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 OpenSSL: With debug enabled output keying information sufficient, server
10 side, to decode a TLS 1.3 packet capture.
12 JH/02 OpenSSL: Suppress the sending of (stateful) TLS1.3 session tickets.
13 Previously the default library behaviour applied, sending two, each in
16 JH/03 Debug output for ACL now gives the config file name and line number for
19 JH/04 The default received_header_text now uses the RFC 8314 tls cipher clause.
21 JH/05 DKIM: ensure that dkim_domain elements are lowercased before use.
23 JH/06 Fix buggy handling of autoreply bounce_return_size_limit, and a possible
24 buffer overrun for (non-chunking) other transports.
26 JH/07 GnuTLS: Our use of late (post-handshake) certificate verification, under
27 TLS1.3, means that a server rejecting a client certificate is not visible
28 to the client until the first read of encrypted data (typically the
29 response to EHLO). Add detection for that case and treat it as a failed
30 TLS connection attempt, so that the normal retry-in-clear can work (if
33 JB/01 Bug 2375: fix expansions of 822 addresses having comments in local-part
34 and/or domain. Found and fixed by Jason Betts.
36 JH/08 Add hardening against SRV & TLSA lookups the hit CNAMEs (a nonvalid
37 configuration). If a CNAME target was not a wellformed name pattern, a
40 JH/09 Logging: Fix initial listening-on line for multiple ports for an IP when
41 the OS reports them interleaved with other addresses.
43 JH/10 OpenSSL: Fix aggregation of messages. Previously, when PIPELINING was
44 used both for input and for a verify callout, both encrypted, SMTP
45 responses being sent by the server could be lost. This resulted in
46 dropped connections and sometimes bounces generated by a peer sending
49 JH/11 Harden plaintext authenticator against a badly misconfigured client-send
50 string. Previously it was possible to cause undefined behaviour in a
51 library routine (usually a crash). Found by "zerons".
53 JH/12 Bug 2384: fix "-bP smtp_receive_timeout". Previously it returned no
56 JH/13 Bug 2386: Fix builds with Dane under LibreSSL 2.9.0 onward. Some old
57 API was removed, so update to use the newer ones.
59 JH/14 Bug 1891: Close the log file if receiving a non-smtp message, without
60 any timeout set, is taking a long time. Previously we would hang on to a
61 rotated logfile "forever" if the input was arriving with long gaps
62 (a previous attempt to fix addressed lack, for a long time, of initial
65 HS/01 Bug 2390: Use message_id for tempfile creation to avoid races in a
66 shared (NFS) environment. The length of the tempfile name is now
67 4 + 16 ("hdr.$message_exim_id") which might break on file
68 systems which restrict the file name length to lower values.
71 HS/02 Bug 2390: Use message_id for tempfile creation to avoid races in a
72 shared (NFS) environment.
74 HS/03 Bug 2392: exigrep does case sensitive *option* processing (as it
75 did for all versions <4.90). Notably -M, -m, --invert, -I may be
78 JH/15 Use unsigned when creating bitmasks in macros, to avoid build errors
79 on some platforms for bit 31.
81 JH/16 GnuTLS: rework ciphersuite strings under recent library versions. Thanks
82 to changes apparently associated with TLS1.3 handling some of the APIs
83 previously used were either nonfunctional or inappropriate. Strings
84 like TLS1.3:ECDHE_SECP256R1__RSA_PSS_RSAE_SHA256__AES_256_GCM__AEAD:256
85 and TLS1.2:ECDHE_SECP256R1__RSA_SHA256__AES_128_CBC__SHA256:128 replace
86 the previous TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256 .
87 This affects log line X= elements, the $tls_{in,out}_cipher variables,
88 and the use of specific cipher names in the encrypted= ACL condition.
90 JH/17 OpenSSL: the default openssl_options now disables ssl_v3.
92 JH/18 GnuTLS: fix $tls_out_ocsp under hosts_request_ocsp. Previously the
93 verification result was not updated unless hosts_require_ocsp applied.
95 JH/19 Bug 2398: fix listing of a named-queue. Previously, even with the option
96 queue_list_requires_admin set to false, non-admin users were denied the
99 JH/20 Bug 2389: fix server advertising of usable certificates, under GnuTLS in
100 directory-of-certs mode. Previously they were advertised despite the
103 JH/21 The smtp transport option "hosts_noproxy_tls" is now unset by default.
104 A single TCP connection by a client will now hold a TLS connection open
105 for multiple message deliveries, by default. Previoud the default was to
108 JH/22 The smtp transport option "hosts_try_dane" now enables all hosts by
109 default. If built with the facility, DANE will be used. The facility
110 SUPPORT_DANE is now enabled in the prototype build Makefile "EDITME".
112 JH/23 The build default is now for TLS to be included; the SUPPORT_TLS define
113 is replaced with DISABLE_TLS. Either USE_GNUTLS or (the new) USE_OPENSSL
114 must be defined and you must still, unless you define DISABLE_TLS, manage
115 the the include-dir and library-file requirements that go with that
116 choice. Non-TLS builds are still supported.
118 JH/24 Fix duplicated logging of peer name/address, on a transport connection-
121 JH/25 The smtp transport option "hosts_try_fastopen" now enables all hosts by
122 default. If the platform supports and has the facility enabled, it will
123 be requested on all coneections.
125 JH/26 The PIPE_CONNECT facility is promoted from experimental status and is now
126 controlled by the build-time option SUPPORT_PIPE_CONNECT.
128 PP/01 Unbreak heimdal_gssapi, broken in 4.92.
130 JH/27 Bug 2404: Use the main-section configuration option "dsn_from" for
131 success-DSN messages. Previously the From: header was always the default
132 one for these; the option was ignored.
134 JH/28 Fix the timeout on smtp response to apply to the whole response.
135 Previously it was reset for every read, so a teergrubing peer sending
136 single bytes within the time limit could extend the connection for a
137 long time. Credit to Qualsys Security Advisory Team for the discovery.
139 JH/29 Fix DSN Final-Recipient: field. Previously it was the post-routing
140 delivery address, which leaked information of the results of local
141 forwarding. Change to the original envelope recipient address, per
144 JH/30 Bug 2411: Fix DSN generation when RFC 3461 failure notification is
145 requested. Previously not bounce was generated and a log entry of
146 error ignored was made.
148 JH/31 Avoid re-expansion in ${sort } expansion. (CVE-2019-13917)
150 JH/32 Introduce a general tainting mechanism for values read from the input
151 channel, and values derived from them. Refuse to expand any tainted
152 values, to catch one form of exploit.
154 JH/33 Bug 2413: Fix dkim_strict option. Previously the expansion result
155 was unused and the unexpanded text used for the test. Found and
156 fixed by Ruben Jenster.
158 JH/34 Fix crash after TLS shutdown. When the TCP/SMTP channel was left open,
159 an attempt to use a TLS library read routine dereffed a nul pointer,
162 JH/35 Bug 2409: filter out-of-spec chars from callout response before using
163 them in our smtp response.
165 JH/36 Have the general router option retry_use_local_part default to true when
166 any of the restrictive preconditions are set (to anything). Previously it
167 was only for check_local user. The change removes one item of manual
168 configuration which is required for proper retries when a remote router
169 handles a subset of addresses for a domain.
171 JH/37 Appendfile: when evaluating quota use (non-quota_size_regex) take the file
172 link count into consideration.
174 HS/04 Fix handling of very log lines in -H files. If a -<key> <value> line
175 caused the extension of big_buffer, the following lines were ignored.
177 JH/38 Bug 1395: Teach the DNS negative-cache about TTL value from the SOA in
178 accordance with RFC 2308. Previously there was no expiry, so a longlived
179 receive process (eg. due to ACL delays) versus a short SOA value could
182 HS/05 Handle trailing backslash gracefully. (CVE-2019-15846)
184 JH/39 Promote DMARC support to mainline.
186 JH/40 Bug 2452: Add a References: header to DSNs.
188 JH/41 With GnuTLS 3.6.0 (and later) do not attempt to manage Diffie-Hellman
189 parameters. The relevant library call is documented as "Deprecated: This
190 function is unnecessary and discouraged on GnuTLS 3.6.0 or later. Since
191 3.6.0, DH parameters are negotiated following RFC7919."
193 HS/06 Change the default of dnssec_request_domains to "*"
195 JH/42 Bug 2545: Fix CHUNKING for all RCPT commands rejected. Previously we
196 carried on and emitted a BDAT command, even when PIPELINING was not
203 JH/01 Remove code calling the customisable local_scan function, unless a new
204 definition "HAVE_LOCAL_SCAN=yes" is present in the Local/Makefile.
206 JH/02 Bug 1007: Avoid doing logging from signal-handlers, as that can result in
207 non-signal-safe functions being used.
209 JH/03 Bug 2269: When presented with a received message having a stupidly large
210 number of DKIM-Signature headers, disable DKIM verification to avoid
211 a resource-consumption attack. The limit is set at twenty.
213 JH/04 Add variables $arc_domains, $arc_oldest_pass for ARC verify. Fix the
214 report of oldest_pass in ${authres } in consequence, and separate out
215 some descriptions of reasons for verification fail.
217 JH/05 Bug 2273: Cutthrough delivery left a window where the received messsage
218 files in the spool were present and unlocked. A queue-runner could spot
219 them, resulting in a duplicate delivery. Fix that by doing the unlock
220 after the unlink. Investigation by Tim Stewart. Take the opportunity to
221 add more error-checking on spoolfile handling while that code is being
224 PP/01 Refuse to open a spool data file (*-D) if it's a symlink.
225 No known attacks, no CVE, this is defensive hardening.
227 JH/06 Bug 2275: The MIME ACL unlocked the received message files early, and
228 a queue-runner could start a delivery while other operations were ongoing.
229 Cutthrough delivery was a common victim, resulting in duplicate delivery.
230 Found and investigated by Tim Stewart. Fix by using the open message data
231 file handle rather than opening another, and not locally closing it (which
232 releases a lock) for that case, while creating the temporary .eml format
233 file for the MIME ACL. Also applies to "regex" and "spam" ACL conditions.
235 JH/07 Bug 177: Make a random-recipient callout success visible in ACL, by setting
236 $sender_verify_failure/$recipient_verify_failure to "random".
238 JH/08 When generating a selfsigned cert, use serial number 1 since zero is not
241 JH/09 Bug 2274: Fix logging of cmdline args when starting in an unlinked cwd.
242 Previously this would segfault.
244 JH/10 Fix ARC signing for case when DKIM signing failed. Previously this would
247 JH/11 Bug 2264: Exim now only follows CNAME chains one step by default. We'd
248 like zero, since the resolver should be doing this for us, But we need one
249 as a CNAME but no MX presence gets the CNAME returned; we need to check
250 that doesn't point to an MX to declare it "no MX returned" rather than
251 "error, loop". A new main option is added so the older capability of
252 following some limited number of chain links is maintained.
254 JH/12 Add client-ip info to non-pass iprev ${authres } lines.
256 JH/13 For receent Openssl versions (1.1 onward) use modern generic protocol
257 methods. These should support TLS 1.3; they arrived with TLS 1.3 and the
258 now-deprecated earlier definitions used only specified the range up to TLS
259 1.2 (in the older-version library docs).
261 JH/14 Bug 2284: Fix DKIM signing for body lines starting with a pair of dots.
263 JH/15 Rework TLS client-side context management. Stop using a global, and
264 explicitly pass a context around. This enables future use of TLS for
265 connections to service-daemons (eg. malware scanning) while a client smtp
266 connection is using TLS; with cutthrough connections this is quite likely.
268 JH/16 Fix ARC verification to do AS checks in reverse order.
270 JH/17 Support a "tls" option on the ${readsocket } expansion item.
272 JH/18 Bug 2287: Fix the protocol name (eg utf8esmtp) for multiple messages
273 using the SMTPUTF8 option on their MAIL FROM commands, in one connection.
274 Previously the "utf8" would be re-prepended for every additional message.
276 JH/19 Reject MAIL FROM commands with SMTPUTF8 when the facility was not advertised.
277 Previously thery were accepted, resulting in issues when attempting to
278 forward messages to a non-supporting MTA.
280 PP/02 Let -n work with printing macros too, not just options.
282 JH/20 Bug 2296: Fix cutthrough for >1 address redirection. Previously only
283 one parent address was copied, and bogus data was used at delivery-logging
284 time. Either a crash (after delivery) or bogus log data could result.
285 Discovery and analysis by Tim Stewart.
287 PP/03 Make ${utf8clean:} expansion operator detect incomplete final character.
288 Previously if the string ended mid-character, we did not insert the
289 promised '?' replacement.
291 PP/04 Documentation: current string operators work on bytes, not codepoints.
293 JH/21 Change as many as possible of the global flags into one-bit bitfields; these
294 should pack well giving a smaller memory footprint so better caching and
295 therefore performance. Group the declarations where this can't be done so
296 that the byte-sized flag variables are not interspersed among pointer
297 variables, giving a better chance of good packing by the compiler.
299 JH/22 Bug 1896: Fix the envelope from for DMARC forensic reports to be possibly
300 non-null, to avoid issues with sites running BATV. Previously reports were
301 sent with an empty envelope sender so looked like bounces.
303 JH/23 Bug 2318: Fix the noerror command within filters. It wasn't working.
304 The ignore_error flag wasn't being returned from the filter subprocess so
305 was not set for later routers. Investigation and fix by Matthias Kurz.
307 JH/24 Bug 2310: Raise a msg:fail:internal event for each undelivered recipient,
308 and a msg:complete for the whole, when a message is manually removed using
309 -Mrm. Developement by Matthias Kurz, hacked on by JH.
311 JH/25 Avoid fixed-size buffers for pathnames in DB access. This required using
312 a "Gnu special" function, asprintf() in the DB utility binary builds; I
313 hope that is portable enough.
315 JH/26 Bug 2311: Fix DANE-TA verification under GnuTLS. Previously it was also
316 requiring a known-CA anchor certificate; make it now rely entirely on the
317 TLSA as an anchor. Checking the name on the leaf cert against the name
318 on the A-record for the host is still done for TA (but not for EE mode).
320 JH/27 Fix logging of proxy address. Previously, a pointless "PRX=[]:0" would be
321 included in delivery lines for non-proxied connections, when compiled with
322 SUPPORT_SOCKS and running with proxy logging enabled.
324 JH/28 Bug 2314: Fire msg:fail:delivery event even when error is being ignored.
325 Developement by Matthias Kurz, tweaked by JH. While in that bit of code,
326 move the existing event to fire before the normal logging of message
327 failure so that custom logging is bracketed by normal logging.
329 JH/29 Bug 2322: A "fail" command in a non-system filter (file) now fires the
330 msg:fail:internal event. Developement by Matthias Kurz.
332 JH/30 Bug 2329: Increase buffer size used for dns lookup from 2k, which was
333 far too small for todays use of crypto signatures stored there. Go all
334 the way to the max DNS message size of 64kB, even though this might be
335 overmuch for IOT constrained device use.
337 JH/31 Fix a bad use of a copy function, which could be used to pointlessly
338 copy a string over itself. The library routine is documented as not
339 supporting overlapping copies, and on MacOS it actually raised a SIGABRT.
341 JH/32 For main options check_spool_space and check_inode_space, where the
342 platform supports 64b integers, support more than the previous 2^31 kB
343 (i.e. more than 2 TB). Accept E, P and T multipliers in addition to
344 the previous G, M, k.
346 JH/33 Bug 2338: Fix the cyrus-sasl authenticator to fill in the
347 $authenticated_fail_id variable on authentication failure. Previously
350 JH/34 Increase RSA keysize of autogen selfsign cert from 1024 to 2048. RHEL 8.0
351 OpenSSL didn't want to use such a weak key. Do for GnuTLS also, and for
352 more-modern GnuTLS move from GNUTLS_SEC_PARAM_LOW to
353 GNUTLS_SEC_PARAM_MEDIUM.
355 JH/35 OpenSSL: fail the handshake when SNI processing hits a problem, server
356 side. Previously we would continue as if no SNI had been received.
358 JH/36 Harden the handling of string-lists. When a list consisted of a sole
359 "<" character, which should be a list-separator specification, we walked
360 off past the nul-terimation.
362 JH/37 Bug 2341: Send "message delayed" warning MDNs (restricted to external
363 causes) even when the retry time is not yet met. Previously they were
364 not, meaning that when (say) an account was over-quota and temp-rejecting,
365 and multiple senders' messages were queued, only one sender would get
366 notified on each configured delay_warning cycle.
368 JH/38 Bug 2351: Log failures to extract envelope addresses from message headers.
370 JH/39 OpenSSL: clear the error stack after an SSL_accept(). With anon-auth
371 cipher-suites, an error can be left on the stack even for a succeeding
372 accept; this results in impossible error messages when a later operation
375 AM/01 Bug 2359: GnuTLS: repeat lowlevel read and write operations while they
376 return error codes indicating retry. Under TLS1.3 this becomes required.
378 JH/40 Fix the feature-cache refresh for EXPERIMENTAL_PIPE_CONNECT. Previously
379 it only wrote the new authenticators, resulting in a lack of tracking of
380 peer changes of ESMTP extensions until the next cache flush.
382 JH/41 Fix the loop reading a message header line to check for integer overflow,
383 and more-often against header_maxsize. Previously a crafted message could
384 induce a crash of the recive process; now the message is cleanly rejected.
386 JH/42 Bug 2366: Fix the behaviour of the dkim_verify_signers option. It had
387 been totally disabled for all of 4.91. Discovery and fix by "Mad Alex".
393 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
394 When redis_servers is set to a list of > 1 element, and the Redis servers
395 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
396 case of MOVED into a DEFER case instead, thus moving the query onto the
397 next server in the list. For a cluster of N elements, all N servers must
398 be defined in redis_servers.
400 GF/02 Catch and remove uninitialized value warning in exiqsumm
401 Check for existence of @ARGV before looking at $ARGV[0]
403 JH/01 Replace the store_release() internal interface with store_newblock(),
404 which internalises the check required to safely use the old one, plus
405 the allocate and data copy operations duplicated in both (!) of the
406 extant use locations.
408 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
409 modifier. This matches the restriction on the commandline.
411 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
412 Previously only the last row was returned.
414 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
415 we assumed that tags in the header were well-formed, and parsed the
416 element content after inspecting only the first char of the tag.
417 Assumptions at that stage could crash the receive process on malformed
420 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
421 While running the DKIM ACL we operate on the Permanent memory pool so that
422 variables created with "set" persist to the DATA ACL. Also (at any time)
423 DNS lookups that fail create cache records using the Permanent pool. But
424 expansions release any allocations made on the current pool - so a dnsdb
425 lookup expansion done in the DKIM ACL releases the memory used for the
426 DNS negative-cache, and bad things result. Solution is to switch to the
427 Main pool for expansions.
428 While we're in that code, add checks on the DNS cache during store_reset,
429 active in the testsuite.
430 Problem spotted, and debugging aided, by Wolfgang Breyha.
432 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
433 When none of the hosts presented to a transport match an already-open
434 connection, close it and proceed with the list. Previously we would
435 queue the message. Spotted by Lena with Yahoo, probably involving
438 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
439 Previously a spurious "250 OK id=" response was appended to the proper
442 JH/08 The "support for" informational output now, which built with Content
443 Scanning support, has a line for the malware scanner interfaces compiled
444 in. Interface can be individually included or not at build time.
446 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
447 by the template makefile "src/EDITME". The "STREAM" support for an older
448 ClamAV interface method is removed.
450 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
451 rows affected is given instead).
453 JH/11 The runtime Berkeley DB library version is now additionally output by
454 "exim -d -bV". Previously only the compile-time version was shown.
456 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
457 SMTP connection. Previously, when one had more receipients than the
458 first, an abortive onward connection was made. Move to full support for
459 multiple onward connections in sequence, handling cutthrough connection
460 for all multi-message initiating connections.
462 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
463 routers. Previously, a multi-recipient message would fail to match the
464 onward-connection opened for the first recipient, and cause its closure.
466 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
467 a timeout on read on a GnuTLS initiating connection, resulting in the
468 initiating connection being dropped. This mattered most when the callout
469 was marked defer_ok. Fix to keep the two timeout-detection methods
472 JH/15 Relax results from ACL control request to enable cutthrough, in
473 unsupported situations, from error to silently (except under debug)
474 ignoring. This covers use with PRDR, frozen messages, queue-only and
477 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
479 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
480 metadata, resulting in a crash in free().
482 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
483 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
484 Broken also in d185889f4, with init system revamp.
486 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
487 Previously we abruptly closed the connection after reading a malware-
488 found indication; now we go on to read the "scan ok" response line,
491 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
492 ACL. Previously, a crash would result.
494 JH/19 Speed up macro lookups during configuration file read, by skipping non-
495 macro text after a replacement (previously it was only once per line) and
496 by skipping builtin macros when searching for an uppercase lead character.
498 JH/20 DANE support moved from Experimental to mainline. The Makefile control
499 for the build is renamed.
501 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
502 was allocated for every new TLS startup, meaning one per message. Fix
503 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
505 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
506 reported the original. Fix to report (as far as possible) the ACL
507 result replacing the original.
509 JH/23 Fix memory leak during multi-message connections using STARTTLS under
510 OpenSSL. Certificate information is loaded for every new TLS startup,
511 and the resources needed to be freed.
513 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
515 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
518 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
519 DATA response info to the (existing) per-recipient response info for
520 the "C=" log element. It can have useful tracking info from the
521 destination system. Patch from Simon Arlott.
523 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
524 length value. Previously this would segfault.
526 HS/02 Support Avast multiline protoocol, this allows passing flags to
527 newer versions of the scanner.
529 JH/28 Ensure that variables possibly set during message acceptance are marked
530 dead before release of memory in the daemon loop. This stops complaints
531 about them when the debug_store option is enabled. Discovered specifically
532 for sender_rate_period, but applies to a whole set of variables.
533 Do the same for the queue-runner and queue-list loops, for variables set
534 from spool message files. Do the same for the SMTP per-message loop, for
535 certain variables indirectly set in ACL operations.
537 JH/29 Bug 2250: Fix a longstanding bug in heavily-pipelined SMTP input (such
538 as a multi-recipient message from a mailinglist manager). The coding had
539 an arbitrary cutoff number of characters while checking for more input;
540 enforced by writing a NUL into the buffer. This corrupted long / fast
541 input. The problem was exposed more widely when more pipelineing of SMTP
542 responses was introduced, and one Exim system was feeding another.
543 The symptom is log complaints of SMTP syntax error (NUL chars) on the
544 receiving system, and refused recipients seen by the sending system
545 (propating to people being dropped from mailing lists).
