF, a bare LF
6498 in a subsequent header line has a space inserted after it, so as not to
6499 terminate the header.
6501 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
6502 unnecessarily, used atoi() instead of strtol(), and didn't check the
6503 termination when getting file sizes from file names by regex.
6505 19. Completely re-implemented the support for maildirsize files, in the light
6506 of a number of problems with the previous contributed implementation
6507 (4.30/29). In particular:
6509 . If the quota is zero, the maildirsize file is maintained, but no quota is
6512 . If the maildir directory does not exist, it is created before any attempt
6513 to write a maildirsize file.
6515 . The quota value in the file is just a cache; if the quota is changed in
6516 the transport, the new value overrides.
6518 . A regular expression is available for excluding directories from the
6521 20. The autoreply transport checks the characters in options that define the
6522 message's headers; it allows continued headers, but it was checking with
6523 isspace() after an embedded newline instead of explicitly looking for a
6526 21. If all the "regular" hosts to which an address was routed had passed their
6527 expiry times, and had not reached their retry times, the address was
6528 bounced, even if fallback hosts were defined. Now Exim should go on to try
6531 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
6532 equivalent code in the SMTP transport. Some hosts send humungous responses
6533 to HELO/EHLO, more than 1024 it seems.
6535 23. Refactoring: code in filter.c used (void *) for "any old type" but this
6536 gives compiler warnings in some environments. I've now done it "properly",
6539 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
6540 (because of problems with the built-in one) was declared to return uschar *
6541 instead of char *, causing compiler failure.
6543 25. Fixed a file descriptor leak when processing alias/forward files.
6545 26. Fixed a minor format string issue in dbfn.c.
6547 27. Typo in exim.c: ("dmbnz" for "dbmnz").
6549 28. If a filter file refered to $h_xxx or $message_headers, and the headers
6550 contained RFC 2047 "words", Exim's memory could, under certain conditions,
6553 29. When a sender address is verified, it is cached, to save repeating the test
6554 when there is more than one recipient in a message. However, when the
6555 verification involves a callout, it is possible for different callout
6556 options to be set for different recipients. It is too complicated to keep
6557 track of this in the cache, so now Exim always runs a verification when a
6558 callout is required, relying on the callout cache for the optimization.
6559 The overhead is duplication of the address routing, but this should not be
6562 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
6563 to be verified with callout=postmaster, and the main callout worked but the
6564 postmaster check failed, the verification correctly failed. However, if a
6565 subsequent RCPT command asked for sender verification *without* the
6566 postmaster check, incorrect caching caused this verification also to fail,
6569 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
6570 it was not caching the DNS options (qualify_single, search_parents) that
6571 were used when the lookup failed. A subsequent lookup with different
6572 options therefore always gave the same answer, though there were cases
6573 where it should not have. (Example: a "domains = !$mx_any" option on a
6574 dnslookup router: the "domains" option is always processed without any
6575 widening, but the router might have qualify_single set.) Now Exim uses the
6576 cached value only when the same options are set.
6578 32. Added John Jetmore's "exipick" utility to the distribution.
6580 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
6581 than a timeout (e.g. a certificate is required, and is not provided), an
6582 Exim server now closes the connection immediately. Previously it waited for
6583 the client to close - but if the client is SSL, it seems that they each
6584 wait for each other, leading to a delay before one of them times out.
6586 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
6587 maintained 0.8.x compatibility because I don't think many are using it, and
6588 it is clearly obsolete.
6590 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
6593 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
6594 client certificate was expired. A simple patch fixes this, though I don't
6595 understand the full logic of why the verify callback is called multiple
6598 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
6599 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
6600 which causes problems with some clients (such as the Certicom SSL Plus
6601 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
6602 disables the coutermeasure allowing Eudora to connect."
6604 38. Exim was not checking that a write() to a log file succeeded. This could
6605 lead to Bad Things if a log got too big, in particular if it hit a file
6606 size limit. Exim now panics and dies if it cannot write to a log file, just
6607 as it does if it cannot open a log file.
6609 39. Modified OS/Makefile-Linux so that it now contains
6611 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
6613 The two -D definitions ensure that Exim is compiled with large file
6614 support, which makes it possible to handle log files that are bigger than
6617 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
6618 instance) a domain was checked against a named list that involved a lookup,
6619 causing $domain_data to be set, then another domain was checked against the
6620 same list, then the first domain was re-checked, the value of $domain_data
6621 after the final check could be wrong. In particular, if the second check
6622 failed, it could be set empty. This bug probably also applied to
6625 41. The strip_trailing_dot option was not being applied to the address given
6626 with the -f command-line option.
6628 42. The code for reading a message's header from the spool was incrementing
6629 $received_count, but never initializing it. This meant that the value was
6630 incorrect (doubled) while delivering a message in the same process in which
6631 it was received. In the most common configuration of Exim, this never
6632 happens - a fresh exec is done - but it can happen when
6633 deliver_drop_privilege is set.
6635 43. When Exim logs an SMTP synchronization error - client data sent too soon -
6636 it now includes up to 150 characters of the unexpected data in the log
6639 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
6640 and building data strings. The size of both of these buffers was 10 000
6641 bytes - far larger than anybody would *ever* want, thought I. Needless to
6642 say, somebody hit the limit. I have increased the maximum line length to
6643 20 000 and the maximum data length of concatenated lines to 100 000. I have
6644 also fixed two bugs, because there was no checking on these buffers. Tsk,
6645 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
6646 buffer is too small.
6648 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6649 lsearch lookups. Now it does.
6651 46. When parsing a route_list item in a manualroute router, a fixed-length
6652 buffer was used for the list of hosts. I made this 1024 bytes long,
6653 thinking that nobody would ever have a list of hosts that long. Wrong.
6654 Somebody had a whole pile of complicated expansion conditions, and the
6655 string was silently truncated, leading to an expansion error. It turns out
6656 that it is easier to change to an unlimited length (owing to other changes
6657 that have happened since this code was originally written) than to build
6658 structure for giving a limitation error. The length of the item that
6659 expands into the list of hosts is now unlimited.
6661 47. The lsearch lookup could not handle data where the length of text line was
6662 more than 4095 characters. Such lines were truncated, leading to shortened
6663 data being returned. It should now handle lines of any length.
6665 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6666 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6669 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6670 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6671 no longer happens in comment lines. A list of replacements is now placed
6672 at the head of all of the source files, except those whose only change is
6673 to replace PERL_COMMAND in the very first #! line.
6675 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6676 messages on the queue, with a bottom-up merge sort, using code contributed
6677 by Michael Haardt. This should make operations like -bp somewhat faster on
6678 large queues. It won't affect queue runners, except when queue_run_in_order
6681 51. Installed eximstats 1.31 in the distribution.
6683 52. Added support for SRV lookups to the dnslookup router.
6685 53. If an ACL referred to $message_body or $message_body_end, the value was not
6686 reset for any messages that followed in the same SMTP session.
6688 54. The store-handling optimization for building very long strings was not
6689 differentiating between the different store pools. I don't think this
6690 actually made any difference in practice, but I've tidied it.
6692 55. While running the routers to verify a sender address, $sender_address
6693 was still set to the sender address. This is wrong, because when routing to
6694 send a bounce to the sender, it would be empty. Therefore, I have changed
6695 it so that, while verifying a sender address, $sender_address is set to <>.
6696 (There is no change to what happens when verifying a recipient address.)
6698 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6699 target A or AAAA records (if not already returned) without resetting the
6700 qualify_single or search_parents options of the DNS resolver. These are
6701 inappropriate in this case because the targets of MX and SRV records must
6702 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6703 target that, when qualified, matched something in the local domain. These
6704 two options are now turned off when doing these lookups.
6706 57. It seems that at least some releases of Reiserfs (which does not have the
6707 concept of a fixed number of inodes) returns zero and not -1 for the
6708 number of available inodes. This interacted badly with check_spool_inodes,
6709 which assumed that -1 was the "no such thing" setting. What I have done is
6710 to check that the total number of inodes is greater than zero before doing
6711 the test of how many are available.
6713 58. When a "warn" ACL statement has a log_message modifier, the message is
6714 remembered, and not repeated. This is to avoid a lot of repetition when a
6715 message has many recipients that cause the same warning to be written.
6716 However, Exim was preserving the list of already written lines for an
6717 entire SMTP session, which doesn't seem right. The memory is now reset if a
6718 new message is started.
6720 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6721 debugging output unless log_rewrite was also set.
6723 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6724 of (int)(handle) when we know that handle contains (void *)(-1).
6726 61. The Exim daemon panic-logs an error return when it closes the incoming
6727 connection. However "connection reset by peer" seems to be common, and
6728 isn't really an error worthy of noting specially, so that particular error
6731 62. When Exim is trying to find all the local interfaces, it used to panic and
6732 die if the ioctl to get the interface flags failed. However, it seems that
6733 on at least one OS (Solaris 9) it is possible to have an interface that is
6734 included in the list of interfaces, but for which you get a failure error
6735 for this call. This happens when the interface is not "plumbed" into a
6736 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6737 failure of the "get flags" call assumes that the interface is down.
6739 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6740 makes life easier for people who are doing arithmetic on fields extracted
6741 from dates, where you often get leading zeros that should not be
6742 interpreted as octal.
6744 64. Added qualify_domain to the redirect router, to override the global
6747 65. If a pathologically long header line contained very many addresses (the
6748 report of this problem mentioned 10 000) and each of them was rewritten,
6749 Exim could use up a very large amount of memory. (It kept on making new
6750 copies of the header line as it rewrote, and never released the old ones.)
6751 At the expense of a bit more processing, the header rewriting function has
6752 been changed so that it no longer eats memory in this way.
6754 66. The generation of the Received: header has been moved from the time that a
6755 message starts to be received, to the time that it finishes. The timestamp
6756 in the Received: header should now be very close to that of the <= log
6757 line. There are two side-effects of this change:
6759 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6760 logged header lines no longer include the local Received: line, because
6761 it has not yet been created. The same applies to a copy of the message
6762 that is returned to a non-SMTP sender when a message is rejected.
6764 (b) When a filter file is tested using -bf, no additional Received: header
6765 is added to the test message. After some thought, I decided that this
6768 This change does not affect the value of $received_for. It is still set
6769 after address rewriting, but before local_scan() is called.
6771 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6773 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6774 gave an unhelpful panic error message, and a defer error. I have managed to
6775 change this behaviour so that it now rejects any supplied certificate,
6776 which seems right, as the list of acceptable certificates is empty.
6778 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6779 gave an unhelpful defer error. I have not managed to make this reject any
6780 supplied certificates, but the error message it gives is "no certificate
6781 supplied", which is not helpful.
6783 70. exigrep's output now also includes lines that are not associated with any
6784 message, but which match the given pattern. Implemented by a patch from
6785 Martin Sluka, which also tidied up the Perl a bit.
6787 71. Recipient callout verification, like sender verification, was using <> in
6788 the MAIL FROM command. This isn't really the right thing, since the actual
6789 sender may affect whether the remote host accepts the recipient or not. I
6790 have changed it to use the actual sender in the callout; this means that
6791 the cache record is now keyed on a recipient/sender pair, not just the
6792 recipient address. There doesn't seem to be a real danger of callout loops,
6793 since a callout by the remote host to check the sender would use <>.
6794 [SEE ABOVE: changed after hitting problems.]
6796 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6797 temporary errors. However, in the case of such a code being given after
6798 the end of a data transmission (i.e. after ".") Exim was failing to write
6799 a retry record for the message. (Yes, there was some broken host that was
6800 actually sending 8xx at this point.)