546 Discovered and pinpointed by David Carter.
548 JH/30 The (EXPERIMENTAL_DMARC) variable $dmarc_ar_header is withdrawn, being
549 replaced by the ${authresults } expansion.
551 JH/31 Bug 2257: Fix pipe transport to not use a socket-only syscall.
553 HS/03 Set a handler for SIGTERM and call exit(3) if running as PID 1. This
554 allows proper process termination in container environments.
556 JH/32 Bug 2258: Fix spool_wireformat in combination with LMTP transport.
557 Previously the "final dot" had a newline after it; ensure it is CR,LF.
559 JH/33 SPF: remove support for the "spf" ACL condition outcome values "err_temp"
560 and "err_perm", deprecated since 4.83 when the RFC-defined words
561 "temperror" and "permerror" were introduced.
563 JH/34 Re-introduce enforcement of no cutthrough delivery on transports having
564 transport-filters or DKIM-signing. The restriction was lost in the
565 consolidation of verify-callout and delivery SMTP handling.
566 Extend the restriction to also cover ARC-signing.
568 JH/35 Cutthrough: for a final-dot response timeout (and nonunderstood responses)
569 in defer=pass mode supply a 450 to the initiator. Previously the message
572 PP/02 DANE: add dane_require_tls_ciphers SMTP Transport option; if unset,
573 tls_require_ciphers is used as before.
575 HS/03 Malware Avast: Better match the Avast multiline protocol. Add
576 "pass_unscanned". Only tmpfails from the scanner are written to
577 the paniclog, as they may require admin intervention (permission
578 denied, license issues). Other scanner errors (like decompression
579 bombs) do not cause a paniclog entry.
581 JH/36 Fix reinitialisation of DKIM logging variable between messages.
582 Previously it was possible to log spurious information in receive log
585 JH/37 Bug 2255: Revert the disable of the OpenSSL session caching. This
586 triggered odd behaviour from Outlook Express clients.
588 PP/03 Add util/renew-opendmarc-tlds.sh script for safe renewal of public
591 JH/38 DKIM: accept Ed25519 pubkeys in SubjectPublicKeyInfo-wrapped form,
592 since the IETF WG has not yet settled on that versus the original
593 "bare" representation.
595 JH/39 Fix syslog logging for syslog_timestamp=no and log_selector +millisec.
596 Previously the millisecond value corrupted the output.
597 Fix also for syslog_pid=no and log_selector +pid, for which the pid
598 corrupted the output.
604 JH/01 Rework error string handling in TLS interface so that the caller in
605 more cases is responsible for logging. This permits library-sourced
606 string to be attached to addresses during delivery, and collapses
607 pairs of long lines into single ones.
609 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
610 during configuration. Wildcards are allowed and expanded.
612 JH/02 Rework error string handling in DKIM to pass more info back to callers.
613 This permits better logging.
615 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
616 do not close it down and have the child transport start it up again on
617 the passed-on TCP connection. Instead, proxy the child (and any
618 subsequent ones) for TLS via a unix-domain socket channel. Logging is
619 affected: the continued delivery log lines do not have any DNSSEC, TLS
620 Certificate or OCSP information. TLS cipher information is still logged.
622 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
623 identical IP addresses on different listening ports. Will also affect
626 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
627 add noisy ifdef guards to special-case this sillyness.
628 Patch from Bernd Kuhls.
630 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
631 than 255 are no longer allowed.
633 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
634 Disable the session-cache too, which might reduce our load. Since we
635 currrectly use a new context for every connection, both as server and
636 client, there is no benefit for these.
637 GnuTLS appears to not support tickets server-side by default (we don't
638 call gnutls_session_ticket_enable_server()) but client side is enabled
639 by default on recent versions (3.1.3 +) unless the PFS priority string
642 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
643 <https://reproducible-builds.org/specs/source-date-epoch/>.
645 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
646 the check for any unsuccessful recipients did not notice the limit, and
647 erroneously found still-pending ones.
649 JH/08 Pipeline CHUNKING command and data together, on kernels that support
650 MSG_MORE. Only in-clear (not on TLS connections).
652 JH/09 Avoid using a temporary file during transport using dkim. Unless a
653 transport-filter is involved we can buffer the headers in memory for
654 creating the signature, and read the spool data file once for the
655 signature and again for transmission.
657 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
658 4.77 as the kernel support then wasn't solid, having issues in 64bit
659 mode. Now, it's been long enough. Add support for FreeBSD also.
661 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
662 case where the routing stage had gathered several addresses to send to
663 a host before calling the transport for the first, we previously failed
664 to close down TLS in the old transport process before passing the TCP
665 connection to the new process. The new one sent a STARTTLS command
666 which naturally failed, giving a failed delivery and bloating the retry
667 database. Investigation and fix prototype from Wolfgang Breyha.
669 JH/12 Fix check on SMTP command input synchronisation. Previously there were
670 false-negatives in the check that the sender had not preempted a response
671 or prompt from Exim (running as a server), due to that code's lack of
672 awareness of the SMTP input buffering.
674 PP/04 Add commandline_checks_require_admin option.
675 Exim drops privileges sanely, various checks such as -be aren't a
676 security problem, as long as you trust local users with access to their
677 own account. When invoked by services which pass untrusted data to
678 Exim, this might be an issue. Set this option in main configuration
679 AND make fixes to the calling application, such as using `--` to stop
682 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
683 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
684 responses to those, into a single TLS record each way (this usually means
685 a single packet). As a side issue, smtp_enforce_sync now works on TLS
688 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
689 affects you only if you're dancing at the edge of the param size limits.
690 If you are, and this message makes sense to you, then: raise the
691 configured limit or use OpenSSL 1.1. Nothing we can do for older
694 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
695 cmdline element to get the documented default one. Previously it was
698 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
699 are unsuable for non-message-specific reasons - eg. connection timeout,
700 banner-time rejection.
702 JH/16 Fix logging of delivery remote port, when specified by router, under
705 PP/06 Repair manualroute's ability to take options in any order, even if one
706 is the name of a transport.
709 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
711 JH/17 Change the list-building routines interface to use the expanding-string
712 triplet model, for better allocation and copying behaviour.
714 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
715 Previously it was constructed the first time a possibly-matching string
716 was met in the configuration file input during startup; now it is done
719 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
720 compatible one, to avoid the (poorly documented) possibility of a config
721 file in the working directory redirecting the DB files, possibly correpting
722 some existing file. CVE-2017-10140 assigned for BDB.
724 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
725 cache-hot. Previously, although the result was properly cached, the
726 initial verify call returned a defer.
728 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
729 the main verify for receipient in uncached-mode.
731 JH/22 Retire historical build files to an "unsupported" subdir. These are
732 defined as "ones for which we have no current evidence of testing".
734 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
735 if present. Previously it was ignored.
737 JH/24 Start using specified-initialisers in C structure init coding. This is
738 a C99 feature (it's 2017, so now considered safe).
740 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
741 if was a fixed-sized field and bitmask ops via macros; it is now more
744 PP/07 GitHub PR 56: Apply MariaDB build fix.
745 Patch provided by Jaroslav Škarvada.
747 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
748 during Coverity cleanups [4.87 JH/47]
749 Diagnosis and fix provided by Michael Fischer v. Mollard.
751 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
752 the right size to place the terminating semicolon on its own folded
753 line, the header hash was calculated to an incorrect value thanks to
754 the (relaxed) space the fold became.
756 HS/02 Fix Bug 2130: large writes from the transport subprocess were chunked
757 and confused the parent.
759 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
760 which could crash as a result. This could lead to undeliverable messages.
762 JH/28 Logging: "next input sent too soon" now shows where input was truncated
765 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
766 matters on fast-turnover and PID-randomising systems, which were getting
767 out-of-order delivery.
769 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
770 a possibly-overlapping copy. The symptom was that "Remote host closed
771 connection in response to HELO" was logged instead of the actual 4xx
774 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
775 Previously only that bufferd was discarded, resulting in SYMTP command
778 JH/32 DKIM: when a message has multiple signatures matching an identity given
779 in dkim_verify_signers, run the dkim acl once for each. Previously only
780 one run was done. Bug 2189.
782 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
783 "panic the current process" to "deliberately defer". The panic log is
784 still written with the problem list name; the mail and reject logs now
785 get a temp-reject line for the message that was being handled, saying
786 something like "domains check lookup or other defer". The SMTP 451
787 message is still "Temporary local problem".
789 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
790 A crafted sequence of BDAT commands could result in in-use memory beeing
791 freed. CVE-2017-16943.
793 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
794 from SMTP input. Previously it was always done; now only done for DATA
795 and not BDAT commands. CVE-2017-16944.
797 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
798 to the message (such as an overlong header line). Previously this was
799 not done and we did not exit BDAT mode. Followon from the previous item
800 though a different problem.
806 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
807 than -2003 did; needs libidn2 in addition to libidn.
809 JH/02 The path option on a pipe transport is now expanded before use.
811 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
812 Patch provided by "Björn", documentation fix added too.
814 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
815 missing a wire-to-host endian conversion.
817 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
818 close after a BDAT command line could be taken as a following command,
819 giving a synch failure. Fix by only checking for synch immediately
820 before acknowledging the chunk.
822 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
823 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
824 macro. Patches provided by Josh Soref.
826 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
827 Previously we did not; the RFC seems ambiguous and VRFY is not listed
828 by IANA as a service extension. However, John Klensin suggests that we
831 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
832 the dkim code may be unix-mode line endings rather than smtp wire-format
833 CRLF, so prepend a CR to any bare LF.
835 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
836 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
838 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
839 the first were themselves being wrongly included in the feed into dkim
840 processing; with most chunk sizes in use this resulted in an incorrect
841 body hash calculated value.
843 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
844 DKIM signature block, for verification. Although advised against by
845 standards it is specifically not ruled illegal.
847 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
849 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
850 missing a body hash (the bh= tag).
852 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
853 It seems that HAProxy sends the Proxy Protocol information in clear and
854 only then does a TLS startup, so do the same.
856 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
857 TCP connections (such as for Spamd) unless the daemon successfully set
858 Fast Open mode on its listening sockets. This fixes breakage seen on
859 too-old kernels or those not configured for Fast Open, at the cost of
860 requiring both directions being enabled for TFO, and TFO never being used
861 by non-daemon-related Exim processes.
863 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
864 endings, at least on the first header line. Try to canonify any that get
865 past that check, despite the cost.
867 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
868 now limited to an arbitrary five deep, while parsing addresses with the
869 strip_excess_angle_brackets option enabled.
871 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
872 instead leave the unprompted TLS handshake in socket buffer for the
873 TLS library to consume.
875 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
877 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
879 JH/16 Drop variables when they go out of scope. Memory management drops a whole
880 region in one operation, for speed, and this leaves assigned pointers
881 dangling. Add checks run only under the testsuite which checks all
882 variables at a store-reset and panics on a dangling pointer; add code
883 explicitly nulling out all the variables discovered. Fixes one known
884 bug: a transport crash, where a dangling pointer for $sending_ip_address
885 originally assigned in a verify callout, is re-used.
887 PP/06 Drop '.' from @INC in various Perl scripts.
889 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
891 PP/08 Reduce a number of compilation warnings under clang; building with
892 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
893 should be warning-free.
895 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
897 HS/01 Fix portability problems introduced by PP/08 for platforms where
898 realloc(NULL) is not equivalent to malloc() [SunOS et al].
900 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
901 chunk. This allows us to accept broken chunked messages. We need a more
902 general solution here.
904 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
905 already-broken messages in the queue.
907 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
909 JH/19 Fix reference counting bug in routing-generated-address tracking.
915 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
916 supports it and a size is available (ie. the sending peer gave us one).
918 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
919 years of being deprecated). The replacements are the ACLs
920 acl_smtp_mime and acl_not_smtp_mime.
922 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
923 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
924 or even an in-clear connection were permitted. Now, if the host lookup
925 was dnssec and dane was requested then the host is only used if the
926 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
927 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
928 if one fails this test.
929 This means that a poorly-configured remote DNS will make it incommunicado;
930 but it protects against a DNS-interception attack on it.
932 JH/04 Bug 1810: make continued-use of an open smtp transport connection
933 non-noisy when a race steals the message being considered.
935 JH/05 If main configuration option tls_certificate is unset, generate a
936 self-signed certificate for inbound TLS connections.
938 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
939 in rewrites and routers.
941 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
942 and logged a warning sing 4.83; now they are a configuration file error.
944 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
945 (lacking @domain). Apply the same qualification processing as RCPT.
947 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
949 JH/10 Support ${sha256:} applied to a string (as well as the previous
952 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
953 a cutthrough deliver is pending, as we always want to make a connection.
954 This also avoids re-routing the message when later placing the cutthrough
955 connection after a verify cache hit.
956 Do not update it with the verify result either.
958 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
959 when routing results in more than one destination address.
961 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
962 signing (which inhibits the cutthrough capability). Previously only
963 the presence of an option was tested; now an expansion evaluating as
964 empty is permissible (obviously it should depend only on data available
965 when the cutthrough connection is made).
967 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
968 the relevant preceding SMTP command did not note the pipelining mode.
970 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
971 Previously they were not counted.
973 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
974 as one having no matching records. Previously we deferred the message
975 that needed the lookup.
977 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
978 distinguished as "(=".
980 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
981 for missing MX records. Previously it only worked for missing A records.
983 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
985 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
986 after the data-go-ahead and data-ack. Patch from Jason Betts.
988 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
989 even for a "none" policy. Patch from Tony Meyer.
991 JH/22 Fix continued use of a connection for further deliveries. If a port was
992 specified by a router, it must also match for the delivery to be
995 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
996 When one of the recipients of a message was unsuitable for the connection
997 (has no matching addresses), we lost track of needing to mark it
998 deferred. As a result mail would be lost.
1000 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
1002 JH/25 Decoding ACL controls is now done using a binary search; the source code
1003 takes up less space and should be simpler to maintain. Merge the ACL
1004 condition decode tables also, with similar effect.
1006 JH/26 Fix problem with one_time used on a redirect router which returned the
1007 parent address unchanged. A retry would see the parent address marked as
1008 delivered, so not attempt the (identical) child. As a result mail would
1011 JH/27 Fix a possible security hole, wherein a process operating with the Exim
1012 UID can gain a root shell. Credit to http://www.halfdog.net/ for
1013 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
1016 JH/28 Enable {spool,log} filesystem space and inode checks as default.
1017 Main config options check_{log,spool}_{inodes,space} are now
1018 100 inodes, 10MB unless set otherwise in the configuration.
1020 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
1021 Previously it only applied to the main-section connection policy
1024 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
1026 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
1027 by me. Added RFC7919 DH primes as an alternative.
1029 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
1030 are not in the system include path.
1032 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
1033 GnuTLS, when a session startup failed (eg because the client disconnected)
1034 Exim did stdio operations after fclose. This was exposed by a recent
1035 change which nulled out the file handle after the fclose.
1037 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
1038 signed directly by the cert-signing cert, rather than an intermediate
1039 OCSP-signing cert. This is the model used by LetsEncrypt.
1041 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
1043 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
1044 an incoming connection.
1046 HS/02 Bug 1802: Do not half-close the connection after sending a request
1049 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
1050 fallback to "prime256v1".
1052 JH/34 SECURITY: Use proper copy of DATA command in error message.
1053 Could leak key material. Remotely exploitable. CVE-2016-9963.
1059 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
1060 and 3.4.4 - once the server is enabled to respond to an OCSP request
1061 it does even when not requested, resulting in a stapling non-aware
1062 client dropping the TLS connection.
1064 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
1065 support variable-length bit vectors. No functional change.
1067 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
1068 The I= interface field on outgoing lines is now after the H= remote
1069 host field, same as incoming lines. There is a separate
1070 outgoing_interface log selector which allows you to disable the
1073 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
1074 If not running log_selector +smtp_connection the mainlog would be held
1075 open indefinitely after a "too many connections" event, including to a
1076 deleted file after a log rotate. Leave the per net connection logging
1077 leaving it open for efficiency as that will be quickly detected by the
1078 check on the next write.
1080 HS/01 Bug 1671: Fix post transport crash.
1081 Processing the wait-<transport> messages could crash the delivery
1082 process if the message IDs didn't exist for some reason. When
1083 using 'split_spool_directory=yes' the construction of the spool
1084 file name failed already, exposing the same netto behaviour.
1086 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
1087 mime_regex ACL conditions.
1089 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
1090 to DSN fail messages (bounces): remote IP, remote greeting, remote response
1091 to HELO, local diagnostic string.
1093 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
1094 log line to debug. Even when configured with a tls authenticator many
1095 client connections are expected to not authenticate in this way, so
1096 an authenticate fail is not an error.
1098 HS/02 Add the Exim version string to the process info. This way exiwhat
1099 gives some more detail about the running daemon.
1101 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
1102 matter for fast-change records such as DNSBLs.
1104 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
1105 retry record, even if constant. There may be multiple transports with
1106 different interface settings and the retry behaviour needs to be kept
1109 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
1111 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
1113 JH/10 Bug 840: fix log_defer_output option of pipe transport
1115 JH/11 Bug 830: use same host for all RCPTS of a message, even under
1116 hosts_randomize. This matters a lot when combined with mua_wrapper.
1118 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
1119 ${quote_pgsql:<string>} operator.
1121 JH/13 Bug 1708: avoid misaligned access in cached lookup.
1123 JH/14 Change header file name for freeradius-client. Relevant if compiling
1124 with Radius support; from the Gentoo tree and checked under Fedora.
1126 JH/15 Bug 1712: Introduce $prdr_requested flag variable
1128 JH/16 Bug 1714: Permit an empty string as expansion result for transport
1129 option transport_filter, meaning no filtering.
1131 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
1133 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
1134 defaults to "*" (all hosts). The variable is now available when not built
1135 with TLS, default unset, mainly to enable keeping the testsuite sane.
1136 If a server certificate is not supplied (via tls_certificate) an error is
1137 logged, and clients will find TLS connections fail on startup. Presumably
1138 they will retry in-clear.
1139 Packagers of Exim are strongly encouraged to create a server certificate
1140 at installation time.
1142 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
1143 with the $config_file variable.
1145 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
1146 in transport context, after the attempt, and per-recipient. The latter type
1147 is per host attempted. The event data is the error message, and the errno
1148 information encodes the lookup type (A vs. MX) used for the (first) host,
1149 and the trailing two digits of the smtp 4xx response.
1151 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
1152 to write to mainlog (or rejectlog, paniclog) in the window between file
1153 creation and permissions/ownership being changed. Particularly affects
1154 installations where exicyclog is run as root, rather than exim user;
1155 result is that the running daemon panics and dies.
1157 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
1159 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
1160 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
1161 "pri" and "weight". Note that the previous implicit priority given by the
1162 list order is no longer honoured.
1164 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
1165 for DKIM processing.
1167 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
1168 by defining SUPPORT_SOCKS.
1170 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
1171 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
1172 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
1173 variables are renamed to proxy_{local,external}_{address,port}.
1175 JH/27 Move Internationalisation support from Experimental to mainline, enabled
1176 for a build by defining SUPPORT_I18N
1178 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
1179 of the query string, and make ${quote_redis:} do that quoting.
1181 JH/29 Move Events support from Experimental to mainline, enabled by default
1182 and removable for a build by defining DISABLE_EVENT.
1184 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
1186 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
1187 cached by the daemon.
1189 JH/32 Move Redis support from Experimental to mainline, enabled for a build
1190 by defining LOOKUP_REDIS. The libhiredis library is required.
1192 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
1193 keys are given for lookup.
1195 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
1196 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
1197 only supported when built with TLS support. The PolarSSL SHA routines
1198 are still used when the TLS library is too old for convenient support.
1200 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
1201 openssl_options), for security. OpenSSL forces this from version 1.1.0
1202 server-side so match that on older versions.
1204 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
1205 allocation for $value could be released as the expansion processing
1206 concluded, but leaving the global pointer active for it.
1208 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
1209 and to use the domains and local_parts ACL conditions.
1211 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
1212 incorrectly not doubled on cutthrough transmission, hence seen as a
1213 body-termination at the receiving system - resulting in truncated mails.
1214 Commonly the sender saw a TCP-level error, and retransmitted the message
1215 via the normal store-and-forward channel. This could result in duplicates
1216 received - but deduplicating mailstores were liable to retain only the
1217 initial truncated version.
1219 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
1221 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
1223 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
1224 we're in there, support oversigning also; bug 1309.
1226 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
1228 HS/04 Add support for keep_environment and add_environment options.
1230 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
1231 either intentional arithmetic overflow during PRNG, or testing config-
1234 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
1235 delivery resulted in actual delivery. Cancel cutthrough before DATA
1238 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
1239 rejects a recipient: pass the reject to the originator.
1241 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
1242 Many were false-positives and ignorable, but it's worth fixing the
1245 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
1246 for the new environment-manipulation done at startup. Move the routines
1247 from being local to tls.c to being global via the os.c file.
1249 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
1250 an extract embedded as result-arg for a map, the first arg for extract
1251 is unavailable so we cannot tell if this is a numbered or keyed
1252 extraction. Accept either.
1258 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
1261 JH/02 The smtp transport option "multi_domain" is now expanded.
1263 JH/03 The smtp transport now requests PRDR by default, if the server offers
1266 JH/04 Certificate name checking on server certificates, when exim is a client,
1267 is now done by default. The transport option tls_verify_cert_hostnames
1268 can be used to disable this per-host. The build option
1269 EXPERIMENTAL_CERTNAMES is withdrawn.
1271 JH/05 The value of the tls_verify_certificates smtp transport and main options
1272 default to the word "system" to access the system default CA bundle.
1273 For GnuTLS, only version 3.0.20 or later.
1275 JH/06 Verification of the server certificate for a TLS connection is now tried
1276 (but not required) by default. The verification status is now logged by
1277 default, for both outbound TLS and client-certificate supplying inbound
1280 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
1283 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
1284 Status Notification (bounce) messages are now MIME format per RFC 3464.
1285 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
1286 under the control of the dsn_advertise_hosts option, and routers may
1287 have a dsn_lasthop option.
1289 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
1290 default, modifiable by a malware= option. The list separator for
1291 the options can now be changed in the usual way. Bug 68.
1293 JH/10 The smtp_receive_timeout main option is now expanded before use.
1295 JH/11 The incoming_interface log option now also enables logging of the
1296 local interface on delivery outgoing connections.
1298 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
1299 if the interface and destination host and port all match.
1301 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
1304 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
1305 Patch from Andrew Lewis.
1307 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
1308 now supports optional time-restrictions, weighting, and priority
1309 modifiers per server. Patch originally by <rommer@active.by>.