6802 73. An unknown lookup type in a host list could cause Exim to panic-die when
6803 the list was checked. (An example that provoked this was putting <; in the
6804 middle of a list instead of at the start.) If this happened during a DATA
6805 ACL check, a -D file could be left lying around. This kind of configuration
6806 error no longer causes Exim to die; instead it causes a defer error. The
6807 incident is still logged to the main and panic logs.
6809 74. Buglet left over from Exim 3 conversion. The message "too many messages
6810 in one connection" was written to the rejectlog but not the mainlog, except
6811 when address rewriting (yes!) was being logged.
6813 75. Added write_rejectlog option.
6815 76. When a system filter was run not as root (that is, when system_filter_user
6816 was set), the values of the $n variables were not being returned to the
6817 main process; thus, they were not subsequently available in the $sn
6820 77. Added +return_path_on_delivery log selector.
6822 78. A connection timeout was being treated differently from recipients deferred
6823 when testing hosts_max_try with a message that was older than the host's
6824 retry timeout. (The host should not be counted, thus allowing all hosts to
6825 be tried at least once before bouncing.) This may have been the cause of an
6826 occasionally reported bug whereby a message would remain on the queue
6827 longer than the retry timeout, but would be bounced if a delivery was
6828 forced. I say "may" because I never totally pinned down the problem;
6829 setting up timeout/retry tests is difficult. See also the next item.
6831 79. The ultimate address timeout was not being applied to errors that involved
6832 a combination of host plus message (for example, a timeout on a MAIL
6833 command). When an address resolved to a number of possible hosts, and they
6834 were not all tried for each delivery (e.g. because of hosts_max_try), a
6835 message could remain on the queue longer than the retry timeout.
6837 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6840 81. Fixed an obscure SMTP outgoing bug which required at least the following
6841 conditions: (a) there was another message waiting for the same server;
6842 (b) the server returned 5xx to all RCPT commands in the first message so
6843 that the message was not completed; (c) the server dropped the connection
6844 or gave a negative response to the RSET that Exim sends to abort the
6845 transaction. The observed case was a dropped connection after DATA that had
6846 been sent in pipelining mode. That is, the server had advertised PIPELINING
6847 but was not implementing it correctly. The effect of the bug was incorrect
6848 behaviour, such as trying another host, and this could lead to a crash.
6854 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6855 and daemon.c were passed as pointers to ints; they should have been
6856 pointers to socklen_t variables (which are typically unsigned ints).
6858 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6861 3. Fixed a really odd bug that affected only the testing scheme; patching a
6862 certain fixed string in the binary changed the value of another string that
6863 happened to be identical to the end of the original first string.
6865 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6866 name", it returns that address as the IP address. On some operating
6867 systems (e.g. Solaris), it also passes back the IP address string as the
6868 "host name". However, on others (e.g. Linux), it passes back an empty
6869 string. Exim wasn't checking for this, and was changing the host name to an
6870 empty string, assuming it had been canonicalized.
6872 5. Although rare, it is permitted to have more than one PTR record for a given
6873 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6874 all the names associated with an address, because they do in Solaris.
6875 However, it seems that they do not in Linux for data that comes from the
6876 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6877 I found this out when I moved to a new Linux workstation and tried to run
6878 the Exim test suite.
6880 To get round this problem I have changed the code so that it now does its
6881 own call to the DNS to look up PTR records when searching for a host name.
6882 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6883 addresses that are only in /etc/hosts are still found.
6885 This behaviour is, however, controlled by an option called host_lookup_
6886 order, which defaults to "bydns:byaddr". If people want to use the other
6887 order, or indeed, just use one or the other means of lookup, they can
6888 specify it in this variable.
6890 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6891 some operating systems, this comes back from gethostbyaddr() as an empty
6892 string, and this is what Exim used to test for. However, it seems that in
6893 other systems, "." is yielded. Exim now tests for this case too.
6895 7. The values of check_spool_space and check_log_space are now held internally
6896 as a number of kilobytes instead of an absolute number of bytes. If a
6897 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6898 kilobyte. This means that much larger values can be stored.
6900 8. Exim monitor: an attempt to get the action menu when not actually pointing
6901 at a message produces an empty menu entitled "No message selected". This
6902 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6903 no entries in it ("Shell widget menu has zero width and/or height"). So I
6904 have added a single, blank menu entry in this case.
6906 9. Added ${quote_local_part.
6908 10. MIME decoding is now applied to the contents of Subject: header lines when
6911 11. Now that a reference to $sender_host_address automatically causes a reverse
6912 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6913 host lookup before query-style lookups in lists that might use this
6914 variable. This has therefore been abolished, and the "net-" prefix is no
6915 longer necessary for query-style lookups.
6917 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6918 have been a typo for LFLAGS, so it has been changed.
6920 13. The install script calls Exim with "-C /dev/null" in order to find the
6921 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6922 to be output. However, since Exim outputs its version number before the
6923 error, it didn't break the script. It just looked ugly. I fixed this by
6924 always allowing "-C /dev/null" if the caller is root.
6926 14. Ignore overlarge ACL variable number when reading spool file - insurance
6927 against a later release with more variables having written the file.
6929 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6930 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6931 this, as well as the form without the "IPv6" on the front.
6933 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6934 OS/Makefile-Darwin file.
6936 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6937 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6939 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6940 variable. However, some OS use ints for inodes. I've added cast to long int
6941 to get rid of the compiler warning.
6943 19. I had forgotten to lock out "/../" in configuration file names when
6944 ALT_CONFIG_PREFIX was set.
6946 20. Routers used for verification do not need to specify transports. However,
6947 if such a router generated a host list, and callout was configured, Exim
6948 crashed, because it could not find a port number from the (non-existent)
6949 transport. It now assumes port 25 in this circumstance.
6951 21. Added the -t option to exigrep.
6953 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6954 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6955 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6956 them is set, LOOKUP_LSEARCH is forced.
6958 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6960 24. Added sender and host information to the "rejected by local_scan()" log
6961 line; previously there was no indication of these.
6963 25. Added .include_if_exists.
6965 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6966 Linux. It turns out that not all file systems support this. Apparently some
6967 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6968 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6969 not supported on the file descriptor, is now ignored when Exim is trying to
6970 sync a directory. This applies only to Linux.
6972 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6974 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6975 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6978 29. Added support for maildirsize files from supplied patch (modified a bit).
6980 30. The use of :fail: followed by an empty string could lead Exim to respond to
6981 sender verification failures with (e.g.):
6983 550 Verification failed for <xxx>
6984 550 Sender verify failed
6986 where the first response line was missing the '-' that indicates it is not
6987 the final line of the response.
6989 31. The loop for finding the name of the user that called Exim had a hardwired
6990 limit of 10; it now uses the value of finduser_retries, which is used for
6991 all other user lookups.
6993 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6996 33. Exim was neglecting to zero errno before one call of strtol() when
6997 expanding a string and expecting an integer value. On some systems this
6998 resulted in spurious "integer overflow" errors. Also, it was casting the
6999 result into an int without checking.
7001 34. Testing for a connection timeout using "timeout_connect" in the retry rules
7002 did not work. The code looks as if it has *never* worked, though it appears
7003 to have been documented since at least release 1.62. I have made it work.
7005 35. The "timeout_DNS" error in retry rules, also documented since at least
7006 1.62, also never worked. As it isn't clear exactly what this means, and
7007 clearly it isn't a major issue, I have abolished the feature by treating it
7008 as "timeout", and writing a warning to the main and panic logs.
7010 36. The display of retry rules for -brt wasn't always showing the error code
7013 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
7014 timeout_connect_A, timeout_connect_MX.
7016 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
7017 to the empty sender.
7019 39. The daemon was not analysing the content of -oX till after it had closed
7020 stderr and disconnected from the controlling terminal. This meant that any
7021 syntax errors were only noted on the panic log, and the return code from
7022 the command was 0. By re-arranging the code a little, I've made the
7023 decoding happen first, so such errors now appear on stderr, and the return
7024 code is 1. However, the actual setting up of the sockets still happens in
7025 the disconnected process, so errors there are still only recorded on the
7028 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
7029 connections (as happens on some IP stacks) was logged at start up time as
7030 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
7031 it from "IPv6 and IPv4", which means that two separate sockets are being
7034 41. The debug output for gethostbyname2() or getipnodebyname() failures now
7035 says whether AF_INET or AF_INET6 was passed as an argument.
7037 42. Exiwhat output was messed up when time zones were included in log
7040 43. Exiwhat now gives more information about the daemon's listening ports,
7041 and whether -tls-on-connect was used.
7043 44. The "port" option of the smtp transport is now expanded.
7045 45. A "message" modifier in a "warn" statement in a non-message ACL was being
7046 silently ignored. Now an error message is written to the main and panic
7049 46. There's a new ACL modifier called "logwrite" which writes to a log file
7050 as soon as it is encountered.
7052 47. Added $local_user_uid and $local_user_gid at routing time.
7054 48. Exim crashed when trying to verify a sender address that was being
7057 49. Exim was recognizing only a space character after ".include". It now also
7058 recognizes a tab character.
7060 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
7061 extracting the relevant information from the specification. The man page no
7062 longer contains scrambled data for the -d option, and I've added a section
7063 at the front about calling Exim under different names.
7065 51. Added "extra_headers" argument to the "mail" command in filter files.
7067 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
7070 53. Installed eximstats 1.29.
7072 54. Added transport_filter_timeout as a generic transport option.
7074 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
7075 Cc: header lines. This was required by RFC 822, but it not required by RFC
7078 56. Exim used to add From:, Date:, and Message-Id: header lines to any
7079 incoming messages that did not have them. Now it does so only if the
7080 message originates locally, that is, if there is no associated remote host
7081 address. When Resent- header lines are present, this applies to the Resent-
7082 lines rather than the non-Resent- lines.
7084 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
7085 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
7087 58. Messages for configuration errors now include the name of the main
7088 configuration file - useful now that there may be more than one file in a
7089 list (.included file names were always shown).
7091 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
7092 for those rare installations that do not start the daemon as root or run it
7093 setuid root. I've cut out the call to initgroups() if the daemon is not
7096 60. The Exim user and group can now be bound into the binary as text strings
7097 that are looked up at the start of Exim's processing.
7099 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
7101 62. Added $mailstore_basename variable.
7103 63. Installed patch to sieve.c from Michael Haardt.
7105 64. When Exim failed to open the panic log after failing to open the main log,
7106 the original message it was trying to log was written to stderr and debug
7107 output, but if they were not available (the usual case in production), it
7108 was lost. Now it is written to syslog before the two lines that record the
7109 failures to open the logs.
7111 65. Users' Exim filters run in subprocesses under the user's uid. It is
7112 possible for a "deliver" command or an alias in a "personal" command to
7113 provoke an address rewrite. If logging of address rewriting is configured,
7114 this fails because the process is not running as root or exim. There may be
7115 a better way of dealing with this, but for the moment (because 4.30 needs
7116 to be released), I have disabled address rewrite logging when running a
7117 filter in a non-root, non-exim process.
7123 1. The buildconfig auxiliary program wasn't quoting the value set for
7124 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
7125 not defined. This bug was masked in 4.22 by the effect that was fixed in
7128 2. Some messages that were rejected after a message id was allocated were
7129 shown as "incomplete" by exigrep. It no longer does this for messages that
7130 are rejected by local_scan() or the DATA or non-SMTP ACLs.
7132 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
7133 have allow_domain_literals set, the ID did not get logged in the <= line.
7134 Domain literals are now always recognized in Message-ID: header lines.
7136 4. The first argument for a ${extract expansion item is the key name or field
7137 number. Leading and trailing spaces in this item were not being ignored,
7138 causing some misleading effects.
7140 5. When deliver_drop_privilege was set, single queue runner processes started
7141 manually (i.e. by the command "exim -q") or by the daemon (which uses the
7142 same command in the process it spins off) were not dropping privilege.