1311 JH/16 The spamd_address main option now supports a mixed list of local
1312 and remote servers. Remote servers can be IPv6 addresses, and
1313 specify a port-range.
1315 JH/17 Bug 68: The spamd_address main option now supports an optional
1316 timeout value per server.
1318 JH/18 Bug 1581: Router and transport options headers_add/remove can
1319 now have the list separator specified.
1321 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
1324 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
1327 JH/21 Support for the A6 type of dns record is withdrawn.
1329 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
1330 rather than the verbs used.
1332 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
1333 from 255 to 1024 chars.
1335 JH/24 Verification callouts now attempt to use TLS by default.
1337 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
1338 are generic router options now. The defaults didn't change.
1340 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
1341 Original patch from Alexander Shikoff, worked over by JH.
1343 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
1344 files if ZCAT_COMMAND is not executable.
1346 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
1348 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
1350 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
1351 Normally benign, it bites when the pair was led to by a CNAME;
1352 modern usage is to not canonicalize the domain to a CNAME target
1353 (and we were inconsistent anyway for A-only vs AAAA+A).
1355 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
1357 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
1358 when evaluating $sender_host_dnssec.
1360 JH/31 Check the HELO verification lookup for DNSSEC, adding new
1361 $sender_helo_dnssec variable.
1363 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
1365 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
1367 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
1369 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
1370 documented as working, but never had. Support all but $spam_report.
1372 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
1373 added for tls authenticator.
1375 HS/03 Add perl_taintmode main config option
1381 TL/01 When running the test suite, the README says that variables such as
1382 no_msglog_check are global and can be placed anywhere in a specific
1383 test's script, however it was observed that placement needed to be near
1384 the beginning for it to behave that way. Changed the runtest perl
1385 script to read through the entire script once to detect and set these
1386 variables, reset to the beginning of the script, and then run through
1387 the script parsing/test process like normal.
1389 TL/02 The BSD's have an arc4random API. One of the functions to induce
1390 adding randomness was arc4random_stir(), but it has been removed in
1391 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
1392 function when detected.
1394 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
1395 cause callback expansion.
1397 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
1398 syntax errors in an expansion can be treated as a string instead of
1399 logging or causing an error, due to the internal use of bool_lax
1400 instead of bool when processing it.
1402 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
1403 server certificates when making smtp deliveries.
1405 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
1407 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
1409 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
1411 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
1412 Merged patch from Sebastian Wiedenroth.
1414 JH/05 Fix results-pipe from transport process. Several recipients, combined
1415 with certificate use, exposed issues where response data items split
1416 over buffer boundaries were not parsed properly. This eventually
1417 resulted in duplicates being sent. This issue only became common enough
1418 to notice due to the introduction of connection certificate information,
1419 the item size being so much larger. Found and fixed by Wolfgang Breyha.
1421 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
1422 size buffer was used, resulting in syntax errors when an expansion
1425 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1426 version 3.3.6 or later.
1428 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1429 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1430 both become "event_action", the variables become $event_name, $event_data
1431 and $event_defer_errno. There is a new variable $verify_mode, usable in
1432 routers, transports and related events. The tls:cert event is now also
1433 raised for inbound connections, if the main configuration event_action
1436 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1437 early OCSP support, but no stapling (appears to be less than 1.0.0).
1439 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1440 server certificate names available under the smtp transport option
1441 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1444 JH/10 Time-related extraction expansions from certificates now use the main
1445 option "timezone" setting for output formatting, and are consistent
1446 between OpenSSL and GnuTLS compilations. Bug 1541.
1448 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1449 encoded parameter in the incoming message. Bug 1558.
1451 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1452 include certificate info, eximon was claiming there were spoolfile
1455 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1457 JH/14 Log delivery-related information more consistently, using the sequence
1458 "H=<name> [<ip>]" wherever possible.
1460 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1461 are problematic for Debian distribution, omit them from the release
1464 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1466 JH/16 Fix string representation of time values on 64bit time_t architectures.
1469 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1470 output list separator was used.
1475 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1476 checkers that were complaining about end of non-void function with no
1479 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1480 This was a regression introduced in 4.83 by another bugfix.
1482 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1484 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1485 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1491 TF/01 Correctly close the server side of TLS when forking for delivery.
1493 When a message was received over SMTP with TLS, Exim failed to clear up
1494 the incoming connection properly after forking off the child process to
1495 deliver the message. In some situations the subsequent outgoing
1496 delivery connection happened to have the same fd number as the incoming
1497 connection previously had. Exim would try to use TLS and fail, logging
1498 a "Bad file descriptor" error.
1500 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1501 utilities have not been installed.
1503 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1504 temporary space as the ACL may create new global variables.
1506 TL/01 LDAP support uses per connection or global context settings, depending
1507 upon the detected version of the libraries at build time.
1509 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1510 to extract and use the src ip:port in logging and expansions as if it
1511 were a direct connection from the outside internet. PPv2 support was
1512 updated based on HAProxy spec change in May 2014.
1514 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1516 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1517 Properly escape header and check for NULL return.
1519 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1522 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1524 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1525 characters in header names, implemented as a verify condition.
1526 Contributed by Michael Fischer v. Mollard.
1528 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1529 results permerror and temperror. Previous values are deprecated but
1530 still accepted. In a future release, err_perm and err_temp will be
1531 completely removed, which will be a backward incompatibility if the
1532 ACL tests for either of these two old results. Patch contributed by
1533 user bes-internal on the mailing list.
1535 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1537 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1538 selectors, in both main and reject logs.
1540 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1543 JH/07 Add malware type "sock" for talking to simple daemon.
1545 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1547 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1548 routers/transports under cutthrough routing.
1550 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1551 numbers. Touch up "bool" conditional to keep the same definition.
1553 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1555 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1557 JH/12 Expand items in router/transport headers_add or headers_remove lists
1558 individually rather than the list as a whole. Bug 1452.
1560 Required for reasonable handling of multiple headers_ options when
1561 they may be empty; requires that headers_remove items with embedded
1562 colons must have them doubled (or the list-separator changed).
1564 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1565 view the policy declared in the DMARC record. Currently, $dmarc_status
1566 is a combined value of both the record presence and the result of the
1569 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1571 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1572 dnslookup router and the smtp transport (applying to the forward
1575 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1576 of ldap servers used for a specific lookup. Patch provided by Heiko
1579 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1580 New variable $lookup_dnssec_authenticated for observability.
1582 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1583 Patch submitted by Lars Timman.
1585 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1587 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1588 Requires trusted mode and valid format message id, aborts otherwise.
1589 Patch contributed by Heiko Schlichting.
1591 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1592 certextract with support for various fields. Bug 1358.
1594 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1595 is requested by default, modifiable by smtp transport option
1598 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1599 operate on certificate variables to give certificate fingerprints
1600 Also new ${sha256:cert_variable}.
1602 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1604 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1605 Christian Aistleitner.
1607 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1609 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1610 file. Patch from Wolfgang Breyha.
1612 JH/25 Expand the coverage of the delivery $host and $host_address to
1613 client authenticators run in verify callout. Bug 1476.
1615 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1616 align with daemon_smtp_ports. Bug 72.
1618 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1619 support and error reporting did not work properly.
1621 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1622 and is readable. Patch from Andrew Colin Kissa.
1624 TL/14 Enhance documentation of ${run expansion and how it parses the
1625 commandline after expansion, particularly in the case when an
1626 unquoted variable expansion results in an empty value.
1628 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1630 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1631 by renaming away from it. Addresses GH issue 15, reported by
1634 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1636 TL/15 SECURITY: prevent double expansion in math comparison functions
1637 (can expand unsanitized data). Not remotely exploitable.
1644 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1646 PP/02 Make -n do something, by making it not do something.
1647 When combined with -bP, the name of an option is not output.
1649 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1652 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1653 $sender_host_name and config options to manage this, and basic check
1656 PP/05 DSCP support for outbound connections and control modifier for inbound.
1658 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1659 (Only plugin which currently uses this is kerberos4, which nobody should
1660 be using, but we should make it available and other future plugins might
1661 conceivably use it, even though it would break NAT; stuff *should* be
1662 using channel bindings instead).
1664 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1665 name; added for Sendmail compatibility; requires admin caller.
1666 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1667 just ignore it); requires trusted caller.
1668 Also parse but ignore: -Ac -Am -X<logfile>
1671 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1673 TL/02 Add +smtp_confirmation as a default logging option.
1675 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1676 Patch by Magnus Holmgren from 2007-02-20.
1678 TL/04 Bugzilla 1281 - Spec typo.
1679 Bugzilla 1283 - Spec typo.
1680 Bugzilla 1290 - Spec grammar fixes.
1682 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1684 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1686 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1687 mailing list by by Dmitry Isaikin.
1689 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1691 JH/02 Support "G" suffix to numbers in ${if comparisons.
1693 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1695 NM/01 Bugzilla 1197 - Spec typo
1696 Bugzilla 1196 - Spec examples corrections
1698 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1700 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1701 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1704 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1705 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1707 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1708 "acl {{name}{arg}...}", and optional args on acl condition
1711 JH/05 Permit multiple router/transport headers_add/remove lines.
1713 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1715 JH/07 Avoid using a waiting database for a single-message-only transport.
1716 Performance patch from Paul Fisher. Bugzilla 1262.
1718 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1721 JH/09 Add $headers_added variable, with content from use of ACL modifier
1722 add_header (but not yet added to the message). Bugzilla 199.
1724 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1725 Pulled from Bugzilla 817 by Wolfgang Breyha.
1727 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1729 (nb: this is the same fix as in Exim 4.80.1)
1731 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1734 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1737 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1739 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1740 advertise SMTP AUTH mechanism to us, instead of a generic
1741 protocol violation error. Also, make Exim more robust to bad
1742 data from the Dovecot auth socket.
1744 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1746 When a queue runner is handling a message, Exim first routes the
1747 recipient addresses, during which it prunes them based on the retry
1748 hints database. After that it attempts to deliver the message to
1749 any remaining recipients. It then updates the hints database using
1752 So if a recipient address works intermittently, it can get repeatedly
1753 deferred at routing time. The retry hints record remains fresh so the
1754 address never reaches the final cutoff time.
1756 This is a fairly common occurrence when a user is bumping up against
1757 their storage quota. Exim had some logic in its local delivery code
1758 to deal with this. However it did not apply to per-recipient defers
1759 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1761 This change adds a proper retry rule check during routing so that the
1762 final cutoff time is checked against the message's age. We only do
1763 this check if there is an address retry record and there is not a
1764 domain retry record; this implies that previous attempts to handle
1765 the address had the retry_use_local_parts option turned on. We use
1766 this as an approximation for the destination being like a local
1767 delivery, as in LMTP.
1769 I suspect this new check makes the old local delivery cutoff check
1770 redundant, but I have not verified this so I left the code in place.
1772 TF/02 Correct gecos expansion when From: is a prefix of the username.
1774 Test 0254 submits a message to Exim with the header
1778 When I ran the test suite under the user fanf2, Exim expanded
1779 the header to contain my full name, whereas it should have added
1780 a Resent-Sender: header. It erroneously treats any prefix of the
1781 username as equal to the username.
1783 This change corrects that bug.
1785 GF/01 DCC debug and logging tidyup
1786 Error conditions log to paniclog rather than rejectlog.
1787 Debug lines prefixed by "DCC: " to remove any ambiguity.
1789 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1791 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1792 Bug spotted by Jeremy Harris; was flawed since initial commit.
1793 Would have resulted in OCSP responses post-SNI triggering an Exim
1794 NULL dereference and crash.
1796 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1798 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1799 Bug detection, analysis and fix by Samuel Thibault.
1800 Bugzilla 1331, Debian bug #698092.
1802 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1804 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1805 Server implementation by Todd Lyons, client by JH.
1806 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1807 config variable "prdr_enable" controls whether the server
1808 advertises the facility. If the client requests PRDR a new
1809 acl_data_smtp_prdr ACL is called once for each recipient, after
1810 the body content is received and before the acl_smtp_data ACL.
1811 The client is controlled by both of: a hosts_try_prdr option
1812 on the smtp transport, and the server advertisement.
1813 Default client logging of deliveries and rejections involving
1814 PRDR are flagged with the string "PRDR".
1816 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1817 fclose(). Diagnosis by Todd Lyons.
1819 PP/17 Update configure.default to handle IPv6 localhost better.
1820 Patch by Alain Williams (plus minor tweaks).
1823 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1824 This is now consistent with GnuTLS, and is now documented: the
1825 previous undocumented portable approach to treating the option as
1826 unset was to force an expansion failure. That still works, and
1827 an empty string is now equivalent.
1829 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1830 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1831 not performing validation itself.
1833 PP/20 Added force_command boolean option to pipe transport.
1834 Patch from Nick Koston, of cPanel Inc.
1836 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1839 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1841 PP/21 Fix eximon continuous updating with timestamped log-files.
1842 Broken in a format-string cleanup in 4.80, missed when I repaired the
1843 other false fix of the same issue.
1844 Report and fix from Heiko Schlichting.
1847 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1848 Report from Prashanth Katuri.
1850 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1851 It's SecureTransport, so affects any MacOS clients which use the
1852 system-integrated TLS libraries, including email clients.
1854 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1855 using a MIME ACL for non-SMTP local injection.
1856 Report and assistance in diagnosis by Warren Baker.
1858 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1860 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1862 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1863 last id that failed so it may be referenced in subsequent ACL's.
1865 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1868 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1869 ldap library initialization, allowing self-signed CA's to be
1870 used. Also properly sets require_cert option later in code by
1871 using NULL (global ldap config) instead of ldap handle (per
1872 session). Bug diagnosis and testing by alxgomz.
1874 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1875 the src/util/ subdirectory.
1877 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1878 renamed to Transport Post Delivery Action by Jeremy Harris, as
1881 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1882 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1883 redis_servers = needs to be configured which will be used by the redis
1884 lookup. Patch from Warren Baker, of The Packet Hub.
1886 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1888 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1889 hostname or reverse DNS when processing a host list. Used suggestions
1890 from multiple comments on this bug.
1892 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1894 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1895 Missed a few lines, added it to make the runtest require no keyboard
1898 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1899 contains upper case chars. Make router use caseful_local_part.
1901 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1902 support when GnuTLS has been built with p11-kit.
1908 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1910 This, or similar/improved, will also be change PP/11 of 4.82.
1916 PP/01 Handle short writes when writing local log-files.
1917 In practice, only affects FreeBSD (8 onwards).
1918 Bugzilla 1053, with thanks to Dmitry Isaikin.
1920 NM/01 Bugzilla 949 - Documentation tweak
1922 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1925 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1927 PP/02 Implemented gsasl authenticator.
1929 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1931 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1932 `pkg-config foo` for cflags/libs.
1934 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1935 with rest of GSASL and with heimdal_gssapi.
1937 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1938 `pkg-config foo` for cflags/libs for the TLS implementation.
1940 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1941 properties get this fed in as external SSF. A number of robustness
1942 and debugging improvements to the cyrus_sasl authenticator.
1944 PP/08 cyrus_sasl server now expands the server_realm option.
1946 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1947 Patch by Jeremy Harris.
1949 PP/10 Added dbmjz lookup type.
1951 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1953 PP/12 MAIL args handles TAB as well as SP, for better interop with
1954 non-compliant senders.
1955 Analysis and variant patch by Todd Lyons.
1957 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1958 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1959 Patch from Dirk Mueller <dmueller@suse.com>
1961 PP/13 tls_peerdn now print-escaped for spool files.
1962 Observed some $tls_peerdn in wild which contained \n, which resulted
1963 in spool file corruption.
1965 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1966 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1967 or write after TLS renegotiation, which otherwise led to messages
1970 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1971 as a tracking header (ie: a signed header comes before the signature).
1972 Patch from Wolfgang Breyha.
1974 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1975 comma-sep list; embedded commas doubled.
1977 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1979 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1981 Report and patch from Dmitry Banschikov.
1983 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1984 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1985 failures. We appear to now support TLS1.1+ with Exim.
1987 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1988 lets Exim select keys and certificates based upon TLS SNI from client.
1989 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1990 before an outbound SMTP session. New log_selector, +tls_sni.
1992 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1993 NULL dereference. Report and patch from Alun Jones.
1995 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1996 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1997 Not seeing resolver debug output on NetBSD, but suspect this is a
1998 resolver implementation change.
2000 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
2001 Left warnings. Added "eximon gdb" invocation mode.
2003 PP/21 Defaulting "accept_8bitmime" to true, not false.
2005 PP/22 Added -bw for inetd wait mode support.
2007 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
2008 locate the relevant includes and libraries. Made this the default.
2010 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
2011 Bugzilla 1246, report and most of solution from Tomasz Kusy.
2013 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
2014 This may cause build issues on older platforms.
2016 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
2017 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
2018 gnutls_require_mac & gnutls_require_protocols (no longer supported).
2019 Added SNI support via GnuTLS too.
2020 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
2022 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
2024 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
2025 Applied second patch from Janne, implementing suggestion to default
2026 multiple-strings-in-record handling to match SPF spec.
2028 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
2030 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
2031 read-only, out of scope).
2032 Patch from Wolfgang Breyha, report from Stuart Northfield.
2034 PP/29 Fix three issues highlighted by clang analyser static analysis.
2035 Only crash-plausible issue would require the Cambridge-specific
2036 iplookup router and a misconfiguration.
2037 Report from Marcin Mirosław.
2039 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
2041 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
2042 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
2043 As part of this, removing so much warning spew let me fix some minor
2044 real issues in debug logging.
2046 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
2047 assignment on my part. Fixed.
2049 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
2050 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
2051 Janne Snabb (who went above and beyond: thank you).
2053 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
2054 string otherwise requires a connection and a bunch more work and it's
2055 relatively easy to get wrong. Should also expose TLS library linkage
2058 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
2059 64-bit ${eval} (JH/03).
2061 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
2062 GNU libc to support some of the 64-bit stuff, should not lead to
2063 conflicts. Defined before os.h is pulled in, so if a given platform
2064 needs to override this, it can.
2066 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
2067 protection layer was required, which is not implemented.
2068 Bugzilla 1254, patch from Wolfgang Breyha.
2070 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
2071 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
2072 tls_dhparam take prime identifiers. Also unbreak combination of
2073 OpenSSL+DH_params+TLSSNI.
2075 PP/39 Disable SSLv2 by default in OpenSSL support.
2081 PP/01 Solaris build fix for Oracle's LDAP libraries.
2082 Bugzilla 1109, patch from Stephen Usher.
2084 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
2086 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
2089 TF/02 Fix a couple more cases where we did not log the error message
2090 when unlink() failed. See also change 4.74-TF/03.
2092 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
2093 lock up or crash if it happened to be inside a call to libc when it
2094 got a SIGUSR1 from exiwhat.
2096 The SIGUSR1 handler appends the current process status to the process
2097 log which is later printed by exiwhat. It used to use the general
2098 purpose logging code to do this, but several functions it calls are
2099 not safe for signals.
2101 The new output code in the SIGUSR1 handler is specific to the process
2102 log, and simple enough that it's easy to inspect for signal safety.
2103 Removing some special cases also simplifies the general logging code.
2104 Removing the spurious timestamps from the process log simplifies
2107 TF/04 Improved ratelimit ACL condition.
2109 The /noupdate option has been deprecated in favour of /readonly which
2110 has clearer semantics. The /leaky, /strict, and /readonly update modes
2111 are mutually exclusive. The update mode is no longer included in the
2112 database key; it just determines when the database is updated. (This
2113 means that when you upgrade Exim will forget old rate measurements.)
2115 Exim now checks that the per_* options are used with an update mode that
2116 makes sense for the current ACL. For example, when Exim is processing a
2117 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
2118 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
2119 must specify per_mail/readonly. If you omit the update mode it defaults to
2120 /leaky where that makes sense (as before) or /readonly where required.
2122 The /noupdate option is now undocumented but still supported for
2123 backwards compatibility. It is equivalent to /readonly except that in
2124 ACLs where /readonly is required you may specify /leaky/noupdate or
2125 /strict/noupdate which are treated the same as /readonly.
2127 A useful new feature is the /count= option. This is a generalization
2128 of the per_byte option, so that you can measure the throughput of other
2129 aggregate values. For example, the per_byte option is now equivalent
2130 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
2132 The per_rcpt option has been generalized using the /count= mechanism
2133 (though it's more complicated than the per_byte equivalence). When it is
2134 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
2135 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
2136 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
2137 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
2138 Note that using per_rcpt with a non-readonly update mode in more than
2139 one ACL will cause the recipients to be double-counted. (The per_mail
2140 and per_byte options don't have this problem.)
2142 The handling of very low rates has changed slightly. If the computed rate
2143 is less than the event's count (usually one) then this event is the first
2144 after a long gap. In this case the rate is set to the same as this event's
2145 count, so that the first message of a spam run is counted properly.
2147 The major new feature is a mechanism for counting the rate of unique
2148 events. The new per_addr option counts the number of different
2149 recipients that someone has sent messages to in the last time period. It
2150 behaves like per_rcpt if all the recipient addresses are different, but
2151 duplicate recipient addresses do not increase the measured rate. Like
2152 the /count= option this is a general mechanism, so the per_addr option
2153 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
2154 example, measure the rate that a client uses different sender addresses
2155 with the options per_mail/unique=$sender_address. There are further
2156 details in the main documentation.
2158 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
2160 TF/06 Removed a few PCRE remnants.
2162 TF/07 Automatically extract Exim's version number from tags in the git
2163 repository when doing development or release builds.
2165 PP/02 Raise smtp_cmd_buffer_size to 16kB.
2166 Bugzilla 879. Patch from Paul Fisher.
2168 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
2169 Heavily based on revision 40f9a89a from Simon Arlott's tree.
2172 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
2174 PP/05 Variable $av_failed, true if the AV scanner deferred.
2175 Bugzilla 1078. Patch from John Horne.
2177 PP/06 Stop make process more reliably on build failure.
2178 Bugzilla 1087. Patch from Heiko Schlittermann.
2180 PP/07 Make maildir_use_size_file an _expandable_ boolean.
2181 Bugzilla 1089. Patch from Heiko Schlittermann.
2183 PP/08 Handle ${run} returning more data than OS pipe buffer size.
2184 Bugzilla 1131. Patch from Holger Weiß.
2186 PP/09 Handle IPv6 addresses with SPF.
2187 Bugzilla 860. Patch from Wolfgang Breyha.
2189 PP/10 GnuTLS: support TLS 1.2 & 1.1.
2191 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
2194 PP/11 match_* no longer expand right-hand-side by default.
2195 New compile-time build option, EXPAND_LISTMATCH_RHS.
2196 New expansion conditions, "inlist", "inlisti".
2198 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
2200 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
2202 PP/14 fix log_write() format string regression from TF/03.
2203 Bugzilla 1152. Patch from Dmitry Isaikin.
2209 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
2211 PP/02 Harmonised TLS library version reporting; only show if debugging.
2212 Layout now matches that introduced for other libraries in 4.74 PP/03.
2214 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
2216 PP/04 New "dns_use_edns0" global option.
2218 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
2221 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
2222 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
2224 TK/01 Updated PolarSSL code to 0.14.2.
2225 Bugzilla 1097. Patch from Andreas Metzler.
2227 PP/07 Catch divide-by-zero in ${eval:...}.
2228 Fixes bugzilla 1102.
2230 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
2233 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
2234 format-string attack -- SECURITY: remote arbitrary code execution.