7144 6. When the daemon running as "exim" started a queue runner, it always
7145 re-executed Exim in the spun-off process. This is a waste of effort when
7146 deliver_drop_privilege is set. The new process now just calls the
7147 queue-runner function directly.
7153 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
7156 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
7157 ignored. Though the use of -oP was forcing the writing of a pid file, it
7158 was always written to the default place.
7160 3. If the message "no IP address found for host xxxx" is generated during
7161 incoming verification, it is now followed by identification of the incoming
7162 connection (so you can more easily find what provoked it).
7164 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
7166 5. Added some features to "harden" Exim a bit more against certain attacks:
7168 (a) There is now a build-time option called FIXED_NEVER_USERS that can
7169 be put in Local/Makefile. This is like the never_users runtime option,
7170 but it cannot be overridden. The default setting is "root".
7172 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
7173 prefix string with which any file named in a -C command line option
7176 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
7177 is retained for -C and -D only if the caller of Exim is root. Without
7178 it, the exim user may also use -C and -D and retain privilege.
7180 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
7181 command line option is disabled.
7183 6. Macro names set by the -D option must start with an upper case letter, just
7184 like macro names defined in the configuration file.
7186 7. Added "dereference=" facility to LDAP.
7188 8. Two instances of the typo "uknown" in the source files are fixed.
7190 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
7191 the Configure-Makefile script screwed up while processing it.
7193 10. Incorporated PCRE 4.4.
7195 11. The SMTP synchronization check was not operating right at the start of an
7196 SMTP session. For example, it could not catch a HELO sent before the client
7197 waited for the greeting. There is now a check for outstanding input at the
7198 point when the greeting is written. Because of the duplex, asynchronous
7199 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
7200 way, but not yet received, when the check is performed.
7202 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
7203 on TCP/IP sockets, because this apparently causes some broken clients to
7206 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
7207 unchanged) from the Cygwin maintainer.
7209 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
7210 was supported instead of testing for mbx. Effectively a typo.
7212 15. The spa authenticator server code was not checking that the input it
7213 received was valid base64.
7215 16. The debug output line for the "set" modifier in ACLs was not showing the
7216 name of the variable that was being set.
7218 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
7220 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
7221 lookup on its own. Something else was needed to trigger the lookup. For
7222 example, a match in host_lookup or the need for a host name in a host list.
7223 Now, if $sender_host_name is referenced and the host name has not yet been
7224 looked up, a lookup is performed. If the lookup fails, the variable remains
7225 empty, and $host_lookup_failed is set to "1".
7227 19. Added "eqi" as a case-independent comparison operator.
7229 20. The saslauthd authentication condition could segfault if neither service
7230 nor realm was specified.
7232 21. If an overflowing value such as "2048M" was set for message_size_limit, the
7233 error message that was logged was misleading, and incoming SMTP
7234 connections were dropped. The message is now more accurate, and temporary
7235 errors are given to SMTP connections.
7237 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
7238 (except RSET) with a 421 error, until QUIT is received. However, it was
7239 failing to send a response to QUIT.
7241 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
7242 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
7243 ACL is now run after the helo_try_verify_hosts code.
7245 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
7246 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
7247 case-independent, so other case variants are also recognized). Apparently
7248 some systems use these upper case variants.
7250 25. If more than two messages were waiting for the same host, and a transport
7251 filter was specified for the transport, Exim sent two messages over the
7252 same TCP/IP connection, and then failed with "socket operation on non-
7253 socket" when it tried to send the third.
7255 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
7257 27. The extern definition of crypt16() in expand.c was not being excluded when
7258 the OS had its own crypt16() function.
7260 28. Added bounce_return_body as a new option, and bounce_return_size_limit
7261 as a preferred synonym for return_size_limit, both as an option and as an
7264 29. Added LIBS=-liconv to OS/Makefile-OSF1.
7266 30. Changed the default configuration ACL to relax the local part checking rule
7267 for addresses that are not in any local domains. For these addresses,
7268 slashes and pipe symbols are allowed within local parts, but the sequence
7269 /../ is explicitly forbidden.
7271 31. SPA server authentication was not clearing the challenge buffer before
7274 32. log_message in a "warn" ACL statement was writing to the reject log as
7275 well as to the main log, which contradicts the documentation and doesn't
7276 seem right (because no rejection is happening). So I have stopped it.
7278 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
7279 However, I am unable to do any testing of this.
7281 34. Fixed an infelicity in the appendfile transport. When checking directories
7282 for a mailbox, to see if any needed to be created, it was accidentally
7283 using path names with one or more superfluous leading slashes; tracing
7284 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
7286 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
7287 contents are added to the log line that is written for every discarded
7288 recipient. (Previously a log_message setting was ignored.)
7290 36. The ${quote: operator now quotes the string if it is empty.
7292 37. The install script runs exim in order to find its version number. If for
7293 some reason other than non-existence or emptiness, which it checks, it
7294 could not run './exim', it was installing it with an empty version number,
7295 i.e. as "exim-". This error state is now caught, and the installation is
7298 38. An argument was missing from the function that creates an error message
7299 when Exim fails to connect to the socket for saslauthd authentication.
7300 This could cause Exim to crash, or give a corrupted message.
7302 39. Added isip, isip4, and isip6 to ${if conditions.
7304 40. The ACL variables $acl_xx are now saved with the message, and can be
7305 accessed later in routers, transports, and filters.
7307 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
7308 strings in the file are not string-expanded.
7310 42. If a MAIL command specified a SIZE value that was too large to fit into an
7311 int variable, the check against message_size_limit failed. Such values are
7312 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
7313 day this will have to be increased, but I don't think I want to be around
7314 when emails are that large.
7321 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
7322 iconv() is not standard in FreeBSD.
7324 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
7325 IPv6 enabled. The observed symptom was a segmentation fault on return from
7326 the function os_common_find_running_interfaces() in src/os.c.
7328 3. In the check_special_case() function in daemon.c I had used "errno" as an
7329 argument name, which causes warnings on some systems. This was basically a
7330 typo, since it was named "eno" in the comments!
7332 4. The code that waits for the clock to tick (at a resolution of some fraction
7333 of a second) so as to ensure message-id uniqueness was always waiting for
7334 at least one whole tick, when it could have waited for less. [This is
7335 almost certainly not relevant at current processor speeds, where it is
7336 unlikely to ever wait at all. But we try to future-proof.]
7338 5. The function that sleeps for a time interval that includes fractions of a
7339 second contained a race. It did not block SIGALRM between setting the
7340 timer, and suspending (a couple of lines later). If the interval was short
7341 and the sigsuspend() was delayed until after it had expired, the suspension
7342 never ended. On busy systems this could lead to processes getting stuck for
7345 6. Some uncommon configurations may cause a lookup to happen in a queue runner
7346 process, before it forks any delivery processes. The open lookup caching
7347 mechanism meant that the open file or database connection was passed into
7348 the delivery process. The problem was that delivery processes always tidy
7349 up cached lookup data. This could cause a problem for the next delivery
7350 process started by the queue runner, because the external queue runner
7351 process does not know about the closure. So the next delivery process
7352 still has data in the lookup cache. In the case of a file lookup, there was
7353 no problem because closing a file descriptor in a subprocess doesn't affect
7354 the parent. However, if the lookup was caching a connection to a database,
7355 the connection was closed, and the second delivery process was likely to
7356 see errors such as "PGSQL: query failed: server closed the connection
7357 unexpectedly". The problem has been fixed by closing all cached lookups
7358 in a queue runner before running a delivery process.
7360 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
7361 seem to have the "const" qualifier which it has on other OS. I've
7364 8. Change 4.21/2 was too strict. It is only if there are two authenticators
7365 *of the same type* (client or server) with the same public name that an
7366 error should be diagnosed.
7368 9. When Exim looked up a host name for an IP address, but failed to find the
7369 original IP address when looking up the host name (a safety check), it
7370 output the message "<ip address> does not match any IP for NULL", which was
7371 confusing, to say the least. The bug was that the host name should have
7372 appeared instead of "NULL".
7374 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
7375 user that is built into the binary, and the -C or -D options is used, root
7376 privilege is dropped before the configuration file is read. In addition,
7377 logging is switched to stderr instead of the normal log files. If the
7378 configuration then re-defines the Exim user, the unprivileged environment
7379 is probably not what is expected, so Exim logs a panic warning message (but
7382 However, if deliver_drop_privilege is set, the unprivileged state may well
7383 be exactly what is intended, so the warning has been cut out in that case,
7384 and Exim is allowed to try to write to its normal log files.
7390 1. smtp_return_error_details was not giving details for temporary sender
7391 or receiver verification errors.
7393 2. Diagnose a configuration error if two authenticators have the same public
7396 3. Exim used not to create the message log file for a message until the first
7397 delivery attempt. This could be confusing when incoming messages were held
7398 for policy or load reasons. The message log file is now created at the time
7399 the message is received, and an initial "Received" line is written to it.
7401 4. The automatically generated man page for command line options had a minor
7402 bug that caused no ill effects; however, a more serious problem was that
7403 the procedure for building the man page automatically didn't always
7404 operate. Consequently, release 4.20 contains an out-of-date version. This
7405 shouldn't happen again.
7407 5. When building Exim with embedded Perl support, the script that builds the
7408 Makefile was calling 'perl' to find its compile-time parameters, ignoring
7409 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
7411 6. The freeze_tell option was not being used for messages that were frozen on
7412 arrival, either by an ACL or by local_scan().
7414 7. Added the smtp_incomplete_transaction log selector.
7416 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
7417 was accepting AUTH without a new EHLO.
7419 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
7420 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
7423 10. Logging of TCP/IP connections (when configured) now happens in the main
7424 daemon process instead of the child process, so that the TCP/IP connection
7425 count is more accurate (but it can never be perfect).
7427 11. The use of "drop" in a nested ACL was not being handled correctly in the
7428 outer ACL. Now, if condition failure induced by the nested "drop" causes
7429 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
7430 or "require"), the connection is dropped.
7432 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
7433 that yield "discard" can now be used with an "accept" or a "discard" verb,
7434 but an error is generated for any others (because I can't see a useful way
7435 to define what should happen).
7437 13. When an ACL is read dynamically from a file (or anywhere else), the lines
7438 are now processed in the same way as lines in the Exim configuration file.
7439 In particular, continuation lines are supported.
7441 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
7443 15. Added -ti meaning -t -i.
7445 16. Check for letters, digits, hyphens, and dots in the names of dnslist
7446 domains, and warn by logging if others are found.
7448 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
7449 returned from GIFCONF when Exim is trying to find the list of interfaces on
7450 a host. The code in os.c has been modified to copy each ifreq to an aligned
7451 structure in all cases.
7453 Also, in some cases, the returned ifreq's were being copied to a 'struct
7454 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
7455 means the last couple of bytes of an IPv6 address could be chopped if the
7456 ifreq contained only a normal sockaddr (14 bytes storage).
7458 18. Named domain lists were not supported in the hosts_treat_as_local option.
7459 An entry such as +xxxx was not recognized, and was treated as a literal
7462 19. Ensure that header lines added by a DATA ACL are included in the reject log
7463 if the ACL subsequently rejects the message.
7465 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
7466 MD5 (which is deprecated).
7468 21. When testing a filter file using -bf, Exim was writing a message when it
7469 took the sender from a "From " line in the message, but it was not doing so
7470 when it took $return_path from a Return-Path: header line. It now does.
7472 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
7473 with a valid header line field name (a series of printing characters
7474 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
7476 23. Changed "disc" in the source to "disk" to conform to the documentation and
7477 the book and for uniformity.
7479 24. Ignore Sendmail's -Ooption=value command line item.