2236 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
2237 time unintentionally subject to list matching rules, letting the header
2238 cause arbitrary Exim lookups (of items which can occur in lists, *not*
2239 arbitrary string expansion). This allowed for information disclosure.
2241 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
2242 INT_MIN/-1 -- value coerced to INT_MAX.
2248 NM/01 Workaround for PCRE version dependency in version reporting
2251 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
2252 This fixes portability to compilers other than gcc, notably
2253 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
2255 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
2256 makefiles for portability to HP-UX and POSIX correctness.
2258 PP/01 Permit LOOKUP_foo enabling on the make command-line.
2259 Also via indented variable definition in the Makefile.
2260 (Debugging by Oliver Heesakkers).
2262 PP/02 Restore caching of spamd results with expanded spamd_address.
2263 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
2265 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
2266 Improves build reliability. Fix from: Frank Elsner
2268 NM/02 Fix wide character breakage in the rfc2047 coding
2269 Fixes bug 1064. Patch from Andrey N. Oktyabrski
2271 NM/03 Allow underscore in dnslist lookups
2272 Fixes bug 1026. Patch from Graeme Fowler
2274 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
2275 Code patches from Adam Ciarcinski of NetBSD.
2277 NM/04 Fixed exiqgrep to cope with mailq missing size issue
2280 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
2281 is logged, to avoid truncation. Patch from John Horne.
2283 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
2284 Patch from Jakob Hirsch.
2286 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
2287 SQL string expansion failure details.
2288 Patch from Andrey Oktyabrski.
2290 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
2291 Patch from Simon Arlott.
2293 PP/09 New lookups functionality failed to compile on old gcc which rejects
2294 extern declarations in function scope.
2295 Patch from Oliver Fleischmann
2297 PP/10 Use sig_atomic_t for flags set from signal handlers.
2298 Check getgroups() return and improve debugging.
2299 Fixed developed for diagnosis in bug 927 (which turned out to be
2302 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
2303 Patch from Mark Zealey.
2305 PP/12 Bugzilla 1056: Improved spamd server selection.
2306 Patch from Mark Zealey.
2308 PP/13 Bugzilla 1086: Deal with maildir quota file races.
2309 Based on patch from Heiko Schlittermann.
2311 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
2312 Patch from Uwe Doering, sign-off by Michael Haardt.
2314 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
2315 variable declaration deep within a block. Bug and patch from
2318 PP/15 lookups-Makefile IRIX compatibility coercion.
2320 PP/16 Make DISABLE_DKIM build knob functional.
2322 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
2323 Patch by Simon Arlott
2325 TF/03 Fix valgrind.h portability to C89 compilers that do not support
2326 variable argument macros. Our copy now differs from upstream.
2332 TF/01 Failure to get a lock on a hints database can have serious
2333 consequences so log it to the panic log.
2335 TF/02 Log LMTP confirmation messages in the same way as SMTP,
2336 controlled using the smtp_confirmation log selector.
2338 TF/03 Include the error message when we fail to unlink a spool file.
2340 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
2341 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
2342 for maintaining out-of-tree patches for some time.
2344 PP/01 Bugzilla 139: Documentation and portability issues.
2345 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
2346 Handle per-OS dynamic-module compilation flags.
2348 PP/02 Let /dev/null have normal permissions.
2349 The 4.73 fixes were a little too stringent and complained about the
2350 permissions on /dev/null. Exempt it from some checks.
2351 Reported by Andreas M. Kirchwitz.
2353 PP/03 Report version information for many libraries, including
2354 Exim version information for dynamically loaded libraries. Created
2355 version.h, now support a version extension string for distributors
2356 who patch heavily. Dynamic module ABI change.
2358 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
2359 privilege escalation vulnerability whereby the Exim run-time user
2360 can cause root to append content of the attacker's choosing to
2363 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
2366 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
2367 If dropping privileges for untrusted macros, we disabled normal logging
2368 on the basis that it would fail; for the Exim run-time user, this is not
2369 the case, and it resulted in successful deliveries going unlogged.
2370 Fixed. Reported by Andreas Metzler.
2376 PP/01 Date: & Message-Id: revert to normally being appended to a message,
2377 only prepend for the Resent-* case. Fixes regression introduced in
2378 Exim 4.70 by NM/22 for Bugzilla 607.
2380 PP/02 Include check_rfc2047_length in configure.default because we're seeing
2381 increasing numbers of administrators be bitten by this.
2383 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
2385 PP/03 Bugzilla 994: added openssl_options main configuration option.
2387 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
2389 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
2391 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
2393 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
2394 without a peer certificate, leading to a segfault because of an
2395 assumption that peers always have certificates. Be a little more
2396 paranoid. Problem reported by Martin Tscholak.
2398 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
2399 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
2400 NB: ClamAV planning to remove STREAM in "middle of 2010".
2401 CL also introduces -bmalware, various -d+acl logging additions and
2402 more caution in buffer sizes.
2404 PP/09 Implemented reverse_ip expansion operator.
2406 PP/10 Bugzilla 937: provide a "debug" ACL control.
2408 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
2410 PP/12 Bugzilla 973: Implement --version.
2412 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
2414 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
2416 PP/15 Bugzilla 816: support multiple condition rules on Routers.
2418 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
2419 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
2420 ignore trailing whitespace.
2422 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2424 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2427 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2428 Notification from Dr Andrew Aitchison.
2430 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2431 ExtendedDetectionInfo response format.
2432 Notification from John Horne.
2434 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2437 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2438 XSL and documented dependency on system catalogs, with examples of how
2441 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2444 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2445 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2446 configuration file which is writeable by the Exim user or group.
2448 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2449 of configuration files to cover files specified with the -C option if
2450 they are going to be used with root privileges, not just the default
2453 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2454 option (effectively making it always true).
2456 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2457 files to be used while preserving root privileges.
2459 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2460 that rogue child processes cannot use them.
2462 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2463 run-time user, instead of root.
2465 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2466 Exim run-time user without dropping privileges.
2468 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2469 result string, instead of calling string_vformat() twice with the same
2472 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2473 for other users. Others should always drop root privileges if they use
2474 -C on the command line, even for a whitelisted configure file.
2476 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2478 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2484 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2485 $data_path, and $header_path variables; fixed documentation bugs and
2488 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2489 exipick to access non-standard spools, including the "frozen" queue
2492 NM/01 Bugzilla 965: Support mysql stored procedures.
2493 Patch from Alain Williams
2495 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2497 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2498 Patch from Andreas Metzler
2500 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2501 Patch from Kirill Miazine
2503 NM/05 Bugzilla 671: Added umask to procmail example.
2505 JJ/03 installed exipick 20100323.0, fixing doc bug
2507 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2508 directory. Notification and patch from Dan Rosenberg.
2510 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2512 TK/02 Improve log output when DKIM signing operation fails.
2514 MH/01 Treat the transport option dkim_domain as a colon separated
2515 list, not as a single string, and sign the message with each element,
2516 omitting multiple occurences of the same signer.
2518 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2519 Bugzilla 985, 986. Patch by Simon Arlott
2521 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2522 Patch by Simon Arlott
2524 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2525 MBX locking. Notification from Dan Rosenberg.
2531 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2533 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2535 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2537 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2539 NM/04 Bugzilla 915: Flag broken perl installation during build.
2545 TK/01 Added patch by Johannes Berg that expands the main option
2546 "spamd_address" if it starts with a dollar sign.
2548 TK/02 Write list of recipients to X-Envelope-Sender header when building
2549 the mbox-format spool file for content scanning (suggested by Jakob
2552 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2553 (http://www.dcc-servers.net/) support via dccifd. Activated by
2554 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2556 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2557 by Mark Daniel Reidel <mr@df.eu>.
2559 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2560 When building exim an external PCRE library is now needed -
2561 PCRE is a system library on the majority of modern systems.
2562 See entry on PCRE_LIBS in EDITME file.
2564 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2565 conversation. Added nologin parameter to request.
2566 Patch contributed by Kirill Miazine.
2568 TF/01 Do not log submission mode rewrites if they do not change the address.
2570 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2572 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2573 log files in place. Contributed by Roberto Lima.
2575 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2577 TF/03 Bugzilla 615: When checking the local_parts router precondition
2578 after a local_part_suffix or local_part_prefix option, Exim now
2579 does not use the address's named list lookup cache, since this
2580 contains cached lookups for the whole local part.
2582 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2583 Robert Millan. Documentation is in experimental-spec.txt.
2585 TF/04 Bugzilla 668: Fix parallel build (make -j).
2587 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2589 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2590 Patch provided by Jan Srzednicki.
2592 TF/05 Leading white space used to be stripped from $spam_report which
2593 wrecked the formatting. Now it is preserved.
2595 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2596 that they are available at delivery time.
2598 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2600 TF/08 TLS error reporting now respects the incoming_interface and
2601 incoming_port log selectors.
2603 TF/09 Produce a more useful error message if an SMTP transport's hosts
2604 setting expands to an empty string.
2606 NM/06 Bugzilla 744: EXPN did not work under TLS.
2607 Patch provided by Phil Pennock.
2609 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2610 Patch provided by Richard Godbee.
2612 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2613 acl_smtp_notquit, added index entry.
2615 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2616 Patch provided by Eugene Bujak.
2618 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2619 accept(). Patch provided by Maxim Dounin.
2621 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2622 Patch provided by Phil Pennock.
2624 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2626 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2627 Patch provided by Brad "anomie" Jorsch.
2629 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2630 Patch provided by Dean Brooks.
2632 TK/05 Add native DKIM support (does not depend on external libraries).
2634 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2635 Patch provided by Graeme Fowler.
2637 NM/16 Bugzilla 851: Documentation example syntax fix.
2639 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2641 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2644 NM/19 Bugzilla 745: TLS version reporting.
2645 Patch provided by Phil Pennock.
2647 NM/20 Bugzilla 167: bool: condition support.
2648 Patch provided by Phil Pennock.
2650 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2651 clients. Patch provided by Phil Pennock.
2653 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2654 Patch provided by Brad "anomie" Jorsch.
2656 NM/23 Bugzilla 687: Fix misparses in eximstats.
2657 Patch provided by Heiko Schlittermann.
2659 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2660 Patch provided by Heiko Schlittermann.
2662 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2663 plus update to original patch.
2665 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2667 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2668 Patch provided by David Brownlee.
2670 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2672 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2674 NM/30 Bugzilla 888: TLS documentation bugfixes.
2676 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2678 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2679 Unlike the original bugzilla I have changed all shell scripts in src tree.
2681 NM/33 Bugzilla 898: Transport filter timeout fix.
2682 Patch by Todd Rinaldo.
2684 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2685 Patch by Serge Demonchaux.
2687 NM/35 Bugzilla 39: Base64 decode bug fixes.
2688 Patch by Jakob Hirsch.
2690 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2692 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2694 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2696 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2702 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2703 ALT-N's libdkim that I have put here:
2704 http://duncanthrax.net/exim-experimental/
2706 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2707 were called 'true' and it seems that C99 defines that as a reserved
2708 keyword to be used with 'bool' variable types. That means you could
2709 not include C99-style headers which use bools without triggering
2710 build errors in sieve.c.
2712 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2713 as mailq or other aliases. Changed the --help handling significantly
2714 to do whats expected. exim_usage() emits usage/help information.
2716 SC/01 Added the -bylocaldomain option to eximstats.
2718 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2720 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2722 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2728 PH/01 Another patch from the Sieve maintainer.
2730 PH/02 When an IPv6 address is converted to a string for single-key lookup
2731 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2732 dots are used instead of colons so that keys in lsearch files need not
2733 contain colons. This was done some time before quoting was made available
2734 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2735 (notated using the quote facility) so as to distinguish them from IPv4
2736 keys. This meant that lookups for IP addresses in host lists did not work
2737 for iplsearch lookups.
2739 This has been fixed by arranging for IPv6 addresses to be expressed with
2740 colons if the lookup type is iplsearch. This is not incompatible, because
2741 previously such lookups could never work.
2743 The situation is now rather anomalous, since one *can* have colons in
2744 ordinary lsearch keys. However, making the change in all cases is
2745 incompatible and would probably break a number of configurations.
2747 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2750 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2751 conversion specification without a maximum field width, thereby enabling
2752 a rogue spamd server to cause a buffer overflow. While nobody in their
2753 right mind would setup Exim to query an untrusted spamd server, an
2754 attacker that gains access to a server running spamd could potentially
2755 exploit this vulnerability to run arbitrary code as the Exim user.
2757 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2758 $primary_hostname instead of what libspf2 thinks the hosts name is.
2760 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2761 a directory entry by the name of the lookup key. Previously, if a
2762 symlink pointed to a non-existing file or a file in a directory that
2763 Exim lacked permissions to read, a lookup for a key matching that
2764 symlink would fail. Now it is enough that a matching directory entry
2765 exists, symlink or not. (Bugzilla 503.)
2767 PH/03 The body_linecount and body_zerocount variables are now exported in the
2770 PH/04 Added the $dnslist_matched variable.
2772 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2773 This means they are set thereafter only if the connection becomes
2776 PH/06 Added the client_condition to authenticators so that some can be skipped
2777 by clients under certain conditions.
2779 PH/07 The error message for a badly-placed control=no_multiline_responses left
2780 "_responses" off the end of the name.
2782 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2784 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2785 (without spaces) instead of just copying the configuration text.
2787 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2789 PH/11 Added $max_received_linelength.
2791 PH/12 Added +ignore_defer and +include_defer to host lists.
2793 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2794 way in which PCRE > 7.0 is built.
2796 PH/14 Implemented queue_only_load_latch.
2798 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2799 MAIL command. The effect was to mangle the value on 64-bit systems.
2801 PH/16 Another patch from the Sieve maintainer.
2803 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2805 PH/18 If a system quota error occurred while trying to create the file for
2806 a maildir delivery, the message "Mailbox is full" was not appended to the
2807 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2808 only to a quota excession during the actual writing of the file.
2810 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2811 characters?) which causes problems in log lines. The DN values are now
2812 passed through string_printing() before being added to log lines.
2814 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2815 and InterBase are left for another time.)
2817 PH/21 Added message_body_newlines option.
2819 PH/22 Guard against possible overflow in moan_check_errorcopy().
2821 PH/23 POSIX allows open() to be a macro; guard against that.
2823 PH/24 If the recipient of an error message contained an @ in the local part
2824 (suitably quoted, of course), incorrect values were put in $domain and
2825 $local_part during the evaluation of errors_copy.
2831 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2832 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2835 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2836 issue a MAIL command.
2838 PH/02 In an ACL statement such as
2840 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2842 if a client was not listed at all, or was listed with a value other than
2843 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2844 the condition was not true (as it should be), so access was not denied.
2845 The bug was that the ! inversion was incorrectly passed on to the second
2846 item. This has been fixed.
2848 PH/03 Added additional dnslists conditions == and =& which are different from
2849 = and & when the dns lookup returns more than one IP address.
2851 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2852 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2854 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2855 FSYNC, which compiles an option called disable_fsync that allows for
2856 bypassing fsync(). The documentation is heavily laced with warnings.
2858 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2860 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2861 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2862 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2863 including adding "make clean"; (3) Added -fPIC when compiling the test
2864 dynamically loaded module, to get rid of a warning.
2866 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2867 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2868 = 0s. The bug is otherwise harmless.
2870 PH/07 There was a bug in the dovecot authenticator such that the value of
2871 $auth1 could be overwritten, and so not correctly preserved, after a
2872 successful authentication. This usually meant that the value preserved by
2873 the server_setid option was incorrect.
2875 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2877 PH/09 Installed PCRE release 7.0.
2879 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2880 run for batched SMTP input. It is now run at the start of every message
2881 in the batch. While fixing this I discovered that the process information
2882 (output by running exiwhat) was not always getting set for -bs and -bS
2883 input. This is fixed, and it now also says "batched" for BSMTP.
2885 PH/11 Added control=no_pipelining.
2887 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2888 patch, slightly modified), and move the expansion of helo_data till after
2889 the connection is made in the smtp transport (so it can use these
2892 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2894 PH/14 Added log_selector = +pid.
2896 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2898 PH/16 Add ${if forany and ${if forall.
2900 PH/17 Added dsn_from option to vary the From: line in DSNs.
2902 PH/18 Flush SMTP output before performing a callout, unless control =
2903 no_callout_flush is set.
2905 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2906 was true (the default) a successful delivery failed to delete the retry
2907 item, thus causing premature timeout of the address. The bug is now
2910 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2912 PH/21 Long custom messages for fakedefer and fakereject are now split up
2913 into multiline responses in the same way that messages for "deny" and
2914 other ACL rejections are.
2916 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2917 with slight modification.
2919 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2920 draft, changing the syntax and factoring some duplicate code".
2922 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2923 for deliveries of the second and subsequent messages over the same SMTP
2926 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2927 ${reduce, with only minor "tidies".
2929 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2931 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2932 expansion side effects.
2934 PH/27 When a message times out after an over-quota error from an Exim-imposed
2935 quota, the bounce message says "mailbox is full". This message was not
2936 being given when it was a system quota that was exceeded. It now should
2939 MH/03 Made $recipients available in local_scan(). local_scan() already has
2940 better access to the recipient list through recipients_list[], but
2941 $recipients can be useful in postmaster-provided expansion strings.
2943 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2944 in the case of a MAIL command with additional options following the
2945 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2946 were accidentally chopped off.
2948 PH/29 SMTP synchronization checks are implemented when a command is read -
2949 there is a check that no more input is waiting when there shouldn't be
2950 any. However, for some commands, a delay in an ACL can mean that it is
2951 some time before the response is written. In this time, more input might
2952 arrive, invalidly. So now there are extra checks after an ACL has run for
2953 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2954 pipelining has not been advertised.
2956 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2958 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2959 returned in the SMTP response when smtp_return_error_details was false.
2960 This has been fixed.
2962 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2963 instead of the C I/O that was originally supplied, because problems were
2964 reported on Solaris.
2966 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2967 Exim which did not show up earlier: it was assuming that a call to
2968 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2969 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2970 was a macro that became an assignment, so it seemed to work. This has
2971 changed to a proper function call with a void return, hence the compile
2972 error. Exim's code has been fixed.
2974 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2977 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2979 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2981 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2982 because it turns out to be too big or there is a timeout) there is no
2983 "Completed" line in the log. When some messages of this type were
2984 selected by exigrep, they were listed as "not completed". Others were
2985 picked up by some special patterns. I have improved the selection
2986 criteria to be more general.
2988 PH/38 The host_find_failed option in the manualroute router can now be set
2989 to "ignore", to completely ignore a host whose IP address cannot be
2990 found. If all hosts are ignored, the behaviour is controlled by the new
2991 host_all_ignored option.
2993 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2994 homing or because of multiple MX records with /mx) generated more than
2995 one IP address, and the following item turned out to be the local host,
2996 all the secondary addresses of the first item were incorrectly removed
2997 from the list, along with the local host and any following hosts (which
2998 is what is supposed to happen).
3000 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
3001 whoever called Exim into the -H file. In the case of the daemon it was
3002 behaving confusingly. When first started, it used values for whoever
3003 started the daemon, but after a SIGHUP it used the Exim user (because it
3004 calls itself on a restart). I have changed the code so that it now always
3007 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
3008 message are rejected with the same error (e.g. no authentication or bad
3009 sender address), and a DATA command is nevertheless sent (as can happen
3010 with PIPELINING or a stupid MUA), the error message that was given to the
3011 RCPT commands is included in the rejection of the DATA command. This is
3012 intended to be helpful for MUAs that show only the final error to their
3015 PH/42 Another patch from the Sieve maintainer.
3017 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
3018 Eximstats - Fixed some broken HTML links and added missing column headers
3020 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
3021 columns for Rejects, Temp Rejects, Ham, and Spam rows.
3023 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
3025 PH/43 Yet another patch from the Sieve maintainer.
3027 PH/44 I found a way to check for a TCP/IP connection going away before sending
3028 the response to the final '.' that terminates a message, but only in the
3029 case where the client has not sent further data following the '.'
3030 (unfortunately, this is allowed). However, in many cases there won't be
3031 any further data because there won't be any more messages to send. A call
3032 to select() can be used: if it shows that the input is "ready", there is
3033 either input waiting, or the socket has been closed. An attempt to read
3034 the next input character can distinguish the two cases. Previously, Exim
3035 would have sent an OK response which the client would never have see.
3036 This could lead to message repetition. This fix should cure that, at
3037 least in a lot of common cases.
3039 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
3040 advertised in response to EHLO.
3046 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
3047 fixed by 4.65/MH/01 (is this a record?) are fixed:
3049 (i) An empty string was always treated as zero by the numeric comparison
3050 operators. This behaviour has been restored.
3052 (ii) It is documented that the numeric comparison operators always treat
3053 their arguments as decimal numbers. This was broken in that numbers
3054 starting with 0 were being interpreted as octal.
3056 While fixing these problems I realized that there was another issue that
3057 hadn't been noticed. Values of message_size_limit (both the global option
3058 and the transport option) were treated as octal if they started with 0.
3059 The documentation was vague. These values are now always treated as
3060 decimal, and I will make that clear in the documentation.
3066 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
3067 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
3070 MH/01 Don't check that the operands of numeric comparison operators are
3071 integers when their expansion is in "skipping" mode (fixes bug
3072 introduced by 4.64-PH/07).
3074 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
3075 child addresses, Exim now panics and dies. Previously, because the count
3076 is held in a short int, deliveries were likely to be lost. As such a
3077 large number of recipients for a single message is ridiculous
3078 (performance will be very, very poor), I have chosen to impose a limit
3079 rather than extend the field.
3085 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
3086 leftover -K file (the existence of which was triggered by #402).