7481 25. When execve() failed while trying to run a command in a pipe transport,
7482 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
7483 could be confused with a return value of 69 from the command itself. This
7484 has been changed to 127, the value the shell returns if it is asked to run
7485 a non-existent command. The wording for the related log line suggests a
7486 non-existent command as the problem.
7488 26. If received_header_text expands to an empty string, do not add a Received:
7489 header line to the message. (Well, it adds a token one on the spool, but
7490 marks it "old" so that it doesn't get used or transmitted.)
7492 27. Installed eximstats 1.28 (addition of -nt option).
7494 28. There was no check for failure on the call to getsockname() in the daemon
7495 code. This can fail if there is a shortage of resources on the system, with
7496 ENOMEM, for example. A temporary error is now given on failure.
7498 29. Contrary to the C standard, it seems that in some environments, the
7499 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
7500 program. Exim now does this explicitly; it affects the formatting of
7501 timestamps using strftime().
7503 30. If exiqsumm was given junk data, it threw up some uninitialized variable
7504 complaints. I've now initialized all the variables, to avoid this.
7506 32. Header lines added by a system filter were not being "seen" during
7507 transport-time rewrites.
7509 33. The info_callback() function passed to OpenSSL is set up with type void
7510 (*)(SSL *, int, int), as described somewhere. However, when calling the
7511 function (actually a macro) that sets it up, the type void(*)() is
7512 expected. I've put in a cast to prevent warnings from picky compilers.
7514 34. If a DNS black list lookup found a CNAME record, but there were no A
7515 records associated with the domain it pointed at, Exim crashed.
7517 35. If a DNS black list lookup returned more than one A record, Exim ignored
7518 all but the first. It now scans all returned addresses if a particular IP
7519 value is being sought. In this situation, the contents of the
7520 $dnslist_value variable are a list of all the addresses, separated by a
7523 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
7524 to accept a host name and all its aliases if the forward lookup for any of
7525 them yielded the IP address of the incoming connection. Now it accepts only
7526 those names whose forward lookup yields the correct IP address. Any other
7527 names are discarded. This closes a loophole whereby a rogue DNS
7528 administrator could create reverse DNS records to break through a
7529 wildcarded host restriction in an ACL.
7531 37. If a user filter or a system filter that ran in a subprocess used any of
7532 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
7533 the wrong values were passed to the pipe command ($thisaddress had the
7534 value of $0, $0 had the value of $1, etc). This bug was introduced by
7535 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
7537 38. Improved the line breaking for long SMTP error messages from ACLs.
7538 Previously, if there was no break point between 40 and 75 characters, Exim
7539 left the rest of the message alone. Two changes have been made: (a) I've
7540 reduced the minimum length to 35 characters; (b) if it can't find a break
7541 point between 35 and 75 characters, it looks ahead and uses the first one
7542 that it finds. This may give the occasional overlong line, but at least the
7543 remaining text gets split now.
7545 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
7546 file descriptors might be low, and that setting 1000 would always raise it.
7547 It turns out that in some environments, the limit is already over 1000 and
7548 that lowering it causes trouble. So now Exim takes care not to decrease it.
7550 40. When delivering a message, the value of $return_path is set to $sender_
7551 address at the start of routing (routers may change the value). By an
7552 oversight, this default was not being set up when an address was tested by
7553 -bt or -bv, which affected the outcome if any router or filter referred to
7556 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
7557 log only when "message" or "log_message" are set. However, if one of the
7558 conditions was an address verification, or a call to a nested ACL, the
7559 messages generated by the underlying test were being passed through. This
7560 no longer happens. The underlying message is available in $acl_verify_
7561 message for both "message" and "log_message" expansions, so it can be
7562 passed through if needed.
7564 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
7565 new expansion $bh_ to give the encoded byte string without charset
7566 translation. Translation happens only if iconv() is available; HAVE_ICONV
7567 indicates this at build time. HEADERS_CHARSET gives the charset to
7568 translate to; headers_charset can change it in the configuration, and
7569 "headers charset" can change it in an individual filter file.
7571 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
7572 that creates RFC 2047 encoded "words" labels them as that charset instead
7573 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
7574 expansion operator; (ii) when Exim creates a From: line for a local
7575 message; (iii) when a header line is rewritten to include a "phrase" part.
7577 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
7578 buggy, causing it to skip the first lines of messages whose message ID
7579 ended in 'D'. This would not have bitten before Exim release 4.14, because
7580 message IDs were unlikely to end in 'D' before then. The effect was to have
7581 incorrect size information for certain domains.
7583 45. #include "config.h" was missing at the start of the crypt16.c module. This
7584 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
7587 46. If there was a timeout during a "random" callout check, Exim treated it as
7588 a failure of the random address, and carried on sending RSET and the real
7589 address. If the delay was just some slowness somewhere, the response to the
7590 original RCPT would be taken as a response to RSET and so on, causing
7591 mayhem of various kinds.
7593 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
7594 when I implemented it. It didn't allow for the fact that some option values
7595 may legitimately be negative (e.g. size_addition), and it didn't even do
7596 the right test for positive values.
7598 48. Domain names in DNS records are case-independent. Exim always looks them up
7599 in lower case. Some resolvers return domain names in exactly the case they
7600 appear in the zone file, that is, they may contain uppercase letters. Not
7601 all resolvers do this - some return always lower case. Exim was treating a
7602 change of case by a resolver as a change of domain, similar to a widening
7603 of a domain abbreviation. This triggered its re-routing code and so it was
7604 trying to route what was effectively the same domain again. This normally
7605 caused routing to fail (because the router wouldn't handle the domain
7606 twice). Now Exim checks for this case specially, and just changes the
7607 casing of the domain that it ultimately uses when it transmits the message
7610 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
7613 50. If a filter generated a file delivery with a non-absolute name (possible if
7614 no home directory exists for the router), the forbid_file option was not
7617 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
7618 the existing equality matching.
7620 52. Exim was using ints instead of ino_t variables in some places where it was
7621 dealing with inode numbers.
7623 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
7624 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
7625 called TMPDIR, and if it finds it is different, it changes its value.