3087 While we were at it, introduced process PID as part of the -K
3088 filename. This should rule out race conditions when creating
3091 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
3092 processing considerably. Previous code took too long for large mails,
3093 triggering a timeout which in turn triggers #401.
3095 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
3096 in the DK code in transports.c. sendfile() is not really portable,
3097 hence the _LINUX specificness.
3099 TF/01 In the add_headers option to the mail command in an Exim filter,
3100 there was a bug that Exim would claim a syntax error in any
3101 header after the first one which had an odd number of characters
3104 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
3105 callout verification, Exim cached a "reject" for the entire domain. This
3106 is correct for most verifications, but it is not correct for a recipient
3107 verification with use_sender or use_postmaster set, because in that case
3108 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
3109 case of MAIL FROM:<> rejection from other early rejections (e.g.
3110 rejection of HELO). When verifying a recipient using a non-null MAIL
3111 address, the cache is ignored if it shows MAIL FROM:<> rejection.
3112 Whatever the result of the callout, the value of the domain cache is
3113 left unchanged (for any other kind of callout, getting as far as trying
3114 RCPT means that the domain itself is ok).
3116 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
3119 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
3120 manifest itself as EPIPE rather than ECONNECT. When tidying away a
3121 session, the daemon ignores ECONNECT errors and logs others; it now
3122 ignores EPIPE as well.
3124 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
3125 (quoted-printable decoding).
3127 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
3128 later the small subsequent patch to fix an introduced bug.
3130 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
3132 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
3134 PH/08 An error is now given if message_size_limit is specified negative.
3136 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
3137 to be given (somewhat) arbitrary names.
3139 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
3142 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
3143 miscellaneous code fixes
3145 PH/10 Added the log_reject_target ACL modifier to specify where to log
3148 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
3149 hostname. This is wrong, because it relates to the incoming message (and
3150 probably the interface on which it is arriving) and not to the outgoing
3151 callout (which could be using a different interface). This has been
3152 changed to use the value of the helo_data option from the smtp transport
3153 instead - this is what is used when a message is actually being sent. If
3154 there is no remote transport (possible with a router that sets up host
3155 addresses), $smtp_active_hostname is used.
3157 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
3158 tweaks were necessary in order to get it to work (see also 21 below):
3159 (a) The code assumed that strncpy() returns a negative number on buffer
3160 overflow, which isn't the case. Replaced with Exim's string_format()
3162 (b) There were several signed/unsigned issues. I just did the minimum
3163 hacking in of casts. There is scope for a larger refactoring.
3164 (c) The code used strcasecmp() which is not a standard C function.
3165 Replaced with Exim's strcmpic() function.
3166 (d) The code set only $1; it now sets $auth1 as well.
3167 (e) A simple test gave the error "authentication client didn't specify
3168 service in request". It would seem that Dovecot has changed its
3169 interface. Fortunately there's a specification; I followed it and
3170 changed what the client sends and it appears to be working now.
3172 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
3175 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
3176 address A is aliased to B and C, where B exists and C does not. Without
3177 -v the output is "A verified" because verification stops after a
3178 successful redirection if more than one address is generated. However,
3179 with -v the child addresses are also verified. Exim was outputting "A
3180 failed to verify" and then showing the successful verification for C,
3181 with its parentage. It now outputs "B failed to verify", showing B's
3182 parentage before showing the successful verification of C.
3184 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
3185 look up a TXT record in a specific list after matching in a combined
3188 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
3189 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
3190 they consult the DNS. I had assumed they would set it the way they
3191 wanted; and indeed my experiments on Linux seem to show that in some
3192 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
3193 To be on the safe side, however, I have now made the interface to
3194 host_find_byname() similar to host_find_bydns(), with an argument
3195 containing the DNS resolver options. The host_find_byname() function now
3196 sets these options at its start, just as host_find_bydns() does. The smtp
3197 transport options dns_qualify_single and dns_search_parents are passed to
3198 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
3199 of host_find_byname() use the default settings of RES_DEFNAMES
3200 (qualify_single) but not RES_DNSRCH (search_parents).
3202 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
3203 spool_read_header() do less string testing, by means of a preliminary
3204 switch on the second character of optional "-foo" lines. (This is
3205 overdue, caused by the large number of possibilities that now exist.
3206 Originally there were few.) While I was there, I also converted the
3207 str(n)cmp tests so they don't re-test the leading "-" and the first
3208 character, in the hope this might squeeze out yet more improvement.
3210 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
3211 flag allowing group syntax was set by the header_syntax check but not
3212 turned off, possible causing trouble later; (2) The flag was not being
3213 set at all for the header_verify test, causing "group"-style headers to
3214 be rejected. I have now set it in this case, and also caused header_
3215 verify to ignore an empty address taken from a group. While doing this, I
3216 came across some other cases where the code for allowing group syntax
3217 while scanning a header line wasn't quite right (mostly, not resetting
3218 the flag correctly in the right place). These bugs could have caused
3219 trouble for malformed header lines. I hope it is now all correct.
3221 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
3222 with the "reply" argument non-NULL. The code, however (which originally
3223 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
3224 but it didn't always do it. This confused somebody who was copying the
3225 code for some other use. I have removed all the tests.
3227 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
3228 feature that was used to support insecure browsers during the U.S. crypto
3229 embargo. It requires special client support, and Exim is probably the
3230 only MTA that supported it -- and would never use it because real RSA is
3231 always available. This code has been removed, because it had the bad
3232 effect of slowing Exim down by computing (never used) parameters for the
3233 RSA_EXPORT functionality.
3235 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
3236 authenticator to fail if there's a tab character in the incoming data
3237 (there should never be unless someone is messing about, as it's supposed
3238 to be base64-encoded). Also added, on Timo's advice, the "secured" option
3239 if the connection is using TLS or if the remote IP is the same as the
3240 local IP, and the "valid-client-cert option" if a client certificate has
3243 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
3244 authenticators. This can be used for authorization after authentication
3245 succeeds. (In the case of plaintext, it servers for both authentication
3248 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
3249 if any retry times were supplied.
3251 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
3252 connection, where there is no client IP address to check. In this
3253 situation, the verify now always succeeds.
3255 PH/25 Applied John Jetmore's -Mset patch.
3257 PH/26 Added -bem to be like -Mset, but loading a message from a file.
3259 PH/27 In a string expansion for a processed (not raw) header when multiple
3260 headers of the same name were present, leading whitespace was being
3261 removed from all of them, but trailing whitespace was being removed only
3262 from the last one. Now trailing whitespace is removed from each header
3263 before concatenation. Completely empty headers in a concatenation (as
3264 before) are ignored.
3266 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
3267 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
3269 PH/29 [Removed. This was a change that I later backed out, and forgot to
3270 correct the ChangeLog entry (that I had efficiently created) before
3271 committing the later change.]
3273 PH/30 Exim was sometimes attempting to deliver messages that had suffered
3274 address errors (4xx response to RCPT) over the same connection as other
3275 messages routed to the same hosts. Such deliveries are always "forced",
3276 so retry times are not inspected. This resulted in far too many retries
3277 for the affected addresses. The effect occurred only when there were more
3278 hosts than the hosts_max_try setting in the smtp transport when it had
3279 the 4xx errors. Those hosts that it had tried were not added to the list
3280 of hosts for which the message was waiting, so if all were tried, there
3281 was no problem. Two fixes have been applied:
3283 (i) If there are any address or message errors in an SMTP delivery, none
3284 of the hosts (tried or untried) are now added to the list of hosts
3285 for which the message is waiting, so the message should not be a
3286 candidate for sending over the same connection that was used for a
3287 successful delivery of some other message. This seems entirely
3288 reasonable: after all the message is NOT "waiting for some host".
3289 This is so "obvious" that I'm not sure why it wasn't done
3290 previously. Hope I haven't missed anything, but it can't do any
3291 harm, as the worst effect is to miss an optimization.
3293 (ii) If, despite (i), such a delivery is accidentally attempted, the
3294 routing retry time is respected, so at least it doesn't keep
3295 hammering the server.
3297 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
3298 in ${readsocket because some servers need this prod.
3300 PH/32 Added some extra debug output when updating a wait-xxx database.
3302 PH/33 The hint "could be header name not terminated by colon", which has been
3303 given for certain expansion errors for a long time, was not being given
3304 for the ${if def:h_colon_omitted{... case.
3306 PH/34 The spec says: "With one important exception, whenever a domain list is
3307 being scanned, $domain contains the subject domain." There was at least
3308 one case where this was not true.
3310 PH/35 The error "getsockname() failed: connection reset by peer" was being
3311 written to the panic log as well as the main log, but it isn't really
3312 panic-worthy as it just means the connection died rather early on. I have
3313 removed the panic log writing for the ECONNRESET error when getsockname()
3316 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
3317 runs only) independently of the message's sender address. This meant
3318 that, if the 4xx error was in fact related to the sender, a different
3319 message to the same recipient with a different sender could confuse
3320 things. In particular, this can happen when sending to a greylisting
3321 server, but other circumstances could also provoke similar problems.
3322 I have changed the default so that the retry time for these errors is now
3323 based a combination of the sender and recipient addresses. This change
3324 can be overridden by setting address_retry_include_sender=false in the
3327 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
3328 remote server are returned as part of bounce messages. This was not
3329 happening for LMTP over a pipe (the lmtp transport), but now it is the
3330 same for both kinds of LMTP.
3332 PH/38 Despite being documented as not happening, Exim was rewriting addresses
3333 in header lines that were in fact CNAMEs. This is no longer the case.
3335 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
3336 and queue runs started by the daemon processed all messages. This has
3337 been fixed so that -R and -S can now usefully be given with -q<time>.
3339 PH/40 Import PCRE release 6.7 (fixes some bugs).
3341 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
3343 PH/42 Give an error if -q is specified more than once.
3345 PH/43 Renamed the variables $interface_address and $interface_port as
3346 $received_ip_address and $received_port, to make it clear that these
3347 values apply to message reception, and not to the outgoing interface when
3348 a message is delivered. (The old names remain recognized, of course.)
3350 PH/44 There was no timeout on the connect() call when using a Unix domain
3351 socket in the ${readsocket expansion. There now is.
3353 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
3354 be meaningful with "accept".
3356 SC/01 Eximstats V1.43
3357 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
3359 SC/02 Eximstats V1.44
3360 Use a glob alias rather than an array ref in the generated
3361 parser. This improves both readability and performance.
3363 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
3364 Collect SpamAssassin and rejection statistics.
3365 Don't display local sender or destination tables unless
3366 there is data to show.
3367 Added average volumes into the top table text output.
3369 SC/04 Eximstats V1.46
3370 Collect data on the number of addresses (recipients)
3371 as well as the number of messages.
3373 SC/05 Eximstats V1.47
3374 Added 'Message too big' to the list of mail rejection
3375 reasons (thanks to Marco Gaiarin).
3377 SC/06 Eximstats V1.48
3378 Mainlog lines which have GMT offsets and are too short to
3379 have a flag are now skipped.
3381 SC/07 Eximstats V1.49 (Alain Williams)
3382 Added the -emptyok flag.
3384 SC/08 Eximstats V1.50
3385 Fixes for obtaining the IP address from reject messages.
3387 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
3388 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
3389 whitespace changes from 4.64-PH/27
3391 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
3394 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
3395 are found, allow negative numbers in numeric criteria)
3397 JJ/06 exipick.20061117.2, added new $message_body_missing variable
3399 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
3400 to match changes made in 4.64-PH/43
3402 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
3404 PH/47 Put in an explicit test for a DNS lookup of an address record where the
3405 "domain" is actually an IP address, and force a failure. This locks out
3406 those revolvers/nameservers that support "A-for-A" lookups, in
3407 contravention of the specifications.
3409 PH/48 When a host name was looked up from an IP address, and the subsequent
3410 forward lookup of the name timed out, the host name was left in
3411 $sender_host_name, contrary to the specification.
3413 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
3414 restricted to single-key lookups, Exim was not diagnosing an error if
3415 * or *@ was used with a query-style lookup.
3417 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
3419 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
3420 long ago, but noone interested enough thought of it. Let's just say that
3421 the "1.1" means that there are some new functions that weren't there at
3422 some point in the past.
3424 PH/51 Error processing for expansion failure of helo_data from an smtp
3425 transport during callout processing was broken.
3427 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3428 tested/used via the -bh/-bhc/-bs options.
3430 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3431 bug, fixed in subsequent PCRE releases).
3433 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3434 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3436 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3442 SC/01 Use a glob alias rather than an array ref in eximstats generated
3443 parser. This improves both readability and performance.
3445 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3446 Don't display local sender or destination tables in eximstats unless
3447 there is data to show.
3448 Added average volumes into the eximstats top table text output.
3450 SC/03 Collect data on the number of addresses (recipients) as well
3451 as the number of messages in eximstats.
3453 TF/01 Correct an error in the documentation for the redirect router. Exim
3454 does (usually) call initgroups() when daemonizing.
3456 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3457 with consistent privilege compared to when running as a daemon.
3459 TF/03 Note in the spec that $authenticated_id is not set for local
3460 submissions from trusted users.
3462 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3463 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3465 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3466 by adding some example configuration directives to the default
3467 configuration file. A little bit of work is required to uncomment the
3468 directives and define how usernames and passwords are checked, but
3469 there is now a framework to start from.
3471 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3472 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3473 without this. I don't know how relevant this is to other LDAP libraries.
3475 PH/02 Add the verb name to the "unknown ACL verb" error.
3477 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3479 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3481 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3482 directory not expanded when it should be if an expanded home directory
3483 was set for the address (which is overridden by the transport).
3485 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3488 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3489 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3490 because it is too late at that time, and has no effect.
3492 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3493 security issue with \' (bugzilla #107). I could not use the
3494 PQescapeStringConn() function, because it needs a PGconn value as one of
3497 PH/08 When testing addresses using -bt, indicate those final addresses that
3498 are duplicates that would not cause an additional delivery. At least one
3499 person was confused, thinking that -bt output corresponded to deliveries.
3500 (Suppressing duplicates isn't a good idea as you lose the information
3501 about possibly different redirections that led to the duplicates.)
3503 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3504 systems where poll() doesn't work, in particular OS X.
3506 PH/10 Added more information to debugging output for retry time not reached.
3508 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3509 operations in malware.c.
3511 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3514 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3515 syslog_duplication set false, log lines that would normally be written
3516 both the the main log and to the reject log were not written to syslog at
3519 PH/14 In the default configuration, change the use of "message" in ACL warn
3520 statements to "add_header".
3522 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3523 not followed by a command (e.g. "seen endif").
3525 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3526 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3529 PH/17 Added extra conditions to the default value of delay_warning_condition
3533 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3534 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3535 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3538 The Auto-Submitted: and various List- headers are standardised, whereas I
3539 don't think Precedence: ever was.
3541 PH/18 Refactored debugging code in route_finduser() to show more information,
3542 in particular, the error code if getpwnam() issues one.
3544 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3545 This is apparently needed in addition to the PH/07 change above to avoid
3546 any possible encoding problems.
3548 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3549 but not after initializing Perl.
3551 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3552 output them only if debugging. By default they are written stderr,
3553 apparently, which is not desirable.
3555 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3558 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3561 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3563 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3564 authenticated or an ident call has been made. Suppress the default
3565 values for $authenticated_id and $authenticated_sender (but permit -oMai
3566 and -oMas) when testing with -bh.
3568 PH/24 Re-jigged the order of the tests in the default configuration so that the
3569 tests for valid domains and recipients precede the DNS black list and CSA
3570 tests, on the grounds that those ones are more expensive.
3572 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3573 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3574 This bug exists in every version of Exim that I still have, right back to
3577 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3578 However, an attempt to turn on case-sensitivity in a regex key by
3579 including (?-i) didn't work because the subject string was already
3580 lowercased, and the effects were non-intuitive. It turns out that a
3581 one-line patch can be used to allow (?-i) to work as expected.
3587 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3588 other effects) broke the use of negated acl sub-conditions.
3590 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3593 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3594 "Deny" causes Exim to reject the incoming connection with a 554 error.
3595 Unfortunately, if there is a major crisis, such as a disk failure,
3596 tcp-wrappers gives "deny", whereas what one would like would be some
3597 kind of temporary error. A kludge has been added to help with this.
3598 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3599 554 error is used if errno is still zero or contains ENOENT (which occurs
3600 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3603 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3605 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3606 errors. Otherwise a message that provokes a temporary error (when other
3607 messages do not) can cause a whole host to time out.
3609 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3610 addresses were routed directly to files or pipes from a redirect router.
3611 File deliveries just didn't batch; pipe deliveries might have suffered
3614 PH/06 A failure to get a lock for a hints database would erroneously always say
3615 "Failed to get write lock", even when it was really a read lock.
3617 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3618 of 0600. This has been changed to use the value of the lockfile_mode
3619 option (which defaults to 0600).
3621 PH/08 Applied small patch from the Sieve maintainer.
3623 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3624 folder from quota calculations, a direct delivery into this folder messed
3625 up the contents of the maildirsize file. This was because the regex was
3626 used only to exclude .Trash (or whatever) when the size of the mailbox
3627 was calculated. There was no check that a delivery was happening into an
3628 excluded directory. This bug has been fixed by ignoring all quota
3629 processing for deliveries into excluded directories.
3631 PH/10 Added the maildirfolder_create_regex option to appendfile.
3637 PH/01 The code for finding all the local interface addresses on a FreeBSD
3638 system running IPv6 was broken. This may well have applied to all BSD
3639 systems, as well as to others that have similar system calls. The broken
3640 code found IPv4 interfaces correctly, but gave incorrect values for the
3641 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3642 that it would not match correctly against @[] and not recognize the IPv6
3645 PH/02 The ipliteral router was not recognizing addresses of the form user@
3646 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3648 PH/03 Added disable_ipv6.
3650 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3651 decoded form, because it is most often used to construct To: headers
3652 lines in autoreplies, and the decoded form may well be syntactically
3653 invalid. However, $reply_address has leading white space removed, and all
3654 newlines turned into spaces so that the autoreply transport does not
3657 PH/05 If group was specified without a user on a router, and no group or user
3658 was specified on a transport, the group from the router was ignored.
3660 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3661 for visible compile-time settings that can be used to change these
3662 numbers, for those that want even more. Backwards compatibility with old
3663 spool files has been maintained. However, going back to a previous Exim
3664 release will lost any variables that are in spool files.
3666 PH/07 Two small changes when running in the test harness: increase delay when
3667 passing a TCP/IP connection to a new process, in case the original
3668 process has to generate a bounce, and remove special handling of
3669 127.0.0.2 (sic), which is no longer necessary.
3671 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3672 be the same on different OS.
3674 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3677 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3678 whether --show-vars was specified or not
3680 JJ/02 exipick: Added support for new ACL variable spool format introduced
3683 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3684 syntactically invalid From: or Reply-to: line, and a filter used this to
3685 generate an autoreply, and therefore failed to obtain an address for the
3686 autoreply, Exim could try to deliver to a non-existent relative file
3687 name, causing unrelated and misleading errors. What now happens is that
3688 it logs this as a hard delivery error, but does not attempt to create a
3691 PH/11 The exinext utility has a -C option for testing purposes, but although
3692 the given file was scanned by exinext itself; it wasn't being passed on
3693 when Exim was called.
3695 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3696 an end-of-file indication when reading a command response.
3698 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3699 compiled. In many other places in Exim, IPv6 addresses are always
3700 recognized, so I have changed this. It also means that IPv4 domain
3701 literals of the form [IPV4:n.n.n.n] are now always recognized.
3703 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3704 used if the router is not running as root, for example, when verifying at
3705 ACL time, or when using -bh. The debugging output from this situation was
3706 non-existent - all you got was a failure to exec. I have made two
3709 (a) Failures to set uid/gid, the current directory, or a process leader
3710 in a subprocess such as that created by queryprogram now generate
3711 suitable debugging output when -d is set.
3713 (b) The queryprogram router detects when it is not running as root,
3714 outputs suitable debugging information if -d is set, and then runs
3715 the subprocess without attempting to change uid/gid.
3717 PH/15 Minor change to Makefile for building test_host (undocumented testing
3720 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3721 additional section of a DNS packet that returns MX or SRV records.
3722 Instead, it always explicitly searches for A/AAAA records. This avoids
3723 major problems that occur when a DNS server includes only records of one
3724 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3725 fixed another bug: if SRV records were looked up and the corresponding
3726 address records were *not* found in the additional section, the port
3727 values from the SRV records were lost.
3729 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3730 using the correct key (the original address) when searching the retry
3731 rules in order to find which one to use for generating the retry hint.
3733 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3734 adding the default one. Similarly, if it contains a Reply-To: header, the
3735 errors_reply_to option, if set, is not used.
3737 PH/19 When calculating a retry time, Exim used to measure the "time since
3738 failure" by looking at the "first failed" field in the retry record. Now
3739 it does not use this if it is later than than the arrival time of the
3740 message. Instead it uses the arrival time. This makes for better
3741 behaviour in cases where some deliveries succeed, thus re-setting the
3742 "first failed" field. An example is a quota failure for a huge message
3743 when small messages continue to be delivered. Without this change, the
3744 "time since failure" will always be short, possible causing more frequent
3745 delivery attempts for the huge message than are intended.
3746 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3748 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3749 $1, $2, $3) because the numerical variables can be reset during some
3750 expansion items (e.g. "match"), thereby losing the authentication data.
3752 PH/21 Make -bV show the size of off_t variables so that the test suite can
3753 decide whether to run tests for quotas > 2G.
3755 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3756 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3757 filecount value is greater than 2G or if a quota value is greater than 2G
3758 on a system where the size of off_t is not greater than 4, a panic error
3761 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3762 never match. The debug and -bh output now contains an explicit error
3763 message indicating a malformed IPv4 address or mask.
3765 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3766 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3767 PH/23 above applies.
3769 PH/25 Do not write to syslog when running in the test harness. The only
3770 occasion when this arises is a failure to open the main or panic logs
3771 (for which there is an explicit test).
3773 PH/26 Added the /no_tell option to "control=freeze".
3775 PH/27 If a host name lookup failed very early in a connection, for example, if
3776 the IP address matched host_lookup and the reverse lookup yielded a name
3777 that did not have a forward lookup, an error message of the form "no IP
3778 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3779 could be logged. Now it outputs the IP address instead of "NULL".
3781 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3782 allows the sender and the authenticated sender to be set when
3783 submitting a message from within Exim. Since child_open_exim() is
3784 documented for local_scan(), the new function should be too.
3786 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3787 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3788 results in an empty string is now treated as unset.
3790 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3792 PH/31 Added sender_verify_fail logging option.
3794 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3795 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3796 tidied the source and removed it altogether.
3798 PH/33 When a queue run was abandoned because the load average was too high, a
3799 log line was always written; now it is written only if the queue_run log
3800 selector is set. In addition, the log line for abandonment now contains
3801 information about the queue run such as the pid. This is always present
3802 in "start" and "stop" lines but was omitted from the "abandon" line.
3804 PH/34 Omit spaces between a header name and the colon in the error message that
3805 is given when verify = headers_syntax fails (if there are lots of them,
3806 the message gets confusing).
3808 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3809 names, as there are now some PTR records that contain slashes. This check
3810 is only to protect against broken name servers that fall over on strange
3811 characters, so the fact that it applies to all lookups doesn't matter.
3813 PH/36 Now that the new test suite is complete, we can remove some of the
3814 special code in Exim that was needed for the old test suite. For example,
3815 sorting DNS records because real resolvers return them in an arbitrary
3816 order. The new test suite's fake resolver always returns records in the
3819 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3820 bounces) except when queue_only is set, to avoid logging races between
3821 the different processes.
3823 PH/38 Panic-die if .include specifies a non-absolute path.
3825 PH/39 A tweak to the "H" retry rule from its user.
3827 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3828 a label. They prevented compilation on older perls.
3830 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3831 a warning to be raised on newish perls.
3833 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3834 on queue. Changes to match documented behaviour of showing count of
3835 messages matching specified criteria.
3837 PH/40 Changed the default ident timeout from 30s to 5s.
3839 PH/41 Added support for the use of login_cap features, on those BSD systems
3840 that have them, for controlling the resources used by pipe deliveries.
3842 PH/42 The content-scanning code uses fopen() to create files in which to put
3843 message data. Previously it was not paying any attention to the mode of
3844 the files. Exim runs with umask(0) because the rest of the code creates
3845 files with open(), and sets the required mode explicitly. Thus, these
3846 files were ending up world-writeable. This was not a big issue, because,
3847 being within the spool directory, they were not world-accessible. I have
3848 created a function called modefopen, which takes an additional mode
3849 argument. It sets umask(777), creates the file, chmods it to the required
3850 mode, then resets the umask. All the relevant calls to fopen() in the
3851 content scanning code have been changed to use this function.
3853 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3854 to 24 hours. This avoids potential overflow problems when processing G
3855 and H retry rules. I suspect nobody ever tinkers with this value.
3857 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3859 PH/45 When the plaintext authenticator is running as a client, the server's
3860 challenges are checked to ensure they are valid base64 strings. By
3861 default, the authentication attempt is cancelled if an invalid string is
3862 received. Setting client_ignore_invalid_base64 true ignores these errors.
3863 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3864 they are received. Thus, the responses can be made to depend on the
3865 challenges. If an invalid string is ignored, an empty string is placed in
3868 PH/46 Messages that are created by the autoreply transport now contains a
3869 References: header, in accordance with RFCs 2822 and 3834.
3871 PH/47 Added authenticated_sender_force to the smtp transport.
3873 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3875 PH/49 Installed latest patch from the Sieve maintainer.
3877 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3878 was also set, the appendfile transport was unnecessarily scanning a
3879 directory of message files (e.g. for maildir delivery) to find the count
3880 of files (along with the size), even though it did not need this
3881 information. It now does the scan only if it needs to find either the
3882 size of the count of files.
3884 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3886 PH/52 Two bugs concerned with error handling when the smtp transport is
3889 (i) Exim was not creating retry information for temporary errors given
3890 for individual recipients after the DATA command when the smtp transport
3891 was used in LMTP mode. This meant that they could be retried too
3892 frequently, and not timed out correctly.
3894 (ii) Exim was setting the flag that allows error details to be returned
3895 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3896 recipients that were returned after the DATA command.
3898 PH/53 This is related to PH/52, but is more general: for any failing address,
3899 when detailed error information was permitted to be returned to the
3900 sender, but the error was temporary, then after the final timeout, only
3901 "retry timeout exceeded" was returned. Now it returns the full error as
3902 well as "retry timeout exceeded".
3904 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3905 do this, and (what is worse) MTAs that accept it.
3907 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3908 will now be deprecated.
3910 PH/56 New os.c-cygwin from the Cygwin maintainer.
3912 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3913 formats (previously only available in exim formats via -bpr, -bpru,
3914 and -bpra. Now also available in native and exiqgrep formats)
3916 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3917 with very large, slow to parse queues
3919 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3921 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3923 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3924 responses to authentication challenges, though it was showing the
3925 challenges; (ii) I've removed the CR characters from the debug output for
3928 PH/58 Allow for the insertion of a newline as well as a space when a string
3929 is turned into more than one encoded-word during RFC 2047 encoding. The
3930 Sieve code now uses this.
3932 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3933 data_4xx, lost_connection, tls_required.
3935 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3936 message was being sent as an SMTP response.
3938 PH/61 Add -l and -k options to exicyclog.
3940 PH/62 When verifying, if an address was redirected to one new address, so that
3941 verification continued, and the new address failed or deferred after
3942 having set something in $address_data, the value of $address_data was not
3943 passed back to the ACL. This was different to the case when no
3944 redirection occurred. The value is now passed back in both cases.
3946 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3947 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3948 use login_cap.h, so on its own it isn't the distinguishing feature. The
3949 new name refers directly to the setclassresources() function.
3951 PH/65 Added configuration files for NetBSD3.
3953 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3955 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3956 is preferred over IPv4.
3958 PH/68 The bounce_return_message and bounce_return_body options were not being
3959 honoured for bounces generated during the reception of non-SMTP messages.
3960 In particular, this applied to messages rejected by the ACL. This bug has
3961 been fixed. However, if bounce_return_message is true and bounce_return_
3962 body is false, the headers that are returned for a non-SMTP message
3963 include only those that have been read before the error was detected.
3964 (In the case of an ACL rejection, they have all been read.)
3966 PH/69 The HTML version of the specification is now built in a directory called
3967 spec_html instead of spec.html, because the latter looks like a path with
3968 a MIME-type, and this confuses some software.
3970 PH/70 Catch two compiler warnings in sieve.c.
3972 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3973 function verify_get_ident() calls ip_connect() to connect a socket, but
3974 if the "connect()" function timed out, ip_connect() used to close the
3975 socket. However, verify_get_ident() also closes the socket later, and in
3976 between Exim writes to the log, which may get opened at this point. When
3977 the socket was closed in ip_connect(), the log could get the same file
3978 descriptor number as the socket. This naturally causes chaos. The fix is
3979 not to close the socket in ip_connect(); the socket should be closed by
3980 the function that creates it. There was only one place in the code where
3981 this was missing, in the iplookup router, which I don't think anybody now
3982 uses, but I've fixed it anyway.
3984 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3985 well as to direct DNS lookups. Otherwise the handling of names in host
3986 lists is inconsistent and therefore confusing.
3992 PH/01 Two changes to the default runtime configuration:
3994 (1) Move the checks for relay_from_hosts and authenticated clients from
3995 after to before the (commented out) DNS black list checks.
3997 (2) Add control=submission to the relay_from_hosts and authenticated
3998 clients checks, on the grounds that messages accepted by these
3999 statements are most likely to be submissions.
4001 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
4003 (1) Generate an error if the third argument for the ${prvs expansion is
4006 (2) Treat a missing third argument of ${prvscheck as if it were an empty
4009 (3) Reset the variables that are obtained from the first argument of
4010 ${prvscheck and used in the second argument before leaving the code,
4011 because their memory is reclaimed, so using them afterwards may do
4014 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
4015 one (it's much easier than Tom thought :-).
4017 (5) Because of (4), we can now allow for the use of $prvscheck_result
4018 inside the third argument.
4020 PH/03 For some reason, the default setting of PATH when running a command from
4021 a pipe transport was just "/usr/bin". I have changed it to
4024 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
4025 anything to be listed in the output from -bV.
4027 PH/05 When a filter generated an autoreply, the entire To: header line was
4028 quoted in the delivery log line, like this:
4030 => >A.N.Other <ano@some.domain> <original@ddress> ...
4032 This has been changed so that it extracts the operative address. There
4033 may be more than one such address. If so, they are comma-separated, like
4036 => >ano@some.domain,ona@other.domain <original@ddress> ...
4038 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
4039 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
4040 not being looked up in the rDNS to get a host name, Exim was showing the
4041 IP address twice in Received: lines, even though the IP addresses were
4042 identical. For example:
4044 Received: from [1.2.3.4] (helo=[1.2.3.4])
4046 However, if the real host name was known, it was omitting the HELO data
4047 if it matched the actual IP address. This has been tidied up so that it
4048 doesn't show the same IP address twice.
4050 PH/07 When both +timestamp and +memory debugging was on, the value given by
4051 $tod_xxx expansions could be wrong, because the tod_stamp() function was
4052 called by the debug printing, thereby overwriting the timestamp buffer.
4053 Debugging no longer uses the tod_stamp() function when +timestamp is set.
4055 PH/08 When the original message was included in an autoreply transport, it
4056 always said "this is a copy of the message, including all the headers",
4057 even if body_only or headers_only was set. It now gives an appropriate
4060 PH/09 Applied a patch from the Sieve maintainer which:
4062 o fixes some comments
4063 o adds the (disabled) notify extension core
4064 o adds some debug output for the result of if/elsif tests
4065 o points to the current vacation draft in the documentation
4066 and documents the missing references header update
4070 o fixes a bug in processing the envelope test (when testing
4071 multiple envelope elements, the last element determined the
4074 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
4075 Electronic Mail") by including:
4077 Auto-submitted: auto-generated
4079 in the messages that it generates (bounce messages and others, such as
4080 warnings). In the case of bounce messages for non-SMTP messages, there was
4081 also a typo: it was using "Auto_submitted" (underscore instead of
4082 hyphen). Since every message generated by Exim is necessarily in response
4083 to another message, thes have all been changed to:
4085 Auto-Submitted: auto-replied
4087 in accordance with these statements in the RFC:
4089 The auto-replied keyword:
4091 - SHOULD be used on messages sent in direct response to another
4092 message by an automatic process,
4094 - MUST NOT be used on manually-generated messages,
4096 - MAY be used on Delivery Status Notifications (DSNs) and Message
4097 Disposition Notifications (MDNs),
4099 - MUST NOT be used on messages generated by automatic or periodic
4100 processes, except for messages which are automatic responses to
4103 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
4104 to the default Received: header definition.
4106 PH/12 Added log selector acl_warn_skipped (default on).
4108 PH/13 After a successful wildlsearch lookup, discard the values of numeric
4109 variables because (a) they are in the wrong storage pool and (b) even if
4110 they were copied, it wouldn't work properly because of the caching.
4112 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
4113 checking when decoding. Apparently there are clients that generate
4114 overlong encoded strings. Why am I not surprised?
4116 PH/15 If the first argument of "${if match_address" was not empty, but did not
4117 contain an "@" character, Exim crashed. Now it writes a panic log message
4118 and treats the condition as false.
4120 PH/16 In autoreply, treat an empty string for "once" the same as unset.
4122 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
4123 extension "envelope-auth". The code is finished and in agreement with
4124 other implementations, but there is no documentation so far and in fact,
4125 nobody wrote the draft yet. This extension is currently #undef'ed, thus
4126 not changing the active code.
4128 Print executed "if" and "elsif" statements when debugging is used. This
4129 helps a great deal to understand what a filter does.
4131 Document more things not specified clearly in RFC3028. I had all this
4132 sorted out, when out of a sudden new issues came to my mind. Oops."
4134 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
4137 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
4138 canonical form (as documented). However, after a host name lookup from
4139 the IP address, check_host() was doing a simple string comparison with
4140 addresses acquired from the DNS when checking that the found name did
4141 have the original IP as one of its addresses. Since any found IPv6
4142 addresses are likely to be in abbreviated form, the comparison could
4143 fail. Luckily, there already exists a function for doing the comparison
4144 by converting both addresses to binary, so now that is used instead of
4145 the text comparison.
4147 PH/20 There was another similar case to PH/19, when a complete host name was
4148 given in a host list; looking up its IP address could give an abbreviated
4149 form, whereas the current host's name might or might not be abbreviated.
4150 The same fix has been applied.
4156 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
4157 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
4160 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
4161 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
4163 PH/03 Typo: missing ".o" in src/pcre/Makefile.
4165 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
4166 header line, restrict the check to what is listed in RFCs 2369 and 2929.
4167 Also, for "Auto-Submitted", treat anything other than "no" as
4168 non-personal, in accordance with RFC 3834. (Previously it treated
4169 anything starting "auto-" as non-personal.)
4171 TF/01 The control=submission/name=... option had a problem with syntax
4172 errors if the name included a slash character. The /name= option
4173 now slurps the rest of the string, so it can include any characters
4174 but it must come last in the list of options (after /sender_retain
4177 PH/05 Some modifications to the interface to the fake nameserver for the new
4185 TK/01 Added the "success_on_redirect" address verification option. See
4186 NewStuff for rationale and an example.
4188 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
4190 PH/02 Patch to exigrep to allow it to work on syslog lines.
4192 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
4193 fread() to skip over the body file's header line, because in Cygwin the
4194 header line is locked and is inaccessible.
4196 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
4197 co-exist for some time) to make it clear that it is the Exim ID that is
4198 referenced, not the Message-ID: header line.
4200 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
4201 string_format() function, because snprintf() does not exist on all
4204 PH/06 The use of forbid_filter_existstest now also locks out the use of the
4205 ${stat: expansion item.
4207 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
4208 protocol synchronization error", to keep the pedants happy.
4210 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
4211 well as for IRIX systems, when gcc is being used. See the host.c source
4214 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
4216 PH/10 Named domain lists were not working if used in a queue_smtp_domains
4219 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
4220 transport and to the smtp transport in LMTP mode.
4222 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
4224 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
4225 run a filter in a subprocess. This could lead to confusion in subsequent
4226 lookups in the parent process. There should also be a search_tidyup() at
4227 the end of the subprocess.
4229 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
4230 only if the host matched helo_try_verify_hosts, which caused the
4231 verification to occur when the EHLO/HELO command was issued. The ACL just
4232 tested the remembered result. Now, if a previous verification attempt has
4233 not happened, "verify = helo" does it there and then.
4235 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
4237 TK/03 Fix log output including CR from clamd.
4239 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
4240 exist provoked a memory error which could cause a segfault.
4242 PH/15 Installed PCRE 6.2
4244 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
4246 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
4247 of the problem. Specifically, suggested +O2 rather than +O1 for the
4250 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
4252 PH/20 If a delivery was routed to a non-standard port by means of an SRV
4253 record, the port was not correctly logged when the outgoing_port log
4254 selector was set (it logged the transort's default port).
4256 PH/21 Added support for host-specific ports to manualroute, queryprogram,
4257 fallback_hosts, and "hosts" in the smtp transport.
4259 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
4260 host errors such as "Connection refused".
4262 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
4263 authentication with radiusclient 0.4.9:
4265 - Error returned from rc_read_config was caught wrongly
4266 - Username/password not passed on to radius server due to wrong length.
4268 The presumption is that some radiusclient API changes for 4.51/PH/17
4269 were not taken care of correctly. The code is still untested by me (my
4270 Linux distribution still has 0.3.2 of radiusclient), but it was
4271 contributed by a Radius user.
4273 PH/24 When doing a callout, the value of $domain wasn't set correctly when
4274 expanding the "port" option of the smtp transport.
4276 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
4277 while reading a MIME header. Thanks to Tom Hughes for a patch.
4279 PH/24 Include config.h inside local_scan.h so that configuration settings are
4282 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
4283 that in an ACL for RCPT (for example), you can examine exactly what was
4286 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
4287 commands, but it was not correctly comparing the address with the actual
4288 client host address. Thus, it would show the EHLO address in Received:
4289 header lines when this was not necessary.
4291 PH/27 Added the % operator to ${eval:}.
4293 PH/28 Exim tries to create and chdir to its spool directory when it starts;
4294 it should be ignoring failures (because with -C, for example, it has lost
4295 privilege). It wasn't ignoring creation failures other than "already
4298 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
4301 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
4302 because an input line was too long, either on its own, or by virtue of
4303 too many continuations, the temporary file was not being removed, and the
4304 return code was incorrect.
4306 PH/31 Missing "BOOL" in function definition in filtertest.c.
4308 PH/32 Applied Sieve patches from the maintainer.
4310 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
4312 PH/33 Added "verify = not_blind".
4314 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
4315 Local/Makefile (with some defaults set). These are used in built scripts
4316 such as exicyclog, but they have never been used in the exim_install
4317 script (though there are many overriding facilities there). I have
4318 arranged that the exim_install script now takes note of these two
4321 PH/35 Installed configuration files for Dragonfly.
4323 PH/36 When a locally submitted message by a trusted user did not contain a
4324 From: header, and the sender address was obtained from -f or from an SMTP
4325 MAIL command, and the trusted user did not use -F to supply a sender
4326 name, $originator_name was incorrectly used when constructing a From:
4327 header. Furthermore, $originator_name was used for submission mode
4328 messages from external hosts without From: headers in a similar way,
4329 which is clearly wrong.
4331 PH/37 Added control=suppress_local_fixups.
4333 PH/38 When log_selector = +received_sender was set, and the addition of the
4334 sender made the log line's construction buffer exactly full, or one byte
4335 less than full, an overflow happened when the terminating "\n" was
4338 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
4339 when the result of a list match is failure because a DNS lookup failed.
4341 PH/40 RM_COMMAND is now used in the building process.
4343 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
4344 the "build-* directories that it finds.
4346 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
4347 address in a domain literal was a prefix of an interface address.
4349 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
4350 when verifying a sender address, unless rewrite_headers is false.
4352 PH/44 Wrote a long comment about why errors_to addresses are verified as
4353 recipients, not senders.
4355 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
4356 the ratelimit ACL was added.
4358 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
4360 PH/46 Added extra information about PostgreSQL errors to the error string.
4362 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
4363 avoiding the need to install special zones in a real server. This is
4364 backwards compatible; if it can't find the fake resolver, it drops back.
4365 Thus, both old and new test suites can be run.
4367 TF/02 Added util/ratelimit.pl
4369 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
4370 clock is set back in time.
4372 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
4373 Candler <B.Candler@pobox.com>.
4375 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
4376 OK for addresses that are the result of redirections.
4378 PH/48 A number of further additions for the benefit of the new test suite,
4379 including a fake gethostbyname() that interfaces to the fake DNS resolver
4382 TF/06 The fix for widen_domains has also been applied to qualify_single and
4383 search_parents which are the other dnslookup options that can cause
4386 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
4389 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
4391 TF/07 Exim produced the error message "an SRV record indicated no SMTP
4392 service" if it encountered an MX record with an empty target hostname.
4393 The message is now "an MX or SRV record indicated no SMTP service".
4395 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
4396 if the DNS of the sending site is misconfigured. This is quite a
4397 common situation. This change restores the behaviour of treating a
4398 helo verification defer as a failure.
4400 PH/51 If self=fail was set on a router, the bounce message did not include the
4401 actual error message.
4407 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
4409 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
4410 command itself ended in error, the underlying message about the transport
4411 filter timeout was being overwritten with the pipe command error. Now the
4412 underlying error message should be appended to the second error message.
4414 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
4416 PH/02 Reduce the amount of output that "make" produces by default. Full output
4417 can still be requested.
4419 PH/03 The warning log line about a condition test deferring for a "warn" verb
4420 was being output only once per connection, rather than after each
4421 occurrence (because it was using the same function as for successful
4422 "warn" verbs). This seems wrong, so I have changed it.
4424 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4425 it should not have, which might have caused a crash in the right
4426 circumstances, but probably never did.
4428 PH/04 Installed a modified version of Tony Finch's patch to make submission
4429 mode fix the return path as well as the Sender: header line, and to
4430 add a /name= option so that you can make the user's friendly name appear
4433 TF/03 Added the control = fakedefer ACL modifier.
4435 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4436 Mark Lowes for thorough testing.
4438 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4440 TK/03 Merged latest SRS patch from Miles Wilton.
4442 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4443 with the definition in sysexits.h (which is #included earlier).
4444 Fortunately, Exim does not actually use EX_OK. The code used to try to
4445 preserve the sysexits.h value, by assuming that macro definitions were
4446 scanned for macro replacements. I have been disabused of this notion,
4447 so now the code just undefines EX_OK before #including unistd.h.
4449 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4450 in the smtp transport. When a block could not be written in a single
4451 write() function, the timeout was being re-applied to each part-write.
4452 This seems wrong - if the receiver was accepting one byte at a time it
4453 would take for ever. The timeout is now adjusted when this happens. It
4454 doesn't have to be particularly precise.
4456 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4457 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4459 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4460 without a domain if the check to <postmaster@domain> fails.
4462 SC/01 Eximstats: added -xls and the ability to specify output files
4463 (patch written by Frank Heydlauf).
4465 SC/02 Eximstats: use FileHandles for outputting results.
4467 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4469 SC/04 Eximstats: fixed display of large numbers with -nvr option
4471 SC/05 Eximstats: fixed merging of reports with empty tables.
4473 SC/06 Eximstats: added the -include_original_destination flag
4475 SC/07 Eximstats: removed tabs and trailing whitespace.
4477 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4479 TK/06 MBOX spool code: Add real "From " MBOX separator line
4480 so the .eml file is really in mbox format (even though
4481 most programs do not really care). Patch from Alex Miller.
4483 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4484 The latter is generated from $received_to and is only set if the
4485 message has one envelope recipient. SA can use these headers,
4486 obviously out-of-the-box. Patch from Alex Miller.
4488 PH/08 The ${def test on a variable was returning false if the variable's
4489 value was "0", contrary to what the specification has always said!
4490 The result should be true unless the variable is empty.
4492 PH/09 The syntax error of a character other than { following "${if
4493 def:variable_name" (after optional whitespace) was not being diagnosed.
4494 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4495 accidental colon was present, for example, could give incorrect results.
4497 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4498 result is used (not including appendfile, where other changes are about
4501 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4502 This involved changing a lot of size variables from int to off_t. It
4503 should work with maildirs and everything.
4505 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4506 spamd dying while we are connected to it.
4508 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4511 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4512 function for BDB 4.3.
4514 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4516 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4517 the sources for PCRE have been re-arranged and more files are now
4520 PH/15 The code I had for printing potentially long long variables in PH/11
4521 above was not the best (it lost precision). The length of off_t variables
4522 is now inspected at build time, and an appropriate printing format (%ld
4523 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4524 to be "long long int" or "long int". This is needed for the internal
4525 formatting function string_vformat().