7627 54. The smtp_printf() function is now made available to local_scan() so
7628 additional output lines can be written before returning. There is also an
7629 smtp_fflush() function to enable the detection of a dropped connection.
7630 The variables smtp_input and smtp_batched_input are exported to
7633 55. Changed the default runtime configuration: the message "Unknown user"
7634 has been removed from the ACL, and instead placed on the localuser router,
7635 using the cannot_route_message feature. This means that any verification
7636 failures that generate their own messages won't get overridden. Similarly,
7637 the "Unrouteable address" message that was in the ACL for unverifiable
7638 relay addresses has also been removed.
7640 56. Added hosts_avoid_esmtp to the smtp transport.
7642 57. The exicyclog script was not checking for the esoteric option
7643 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
7644 will work only if exicyclog is run under the appropriate euid.
7646 58. Following a discussion on the list, the rules by which Exim recognises line
7647 endings on incoming messages have been changed. The -dropcr and drop_cr
7648 options are now no-ops, retained only for backwards compatibility. The
7649 following line terminators are recognized: LF CRLF CR. However, special
7650 processing applies to CR:
7652 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7653 nor a local message in the state where . is a terminator.
7655 (ii) If a bare CR is encountered in a header line, an extra space is added
7656 after the line terminator so as not to end the header. The reasoning
7657 behind this is that bare CRs in header lines are most likely either
7658 to be mistakes, or people trying to play silly games.
7660 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7661 was being incorrectly given as 18 bytes larger than it should have been.
7662 This is a VOB (very old bug).
7664 60. This may never have affected anything current, but just in case it has:
7665 When the local host is found other than at the start of a list of hosts,
7666 the local host, those with the same MX, and any that follow, are discarded.
7667 When the list in question was part of a longer list of hosts, the following
7668 hosts (not currently being processed) were also being discarded. This no
7669 longer happens. I'm not sure if this situation could ever has previously
7672 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7675 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7676 header. This is something that is recommended in a new Internet Draft, and
7677 is something that is documented as being done by Sendmail. There are two
7678 possible values. For messages generated by the autoreply transport, Exim
7681 Auto-Submitted: auto-replied
7683 whereas for all other generated messages (e.g. bounces) it adds
7685 Auto-Submitted: auto-generated
7687 63. The "personal" condition in filters now includes a test for the
7688 Auto-Submitted: header. If it contains the string "auto-" the message it
7689 not considered personal.
7691 64. Added rcpt_include_affixes as a generic transport option.
7693 65. Added queue_only_override (default true).
7695 66. Added the syslog_duplication option.
7697 67. If what should have been the first header line of a message consisted of
7698 a space followed by a colon, Exim was mis-interpreting it as a header line.
7699 It isn't of course - it is syntactically invalid and should therefore be
7700 treated as the start of the message body. The misbehaviour could have
7701 caused a number of strange effects, including loss of data in subsequent
7702 header lines, and spool format errors.
7704 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7705 client host had authenticated. This control can now be exercised by an ACL
7706 for more flexibility.
7708 69. By default, callouts do not happen when testing with -bh. There is now a
7709 variant, -bhc, which does actually run the callout code, including
7710 consulting and updating the callout cache.
7712 70. Added support for saslauthd authentication, courtesy of Alexander
7715 71. If statvfs() failed on the spool or log directories while checking their
7716 size for availability, Exim confusingly gave the error "space shortage".
7717 Furthermore, in debugging mode it crashed with a floating point exception.
7718 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7719 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7720 serious problem, Exim now writes to the main and panic logs when this
7721 happens, with details of the failure. It then refuses to accept the
7722 incoming message, giving the message "spool directory problem" or "log
7723 directory problem" with a 421 code for SMTP messages.
7725 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7726 0, 1, and 2 exist, because some OS complain for execs without them (see
7727 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7728 descriptors. However, the code omitted to check that the open succeeded,
7729 causing mysterious errors if for some reason the permissions on /dev/null
7730 got screwed. Now Exim writes a message to the main and panic logs, and
7731 bombs out if it can't open /dev/null.
7733 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7734 interact so that it is all more flexible. It is supposed to remain
7735 backwards compatible. Also added extra_local_interfaces.
7737 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7738 to bomb out with an assertion failure - to the client this appears as a
7739 connection drop. This problem occurs in the part of the code that was taken
7740 from the Samba project. Fortunately, the assertion is in a very simple
7741 function, so I have fixed this by reproducing the function inline in the
7742 one place where it is called, and arranging for authentication to fail
7743 instead of killing the process with assert().
7745 75. The SPA client code was not working when the server requested OEM rather
7746 than Unicode encoding.
7748 76. Added code to make require_files with a specific uid setting more usable in
7749 the case where statting the file as root fails - usually a non-root-mounted
7750 NFS file system. When this happens and the failure is EACCES, Exim now
7751 forks a subprocess and does the per-uid checking as the relevant uid.
7753 77. Added process_log_path.
7755 78. If log_file_path was not explicitly set, a setting of check_log_space or
7756 check_log_inodes was ignored.
7758 79. If a space check for the spool or log partitions fails, the incident is now
7759 logged. Of course, in the latter case the data may get lost...
7761 80. Added the %p formatting code to string_format() so that it can be used to
7762 print addresses in debug_print(). Adjusted all the address printing in the
7763 debugging in store.c to use %p rather than %d.
7765 81. There was a concern that a line of code in smtp_in.c could overflow a
7766 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7767 initially expressed, the concern was not well-founded, because trailing
7768 spaces are removed early. However, if the trailing spaces were followed by
7769 a NULL, they did not get removed, so the overflow was possible. Two fixes
7772 (a) I re-wrote the offending code in a cleaner fashion.
7773 (b) If an incoming SMTP command contains a NULL character, it is rejected
7776 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7777 runs initgroups(), so that if the Exim user is in any additional groups,
7778 they will be used during message reception.
7784 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / jgh / exim.git / blob