4527 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4528 the configuration file to be ":syslog", then the script "guesses" where
4529 the logs files are, rather than using the compiled in default. In our
4530 case the guess is not the same as the compiled default, so the script
4531 suddenly stopped working when I started to use syslog. The patch checks
4532 to see if log_file_path is "". If so, it attempts to read it from exim
4533 with no configuration file to get the compiled in version, before it
4534 falls back to the previous guessing code."
4536 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4537 implementing BATV in an Exim configuration. See NewStuff for the gory
4540 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4541 Makefile that are specific to HP-UX.
4543 PH/18 If the "use_postmaster" option was set for a recipient callout together
4544 with the "random" option, the postmaster address was used as the MAIL
4545 FROM address for the random test, but not for the subsequent recipient
4546 test. It is now used for both.
4548 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4549 patch removes a few documentation additions to RFC 3028, because the
4550 latest draft now contains them. It adds the new en;ascii-case comparator
4551 and a new error check for 8bit text in MIME parts. Comparator and
4552 require names are now matched exactly. I enabled the subaddress
4553 extension, but it is not well tested yet (read: it works for me)."
4555 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4556 rework some of the code of TK/09 above to avoid the hardwired use of
4557 "%lld" and "long long". Replaced the call to snprintf() with a call to
4560 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4561 records point to non-existent hosts", "retry timeout exceeded", and
4562 "retry time not reached for any host after a long failure period".
4564 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4565 experimental DomainKeys support:
4567 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4568 (2) On an error such as an illegally used "control", the wrong name for
4569 the control was given.
4571 These problems did NOT occur unless DomainKeys support was compiled.
4573 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4575 PH/24 Added ${if match_ip condition.
4577 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4578 databases so that it will be absolutely obvious if a crash occurs in the
4579 DB library. This is a regular occurrence (often caused by mis-matched
4582 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4583 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4584 picked up on a user's system that detects such things. There doesn't seem
4585 to be a gcc warning option for this - only an attribute that has to be
4586 put on the function's prototype. It seems that in Fedora Core 4 they have
4587 set this on a number of new functions. No doubt there will be more in due
4590 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4591 specify a transport. However, if an address that was verified by such a
4592 router was the subject of a callout, Exim crashed because it tried to
4593 read the rcpt_include_affixes from the non-existent transport. Now it
4594 just assumes that the setting of that option is false. This bug was
4595 introduced by 4.51/PH/31.
4597 PH/28 Changed -d+all to exclude +memory, because that information is very
4598 rarely of interest, but it makes the output a lot bigger. People tend to
4599 do -d+all out of habit.
4601 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4602 code in os-type was giving problems when libc.so lives in lib64, like on
4605 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4606 aren't the modern standard, and it seems that some systems' include files
4607 don't always have them. Exim was already checking for some of the newer
4608 ones like T_AAAA, and defining it itself. I've added checks for all the
4609 record types that Exim uses.
4611 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4612 not automatically generating a new one, as it is supposed to. This
4613 prevented TLS from working. If the file did exist, but contained invalid
4614 data, a new version was generated, as expected. It was only the case of a
4615 non-existent file that was broken.
4617 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4618 with a change in libdomainkeys > 0.64.
4620 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4621 from DNS. If the selector record carries the flag, it now has
4622 precedence over the domain-wide flag.
4624 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4626 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4627 the use of a port name that isn't defined in /etc/services) occurred, the
4628 message was deferred as in a normal delivery, and thus remained on the
4629 spool, instead of being failed because of the mua_wrapper setting. This
4630 is now fixed, and I tidied up some of the mua_wrapper messages at the
4633 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4634 the messages in a hash of arrays rather than using individual hashes.
4635 This is a bit cleaner and results in dramatic memory savings, albeit
4636 at a slight CPU cost.
4638 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4639 as requested by Marc Sherman.
4641 SC/10 Eximstats: added histograms for user specified patterns as requested
4644 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4646 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4647 fopen() in the content-scanning modules that did not already have it.
4653 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4654 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4656 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4658 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4660 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4661 to test Sieve filters that use "vacation".
4663 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4664 that changes the way the GnuTLS parameters are stored in the cache file.
4665 The new format can be generated externally. For backward compatibility,
4666 if the data in the cache doesn't make sense, Exim assumes it has read an
4667 old-format file, and it generates new data and writes a new file. This
4668 means that you can't go back to an older release without removing the
4671 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4672 work if there are any delivery delays because "one_time" forces the
4673 parent to be marked "delivered", so its unseen clone is never tried
4674 again. For this reason, Exim now forbids the simultaneous setting of
4677 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4678 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4679 entry if you want to know the details. The fix, however, neglected to
4680 consider the case when local delivery batching is involved. The test for
4681 "previously delivered" was not happening when checking to see if an
4682 address could be batched with a previous (undelivered) one; under
4683 certain circumstances this could lead to multiple deliveries to the same
4686 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4687 in its include files, and this causes problems building Exim.
4689 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4690 header_syntax) but Exim was just ignoring anything given after a slash.
4691 In particular, this caused confusion with an attempt to use "verify =
4692 reverse_host_lookup/defer_ok". An error is now given when options are
4693 supplied for verify items that do not have them. (Maybe reverse_host_
4694 lookup should have a defer_ok option, but that's a different point.)
4696 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4697 defined by RFC 821) to 2048, because there were problems with some AUTH
4698 commands, and RFC 1869 says the size should be increased for extended
4699 SMTP commands that take arguments.
4701 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4704 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4705 "unknown" error; now it says that the functionality isn't in the binary.
4707 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4708 an address' error message when a string expansion fails (syntax or
4709 whatever). Otherwise the password may appear in the log. Following change
4710 PH/42 below, there is no longer a chance of it appearing in a bounce
4713 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4715 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4716 of its addresses was ever tried. (Bugzilla bug #2.)
4718 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4719 the result incorrectly in the debug output. (It correctly added a newline
4720 to what was transported.)
4722 TF/01 Added $received_time.
4724 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4725 commented out examples of how to interface to a virus scanner and to
4726 SpamAssassin. Also added commented examples of av_scanner and
4727 spamd_address settings.
4729 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4730 and controls are allowed in which ACLs. There were a couple of minor
4731 errors. Some of the entries in the conditions table (which is a table of
4732 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4733 negation of where the condition IS allowed.
4735 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4737 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4738 header file does not have a version number, so I've had to invent a new
4739 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4740 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4741 radiusclient), but it was contributed by a Radius user.
4743 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4744 files or directories, for OpenSSL.
4746 PH/19 When an Exim process that is running as root has to create an Exim log
4747 file, it does so in a subprocess that runs as exim:exim so as to get the
4748 ownership right at creation (otherwise, other Exim processes might see
4749 the file with the wrong ownership). There was no test for failure of this
4750 fork() call, which would lead to the process getting stuck as it waited
4751 for a non-existent subprocess. Forks do occasionally fail when resources
4752 run out. I reviewed all the other calls to fork(); they all seem to check
4755 PH/20 When checking for unexpected SMTP input at connect time (before writing
4756 the banner), Exim was not dealing correctly with a non-positive return
4757 from the read() function. If the client had disconnected by this time,
4758 the result was a log entry for a synchronization error with an empty
4759 string after "input=" when read() returned zero. If read() returned -1
4760 (an event I could not check), uninitialized data bytes were printed.
4761 There were reports of junk text (parts of files, etc) appearing after
4764 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4766 PH/22 Added support for macro redefinition, and (re)definition in between
4767 driver and ACL definitions.
4769 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4770 forgetting to use the resulting value; it was using the unexpanded value.
4772 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4773 hadn't been configured. The fix is from Juergen Kreileder, who
4774 understands it better than I do:
4776 "Here's what I see happening with three configured cyrus_sasl
4777 authenticators configured (plain, login, cram-md5):
4779 On startup auth_cyrus_sasl_init() gets called for each of these.
4780 This means three calls to sasl_listmech() without a specified mech_list.
4781 => SASL tests which mechs of all available mechs actually work
4782 => three warnings about OTP not working
4783 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4785 With the patch, sasl_listmech() also gets called three times. But now
4786 SASL's mech_list option is set to the server_mech specified in the the
4787 authenticator. Or in other words, the answer from sasl_listmech()
4788 gets limited to just the mech you're testing for (which is different
4790 => the return list contains just 'plain' or 'login', 'cram-md5' or
4791 nothing depending on the value of ob->server_mech.
4793 I've just tested the patch: Authentication still works fine,
4794 unavailable mechs specified in the exim configuration are still
4795 caught, and the auth.log warnings about OTP are gone."
4797 PH/25 When debugging is enabled, the contents of the command line are added
4798 to the debugging output, even when log_selector=+arguments is not
4801 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4802 answer is "GNU", and only if the return is "GNU/something" is the answer
4805 PH/27 $acl_verify_message is now set immediately after the failure of a
4806 verification in an ACL, and so is available in subsequent modifiers. In
4807 particular, the message can be preserved by coding like this:
4809 warn !verify = sender
4810 set acl_m0 = $acl_verify_message
4812 Previously, $acl_verify_message was set only while expanding "message"
4813 and "log_message" when a very denied access.
4815 PH/28 Modified OS/os.c-Linux with
4817 -#ifndef OS_LOAD_AVERAGE
4818 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4820 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4821 nomenclature these days.)
4823 PH/29 Installed patch from the Sieve maintainer that adds the options
4824 sieve_useraddress and sieve_subaddress to the redirect router.
4826 PH/30 In these circumstances:
4827 . Two addresses routed to the same list of hosts;
4828 . First host does not offer TLS;
4829 . First host accepts first address;
4830 . First host gives temporary error to second address;
4831 . Second host offers TLS and a TLS session is established;
4832 . Second host accepts second address.
4833 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4834 and peerdn, if requested) that were in fact used only for the second
4837 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4838 attention to any local part prefix or suffix that was matched by the
4839 router that accepted the address. It now behaves in the same way as it
4840 does for delivery: the affixes are removed from the local part unless
4841 rcpt_include_affixes is set on the transport.
4843 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4844 timeout during the DATA phase of an incoming message.
4846 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4847 applied a patch sanctioned by the Sieve maintainer.
4849 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4850 the uid or gid is negative. A case of a negative gid caused this to be
4851 noticed. The fix allows for either to be negative.
4853 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4854 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4857 PH/36 The change PH/12 above was broken. Fixed it.
4859 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4860 the grounds that routing the same address twice would always produce the
4861 same answer. This might have been true once, but it is certainly no
4862 longer true now. Routing a child address may depend on the previous
4863 routing that produced that child. Some complicated redirection strategies
4864 went wrong when messages had multiple recipients, and made Exim's
4865 behaviour dependent on the order in which the addresses were given.
4867 I have moved the duplicate checking until after the routing is complete.
4868 Exim scans the addresses that are assigned to local and remote
4869 transports, and removes any duplicates. This means that more work will be
4870 done, as duplicates will always all be routed, but duplicates are
4871 presumably rare, so I don't expect this is of any significance.
4873 For deliveries to pipes, files, and autoreplies, the duplicate checking
4874 still happens during the routing process, since they are not going to be
4877 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4878 It corrects a timeout issue with spamd. This is Ian's comment: "The
4879 background is that sometimes spamd either never reads data from a
4880 connection it has accepted, or it never writes response data. The exiscan
4881 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4882 blindly assumes that writes won't block so it may never time out."
4884 PH/39 Allow G after quota size as well as K and M.
4886 PH/40 The value set for $authenticated_id in an authenticator may not contain
4887 binary zeroes or newlines because the value is written to log lines and
4888 to spool files. There was no check on this. Now the value is run through
4889 the string_printing() function so that such characters are converted to
4890 printable escape sequences.
4892 PH/41 $message_linecount is a new variable that contains the total number of
4893 lines in the message. Compare $body_linecount, which is the count for the
4896 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4897 bounce and delay warning messages, except in certain special cases, which
4900 (a) An SMTP error message from a remote host;
4901 (b) A message specified in a :fail: redirection;
4902 (c) A message specified in a "fail" command in a system filter;
4903 (d) A message specified in a FAIL return from the queryprogram router;
4904 (e) A message specified by the cannot_route_message router option.
4906 In these cases only, Exim does include the error details in bounce and
4907 warning messages. There are also a few cases where bland messages such
4908 as "unrouteable address" or "local delivery error" are given.
4910 PH/43 $value is now also set for the "else" part of a ${run expansion.
4912 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4913 being worked on, but at least Exim now implements the latest version to
4916 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4917 process to complete was treated as a delivery failure, a timeout while
4918 writing the message to the pipe was logged, but erroneously treated as a
4919 successful delivery. Such timeouts include transport filter timeouts. For
4920 consistency with the overall process timeout, these timeouts are now
4921 treated as errors, giving rise to delivery failures by default. However,
4922 there is now a new Boolean option for the pipe transport called
4923 timeout_defer, which, if set TRUE, converts the failures into defers for
4924 both kinds of timeout. A transport filter timeout is now identified in
4927 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4928 systems where "make" and "gmake" are different, calling "gmake" at top
4929 level broke things. I've arranged for the value of $(MAKE) to be passed
4930 from the Makefile to this script so that it can call the same version of
4934 A note about Exim versions 4.44 and 4.50
4935 ----------------------------------------
4937 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4938 changes of various kinds. As a consequence, a big documentation update was
4939 needed. This delayed the release for rather longer than seemed good, especially
4940 in the light of a couple of (minor) security issues. Therefore, the changes
4941 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4942 release. So 4.44 and 4.50 are in effect two different branches that both start
4945 I have left the 4.50 change log unchanged; it contains all the changes since
4946 4.43. The change log for 4.44 is below; many of its items are identical to
4947 those for 4.50. This seems to be the most sensible way to preserve the
4948 historical information.
4954 1. Minor wording change to the doc/README.SIEVE file.
4956 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4957 computation of the current number of files was incorrect.
4959 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4960 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4963 4. Give more explanation in the error message when the command for a transport
4964 filter fails to execute.
4966 5. There are several places where Exim runs a non-Exim command in a
4967 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4968 was being done only for the command run by the queryprogram router. It is
4969 now done for all such subprocesses. The other cases are: ${run, transport
4970 filters, and the commands run by the lmtp and pipe transports.
4972 6. Added CONFIGURE_GROUP build-time option.
4974 7. Some older OS have a limit of 256 on the maximum number of file
4975 descriptors. Exim was using setrlimit() to set 1000 as a large value
4976 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4977 systems. I've change it so that if it can't get 1000, it tries for 256.
4979 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4980 was an oversight, and furthermore, ever since the addition of extra
4981 controls (e.g. 4.43/32), the checks on when to allow different forms of
4982 "control" were broken. There should now be diagnostics for all cases when a
4983 control that does not make sense is encountered.
4985 9. Added the /retain_sender option to "control=submission".
4987 10. $recipients is now available in the predata ACL (oversight).
4989 11. Tidy the search cache before the fork to do a delivery from a message
4990 received from the command line. Otherwise the child will trigger a lookup
4991 failure and thereby defer the delivery if it tries to use (for example) a
4992 cached ldap connection that the parent has called unbind on.
4994 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4995 of $address_data from the recipient verification was clobbered by the
4996 sender verification.
4998 13. The value of address_data from a sender verification is now available in
4999 $sender_address_data in subsequent conditions in the ACL statement.
5001 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
5003 15. Added a new option "connect=<time>" to callout options, to set a different
5006 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5007 was its contents. (It was OK if the option was not defined at all.)
5009 17. A "Completed" log line is now written for messages that are removed from
5010 the spool by the -Mrm option.
5012 18. New variables $sender_verify_failure and $recipient_verify_failure contain
5013 information about exactly what failed.
5015 19. Added -dd to debug only the daemon process.
5017 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
5018 handles timeouts, both on the server side and network timeouts. Renamed the
5019 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
5021 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
5022 It is now set to "smtps".
5024 22. $host_address is now set to the target address during the checking of
5025 ignore_target_hosts.
5027 23. When checking ignore_target_hosts for an ipliteral router, no host name was
5028 being passed; this would have caused $sender_host_name to have been used if
5029 matching the list had actually called for a host name (not very likely,
5030 since this list is usually IP addresses). A host name is now passed as
5033 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5034 code that specifies a non-restarting handler (typically sigaction() in
5035 modern systems) in an attempt to fix a rare and obscure crash bug.
5037 25. Narrowed the window for a race in the daemon that could cause it to ignore
5038 SIGCHLD signals. This is not a major problem, because they are used only to
5039 wake it up if nothing else does.
5041 26. A malformed maildirsize file could cause Exim to calculate negative values
5042 for the mailbox size or file count. Odd effects could occur as a result.
5043 The maildirsize information is now recalculated if the size or filecount
5046 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5047 support for a long time. Removed HAVE_SYS_VFS_H.
5049 28. Installed the latest version of exipick from John Jetmore.
5051 29. In an address list, if the pattern was not a regular expression, an empty
5052 subject address (from a bounce message) matched only if the pattern was an
5053 empty string. Non-empty patterns were not even tested. This was the wrong
5054 because it is perfectly reasonable to use an empty address as part of a
5055 database query. An empty address is now tested by patterns that are
5056 lookups. However, all the other forms of pattern expect the subject to
5057 contain a local part and a domain, and therefore, for them, an empty
5058 address still always fails if the pattern is not itself empty.
5060 30. Exim went into a mad DNS loop when attempting to do a callout where the
5061 host was specified on an smtp transport, and looking it up yielded more
5062 than one IP address.
5064 31. Re-factored the code for checking spool and log partition space into a
5065 function that finds that data and another that does the check. The former
5066 is then used to implement four new variables: $spool_space, $log_space,
5067 $spool_inodes, and $log_inodes.
5069 32. The RFC2047 encoding function was originally intended for short strings
5070 such as real names; it was not keeping to the 75-character limit for
5071 encoded words that the RFC imposes. It now respects the limit, and
5072 generates multiple encoded words if necessary. To be on the safe side, I
5073 have increased the buffer size for the ${rfc2047: expansion operator from
5076 33. It is now permitted to omit both strings after an "if" condition; if the
5077 condition is true, the result is "true". As before, when the second string
5078 is omitted, a false condition yields an empty string. This makes it less
5079 cumbersome to write custom ACL and router conditions.
5081 34. Failure to deliver a bounce message always caused it to be frozen, even if
5082 there was an errors_to setting on the router. The errors_to setting is now
5085 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5086 canonical form (fully expanded) before being placed in
5087 $sender_host_address.
5089 36. The table in the code that translates DNS record types into text (T_A to
5090 "A" for instance) was missing entries for NS and CNAME. It is just possible
5091 that this could have caused confusion if both these types were looked up
5092 for the same domain, because the text type is used as part of Exim's
5093 per-process caching. But the chance of anyone hitting this buglet seems
5096 37. The dnsdb lookup has been extended in a number of ways.
5098 (1) There is a new type, "zns", which walks up the domain tree until it
5099 finds some nameserver records. It should be used with care.
5101 (2) There is a new type, "mxh", which is like "mx" except that it returns
5102 just the host names, not the priorities.
5104 (3) It is now possible to give a list of domains (or IP addresses) to be
5105 looked up. The behaviour when one of the lookups defers can be
5106 controlled by a keyword.
5108 (4) It is now possible to specify the separator character for use when
5109 multiple records are returned.
5111 38. The dnslists ACL condition has been extended: it is now possible to supply
5112 a list of IP addresses and/or domains to be looked up in a particular DNS
5115 39. Added log_selector=+queue_time_overall.
5117 40. When running the queue in the test harness, wait just a tad after forking a
5118 delivery process, to get repeatability of debugging output.
5120 41. Include certificate and key file names in error message when GnuTLS fails
5121 to set them up, because the GnuTLS error message doesn't include the name
5122 of the failing file when there is a problem reading it.
5124 42. Allow both -bf and -bF in the same test run.
5126 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
5128 44. The "Exiscan patch" is now merged into the mainline Exim source.
5130 45. Sometimes the final signoff response after QUIT could fail to get
5131 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5132 < 0 before doing a fflush(). This bug looks as though it goes back to the
5133 introduction of TLS in release 3.20, but "sometimes" must have been rare
5134 because the tests only now provoked it.
5136 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
5137 (this can affect the format of dates).
5139 47. exim_tidydb, when checking for the continued existence of a message for
5140 which it has found a message-specific retry record, was not finding
5141 messages that were in split spool directories. Consequently, it was
5142 deleting retry records that should have stayed in existence.
5144 48. Steve fixed some bugs in eximstats.
5146 49. The SPA authentication driver was not abandoning authentication and moving
5147 on to the next authenticator when an expansion was forced to fail,
5148 contradicting the general specification for all authenticators. Instead it
5149 was generating a temporary error. It now behaves as specified.
5151 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
5152 (the order specifies the preference for clients). The order is now AES256,
5153 AES128, 3DES, ARCFOUR128.
5155 51. Small patch to Sieve code - explicitly set From: when generating an
5158 52. Exim crashed if a remote delivery caused a very long error message to be
5159 recorded - for instance if somebody sent an entire SpamAssassin report back
5160 as a large number of 550 error lines. This bug was coincidentally fixed by
5161 increasing the size of one of Exim's internal buffers (big_buffer) that
5162 happened as part of the Exiscan merge. However, to be on the safe side, I
5163 have made the code more robust (and fixed the comments that describe what
5166 53. Now that there can be additional text after "Completed" in log lines (if
5167 the queue_time_overall log selector is set), a one-byte patch to exigrep
5168 was needed to allow it to recognize "Completed" as not the last thing in
5171 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
5172 patch that reportedly fixes this has been added. I am not expert enough to
5173 create a test for it. This is what the patch creator wrote:
5175 "I found a little strange behaviour of ldap code when working with
5176 Windows 2003 AD Domain, where users was placed in more than one
5177 Organization Units. When I tried to give exim partial DN, the exit code
5178 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
5179 But simultaneously result of request was absolutely normal ldap result,
5180 so I produce this patch..."
5182 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
5183 REFERENCE, so I have modified the code to exclude the patch when that macro
5186 55. Some experimental protocols are using DNS PTR records for new purposes. The
5187 keys for these records are domain names, not reversed IP addresses. The
5188 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5189 leaves it alone. Component reversal etc. now happens only for IP addresses.
5192 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
5194 57. Double the size of the debug message buffer (to 2048) so that more of very
5195 long debug lines gets shown.
5197 58. The exicyclog utility now does better if the number of log files to keep
5198 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
5200 59. Two changes related to the smtp_active_hostname option:
5202 (1) $smtp_active_hostname is now available as a variable.
5203 (2) The default for smtp_banner uses $smtp_active_hostname instead
5204 of $primary_hostname.
5206 60. The host_aton() function is supposed to be passed a string that is known
5207 to be a valid IP address. However, in the case of IPv6 addresses, it was
5208 not checking this. This is a hostage to fortune. Exim now panics and dies
5209 if the condition is not met. A case was found where this could be provoked
5210 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5211 components; fortuitously, this particular loophole had already been fixed
5212 by change 4.50/55 above.
5214 If there are any other similar loopholes, the new check in host_aton()
5215 itself should stop them being exploited. The report I received stated that
5216 data on the command line could provoke the exploit when Exim was running as
5217 exim, but did not say which command line option was involved. All I could
5218 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5219 running as the user.
5222 61. There was a buffer overflow vulnerability in the SPA authentication code
5223 (which came originally from the Samba project). I have added a test to the
5224 spa_base64_to_bits() function which I hope fixes it.
5227 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
5228 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
5230 63. The daemon start-up calls getloadavg() while still root for those OS that
5231 need the first call to be done as root, but it missed one case: when
5232 deliver_queue_load_max is set with deliver_drop_privilege. This is
5233 necessary for the benefit of the queue runner, because there is no re-exec
5234 when deliver_drop_privilege is set.
5236 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
5237 This has been fixed.
5239 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5240 was in use, was not putting the data itself into the right store pool;
5241 consequently, it could be overwritten for a subsequent message in the same
5242 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5245 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
5247 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
5248 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
5249 the function were treating the return as a boolean value, which happened to
5250 work because 0=false and not-0=true, but is not correct code.
5252 68. The host_aton() function was not handling scoped IPv6 addresses (those
5253 with, for example, "%eth0" on the end) correctly.
5255 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
5256 negated items (that is, ~something) in unsigned ints. Some compilers
5257 apparently mutter when there is no cast.
5259 70. If an address verification called from an ACL failed, and did not produce a
5260 user-specific message (i.e. there was only a "system" message), nothing was
5261 put in $acl_verify_message. In this situation, it now puts the system
5264 71. Change 4.23/11 added synchronization checking at the start of an SMTP
5265 session; change 4.31/43 added the unwanted input to the log line - except
5266 that it did not do this in the start of session case. It now does.
5268 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
5269 This is wrong and can cause the other end to generate a synchronization
5270 error if it is another Exim or anything else that does the synchronization
5271 check. A QUIT command is no longer sent after a timeout.
5273 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
5274 during host lookups.
5276 74. The defer_ok option of callout verification was not working if it was used
5277 when verifying addresses in header lines, that is, for this case:
5279 verify = header_sender/callout=defer_ok
5281 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
5282 those file descriptors could be used for SMTP connections. If anything
5283 wrote to stderr (the example that came up was "warn" in embedded Perl), it
5284 could be sent to the SMTP client, causing chaos. The daemon now opens
5285 stdin, stdout, and stderr to /dev/null when it puts itself into the
5288 76. Arrange for output from Perl's "warn" command to be written to Exim's main
5289 log by default. The user can override this with suitable Perl magic.
5291 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
5292 the log message when discard triggers, was not working for the DATA ACL or
5293 for the non-SMTP ACL.
5295 78. Error message wording change in sieve.c.
5297 79. If smtp_accept_max_per_host was set, the number of connections could be
5298 restricted to fewer than expected, because the daemon was trying to set up
5299 a new connection before checking whether the processes handling previous
5300 connections had finished. The check for completed processes is now done
5301 earlier. On busy systems, this bug wouldn't be noticed because something
5302 else would have woken the daemon, and it would have reaped the completed
5305 80. If a message was submitted locally by a user whose login name contained one
5306 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
5307 It caused a spool format error. I have fixed the spool reading code. A
5308 related problem was that the "from" clause in the Received: line became
5309 illegal because of the space(s). It is now covered by ${quote_local_part.
5311 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
5314 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
5316 83. Patch from Sieve maintainer to fix unterminated string problem in
5317 "vacation" handling.
5319 84. Some minor changes to the Linux configuration files to help with other
5320 OS variants using glibc.
5322 85. One more patch for Sieve to update vacation handling to latest spec.
5325 ----------------------------------------------------
5326 See the note above about the 4.44 and 4.50 releases.
5327 ----------------------------------------------------
5333 1. Change 4.43/35 introduced a bug that caused file counts to be
5334 incorrectly computed when quota_filecount was set in an appendfile
5337 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5338 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5341 3. Give more explanation in the error message when the command for a transport
5342 filter fails to execute.
5344 4. There are several places where Exim runs a non-Exim command in a
5345 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5346 was being done only for the command run by the queryprogram router. It is
5347 now done for all such subprocesses. The other cases are: ${run, transport
5348 filters, and the commands run by the lmtp and pipe transports.
5350 5. Some older OS have a limit of 256 on the maximum number of file
5351 descriptors. Exim was using setrlimit() to set 1000 as a large value
5352 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5353 systems. I've change it so that if it can't get 1000, it tries for 256.
5355 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5356 was an oversight, and furthermore, ever since the addition of extra
5357 controls (e.g. 4.43/32), the checks on when to allow different forms of
5358 "control" were broken. There should now be diagnostics for all cases when a
5359 control that does not make sense is encountered.
5361 7. $recipients is now available in the predata ACL (oversight).
5363 8. Tidy the search cache before the fork to do a delivery from a message
5364 received from the command line. Otherwise the child will trigger a lookup
5365 failure and thereby defer the delivery if it tries to use (for example) a
5366 cached ldap connection that the parent has called unbind on.
5368 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5369 of $address_data from the recipient verification was clobbered by the
5370 sender verification.
5372 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5373 was its contents. (It was OK if the option was not defined at all.)
5375 11. A "Completed" log line is now written for messages that are removed from
5376 the spool by the -Mrm option.
5378 12. $host_address is now set to the target address during the checking of
5379 ignore_target_hosts.
5381 13. When checking ignore_target_hosts for an ipliteral router, no host name was
5382 being passed; this would have caused $sender_host_name to have been used if
5383 matching the list had actually called for a host name (not very likely,
5384 since this list is usually IP addresses). A host name is now passed as
5387 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5388 code that specifies a non-restarting handler (typically sigaction() in
5389 modern systems) in an attempt to fix a rare and obscure crash bug.
5391 15. Narrowed the window for a race in the daemon that could cause it to ignore
5392 SIGCHLD signals. This is not a major problem, because they are used only to
5393 wake it up if nothing else does.
5395 16. A malformed maildirsize file could cause Exim to calculate negative values
5396 for the mailbox size or file count. Odd effects could occur as a result.
5397 The maildirsize information is now recalculated if the size or filecount
5400 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5401 support for a long time. Removed HAVE_SYS_VFS_H.
5403 18. Updated exipick to current release from John Jetmore.
5405 19. Allow an empty sender to be matched against a lookup in an address list.
5406 Previously the only cases considered were a regular expression, or an
5409 20. Exim went into a mad DNS lookup loop when doing a callout where the
5410 host was specified on the transport, if the DNS lookup yielded more than
5413 21. The RFC2047 encoding function was originally intended for short strings
5414 such as real names; it was not keeping to the 75-character limit for
5415 encoded words that the RFC imposes. It now respects the limit, and
5416 generates multiple encoded words if necessary. To be on the safe side, I
5417 have increased the buffer size for the ${rfc2047: expansion operator from
5420 22. Failure to deliver a bounce message always caused it to be frozen, even if
5421 there was an errors_to setting on the router. The errors_to setting is now
5424 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5425 canonical form (fully expanded) before being placed in
5426 $sender_host_address.
5428 24. Updated eximstats to version 1.33
5430 25. Include certificate and key file names in error message when GnuTLS fails
5431 to set them up, because the GnuTLS error message doesn't include the name
5432 of the failing file when there is a problem reading it.
5434 26. Expand error message when OpenSSL has problems setting up cert/key files.
5437 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5438 (this can affect the format of dates).
5440 28. exim_tidydb, when checking for the continued existence of a message for
5441 which it has found a message-specific retry record, was not finding
5442 messages that were in split spool directories. Consequently, it was
5443 deleting retry records that should have stayed in existence.
5445 29. eximstats updated to version 1.35
5446 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5447 1.35 - bugfix such that pie charts by volume are generated correctly
5449 30. The SPA authentication driver was not abandoning authentication and moving
5450 on to the next authenticator when an expansion was forced to fail,
5451 contradicting the general specification for all authenticators. Instead it
5452 was generating a temporary error. It now behaves as specified.
5454 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5455 (the order specifies the preference for clients). The order is now AES256,
5456 AES128, 3DES, ARCFOUR128.
5458 31. Small patch to Sieve code - explicitly set From: when generating an
5461 32. Exim crashed if a remote delivery caused a very long error message to be
5462 recorded - for instance if somebody sent an entire SpamAssassin report back
5463 as a large number of 550 error lines. This bug was coincidentally fixed by
5464 increasing the size of one of Exim's internal buffers (big_buffer) that
5465 happened as part of the Exiscan merge. However, to be on the safe side, I
5466 have made the code more robust (and fixed the comments that describe what
5469 33. Some experimental protocols are using DNS PTR records for new purposes. The
5470 keys for these records are domain names, not reversed IP addresses. The
5471 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5472 leaves it alone. Component reversal etc. now happens only for IP addresses.
5475 34. The host_aton() function is supposed to be passed a string that is known
5476 to be a valid IP address. However, in the case of IPv6 addresses, it was
5477 not checking this. This is a hostage to fortune. Exim now panics and dies
5478 if the condition is not met. A case was found where this could be provoked
5479 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5480 components; fortuitously, this particular loophole had already been fixed
5481 by change 4.50/55 or 4.44/33 above.
5483 If there are any other similar loopholes, the new check in host_aton()
5484 itself should stop them being exploited. The report I received stated that
5485 data on the command line could provoke the exploit when Exim was running as
5486 exim, but did not say which command line option was involved. All I could
5487 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5488 running as the user.
5491 35. There was a buffer overflow vulnerability in the SPA authentication code
5492 (which came originally from the Samba project). I have added a test to the
5493 spa_base64_to_bits() function which I hope fixes it.
5496 36. The daemon start-up calls getloadavg() while still root for those OS that
5497 need the first call to be done as root, but it missed one case: when
5498 deliver_queue_load_max is set with deliver_drop_privilege. This is
5499 necessary for the benefit of the queue runner, because there is no re-exec
5500 when deliver_drop_privilege is set.
5502 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5503 was in use, was not putting the data itself into the right store pool;
5504 consequently, it could be overwritten for a subsequent message in the same
5505 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5508 38. Sometimes the final signoff response after QUIT could fail to get
5509 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5510 < 0 before doing a fflush(). This bug looks as though it goes back to the
5511 introduction of TLS in release 3.20, but "sometimes" must have been rare
5512 because the tests only now provoked it.
5518 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5519 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5520 Now it returns an integer. A number of calls were still expecting a T/F
5521 return. Fortuitously, in all cases, the tests worked in OK situations,
5522 which is the norm. However, things would have gone wrong on any write
5523 failures on the smtp file descriptor. This function is used when sending
5524 messages over SMTP and also when doing verify callouts.
5526 2. When Exim is called to do synchronous delivery of a locally submitted
5527 message (the -odf or -odi options), it no longer closes stderr before doing
5530 3. Implemented the mua_wrapper option.
5532 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5534 5. Implemented the functions header_remove(), header_testname(),
5535 header_add_at_position(), and receive_remove_recipient(), and exported them
5538 6. If an ACL "warn" statement specified the addition of headers, Exim already
5539 inserted X-ACL-Warn: at the start if there was no header name. However, it
5540 was not making this test for the second and subsequent header lines if
5541 there were newlines in the string. This meant that an invalid header could
5542 be inserted if Exim was badly configured.
5544 7. Allow an ACL "warn" statement to add header lines at the start or after all
5545 the Received: headers, as well as at the end.
5547 8. Added the rcpt_4xx retry error code.
5549 9. Added postmaster_mailfrom=xxx to callout verification option.
5551 10. Added mailfrom=xxxx to the callout verification option, for verify=
5554 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5555 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5557 12. Inserted some casts to stop certain compilers warning when using pointer
5558 differences as field lengths or precisions in printf-type calls (mostly
5559 affecting debugging statements).
5561 13. Added optional readline() support for -be (dynamically loaded).
5563 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5564 same clock tick as a message's arrival, so that its received time was the
5565 same as the "first fail" time on the retry record, and that message
5566 remained on the queue past the ultimate address timeout, every queue runner
5567 would try a delivery (because it was past the ultimate address timeout) but
5568 after another failure, the ultimate address timeout, which should have then
5569 bounced the address, did not kick in. This was a "< instead of <=" error;
5570 in most cases the first failure would have been in the next clock tick
5571 after the received time, and all would be well.
5573 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5574 being recognized when the domain list was tested by the match_domain
5575 condition in an expansion string.
5577 16. Added the ${str2b64: operator.
5579 17. Exim was always calling setrlimit() to set a large limit for the number of
5580 processes, without checking whether the existing limit was already
5581 adequate. (It did check for the limit on file descriptors.) Furthermore,
5582 errors from getrlimit() and setrlimit() were being ignored. Now they are
5583 logged to the main and panic logs, but Exim does carry on, to try to do its
5584 job under whatever limits there are.
5586 18. Imported PCRE 5.0.
5588 19. Trivial typo in log message " temporarily refused connection" (the leading
5591 20. If the log selector return_path_on_delivery was set and an address was
5592 redirected to /dev/null, the delivery process crashed because it assumed
5593 that a return path would always be set for a "successful" delivery. In this
5594 case, the whole delivery is bypassed as an optimization, and therefore no
5597 21. Internal re-arrangement: the function for sending a challenge and reading
5598 a response while authentication was assuming a zero-terminated challenge
5599 string. It's now changed to take a pointer and a length, to allow for
5600 binary data in such strings.
5602 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5604 23. Exim was not respecting finduser_retries when seeking the login of the
5605 uid under which it was called; it was always trying 10 times. (The default
5606 setting of finduser_retries is zero.) Also, it was sleeping after the final
5607 failure, which is pointless.
5609 24. Implemented tls_on_connect_ports.
5611 25. Implemented acl_smtp_predata.
5613 26. If the domain in control=submission is set empty, Exim assumes that the
5614 authenticated id is a complete email address when it generates From: or
5615 Sender: header lines.
5617 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5618 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5619 chown and chgrp in /bin and hostname in /usr/bin.
5621 28. Exim was keeping the "process log" file open after each use, just as it
5622 does for the main log. This opens the possibility of it remaining open for
5623 long periods when the USR1 signal hits a daemon. Occasional processlog
5624 errors were reported, that could have been caused by this. Anyway, it seems
5625 much more sensible not to leave this file open at all, so that is what now
5628 29. The long-running daemon process does not normally write to the log once it
5629 has entered its main loop, and it closes the log before doing so. This is
5630 so that log files can straightforwardly be renamed and moved. However,
5631 there are a couple of unusual error situations where the daemon does write
5632 log entries, and I had neglected to close the log afterwards.
5634 30. The text of an SMTP error response that was received during a remote
5635 delivery was being truncated at 512 bytes. This is too short for some of
5636 the long messages that one sometimes sees. I've increased the limit to
5639 31. It is now possible to make retry rules that apply only when a message has a
5640 specific sender, in particular, an empty sender.
5642 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5643 it possible to be selective about when SMTP synchronization is enforced.
5645 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5647 32. Implemented hosts_connection_nolog.
5649 33. Added an ACL for QUIT.
5651 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5654 35. Added mailbox_size and mailbox_filecount to appendfile.
5656 36. Added control = no_multiline_responses to ACLs.
5658 37. There was a bug in the logic of the code that waits for the clock to tick
5659 in the case where the clock went backwards by a substantial amount such
5660 that the microsecond fraction of "now" was more than the microsecond
5661 fraction of "then" (but the whole seconds number was less).
5663 38. Added support for the libradius Radius client library this is found on
5664 FreeBSD (previously only the radiusclient library was supported).
5670 1. When certain lookups returned multiple values in the form name=value, the
5671 quoting of the values was not always being done properly. Specifically:
5672 (a) If the value started with a double quote, but contained no whitespace,
5674 (b) If the value contained whitespace other than a space character (i.e.
5675 tabs or newlines or carriage returns) it was not quoted.
5676 This fix has been applied to the mysql and pgsql lookups by writing a
5677 separate quoting function and calling it from the lookup code. The fix
5678 should probably also be applied to nisplus, ibase and oracle lookups, but
5679 since I cannot test any of those, I have not disturbed their existing code.
5681 2. A hit in the callout cache for a specific address caused a log line with no
5682 reason for rejecting RCPT. Now it says "Previous (cached) callout
5683 verification failure".
5685 3. There was an off-by-one bug in the queryprogram router. An over-long
5686 return line was truncated at 256 instead of 255 characters, thereby
5687 overflowing its buffer with the terminating zero. As well as fixing this, I
5688 have increased the buffer size to 1024 (and made a note to document this).
5690 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5691 when Exim is waiting for an SMTP response from a remote server, Exim
5692 restarts its select() call on the socket, thereby resetting its timeout.
5693 This is not a problem when such interrupts are rare. Somebody set up a cron
5694 job to run exiwhat every 2 minutes, which is less than the normal select()
5695 timeout (5 or 10 minutes). This meant that the select() timeout never
5696 kicked in because it was always reset. I have fixed this by comparing the
5697 time when an interrupt arrives with the time at the start of the first call
5698 to select(). If more time than the timeout has elapsed, the interrupt is
5699 treated as a timeout.
5701 5. Some internal re-factoring in preparation for the addition of Sieve
5702 extensions (by MH). In particular, the "personal" test is moved to a
5703 separate function, and given an option for scanning Cc: and Bcc: (which is
5704 not set for Exim filters).
5706 6. When Exim created an email address using the login of the caller as the
5707 local part (e.g. when creating a From: or Sender: header line), it was not
5708 quoting the local part when it contained special characters such as @.
5710 7. Installed new OpenBSD configuration files.
5712 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5713 try to make them clearer.
5715 9. Callout options, other than the timeout value, were being ignored when
5716 verifying sender addresses in header lines. For example, when using
5718 verify = header_sender/callout=no_cache
5720 the cache was (incorrectly) being used.
5722 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5723 only the Cygwin environment.
5725 11. When return_path_on_delivery was set as a log selector, if different remote
5726 addresses in the same message used different return paths and parallel
5727 remote delivery occurred, the wrong values would sometimes be logged.
5728 (Whenever a remote delivery process finished, the return path value from
5729 the most recently started remote delivery process was logged.)
5731 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5732 lines when AUTH and/or TLS are in use. This is the "received protocol"
5733 field. Exim used to use "asmtp" for authenticated SMTP, without any
5734 indication (in the protocol name) for TLS use. Now it follows the RFC and
5735 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5736 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5737 names appear in log lines as well as in Received: header lines.
5739 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5740 extensions, and comparison tests, and to fix some bugs.
5742 14. Changes to the "personal" filter test:
5744 (1) The test was buggy in that it was just doing the equivalent of
5745 "contains" tests on header lines. For example, if a user's address was
5746 anne@some.where, the "personal" test would incorrectly be true for
5748 To: susanne@some.where
5750 This test is now done by extracting each address from the header in turn,
5751 and checking the entire address. Other tests that are part of "personal"
5752 are now done using regular expressions (for example, to check local parts
5753 of addresses in From: header lines).
5755 (2) The list of non-personal local parts in From: addresses has been
5756 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5757 taken from the Sieve specification recommendations.
5759 (3) If the message contains any header line starting with "List-" it is
5760 treated as non-personal.
5762 (4) The test for "circular" in the Subject: header line has been removed
5763 because it now seems ill-conceived.
5765 15. Minor typos in src/EDITME comments corrected.
5767 16. Installed latest exipick from John Jetmore.
5769 17. If headers_add on a router specified a text string that was too long for
5770 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5771 of string_sprintf() is now avoided.
5773 18. $message_body_size was not set (it was always zero) when running the DATA
5774 ACL and the local_scan() function.
5776 19. For the "mail" command in an Exim filter, no default was being set for
5777 the once_repeat time, causing a random time value to be used if "once" was
5778 specified. (If the value happened to be <= 0, no repeat happened.) The
5779 default is now 0s, meaning "never repeat". The "vacation" command was OK
5780 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5781 (I found it when inspecting the code).
5783 20. There is now an overall timeout for performing a callout verification. It
5784 defaults to 4 times the callout timeout, which applies to individual SMTP
5785 commands during the callout. The overall timeout applies when there is more
5786 than one host that can be tried. The timeout is checked before trying the
5787 next host. This prevents very long delays if there are a large number of
5788 hosts and all are timing out (e.g. when the network connections are timing
5789 out). The value of the overall timeout can be changed by specifying an
5790 additional sub-option for "callout", called "maxwait". For example:
5792 verify = sender/callout=5s,maxwait=20s
5794 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5795 to the end before writing, but this should make it even safer).
5797 22. Exim was forgetting that it had advertised PIPELINING for the second and
5798 subsequent messages on an SMTP connection. It was also not resetting its
5799 memory on STARTTLS and an internal HELO.
5801 23. When Exim logs an SMTP synchronization error within a session, it now
5802 records whether PIPELINING has been advertised or not.
5804 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5805 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5806 rather than long int.
5808 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5810 26. Added the never_mail option to autoreply.
5816 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5817 crash if the getsockname() call failed; this can happen if a connection is
5818 closed very soon after it is established. The problem was simply in the
5819 order in which certain operations were done, causing Exim to try to write
5820 to the SMTP stream before it had set up the file descriptor. The bug has
5821 been fixed by making things happen in the correct order.
5827 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5828 before the connection was closed, thus losing the rejection response.
5830 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5831 some early Solaris releases, but causes trouble in current releases where
5832 socklen_t is defined.
5834 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5837 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5840 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5841 that releasing the top of it at the end releases what was used for sub-
5842 expansions (unless the block got too big). However, discard this block if
5843 the first thing is a variable or header, so that we can use its block when
5844 it is dynamic (useful for very large $message_headers, for example).
5846 6. Lookups now cache *every* query, not just the most recent. A new, separate
5847 store pool is used for this. It can be recovered when all lookup caches are
5848 flushed. Lookups now release memory at the end of their result strings.
5849 This has involved some general refactoring of the lookup sources.
5851 7. Some code has been added to the store_xxx() functions to reduce the amount
5852 of flapping under certain conditions.
5854 8. log_incoming_interface used to affect only the <= reception log lines. Now
5855 it causes the local interface and port to be added to several more SMTP log
5856 lines, for example "SMTP connection from", and rejection lines.
5858 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5860 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5862 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5863 could be overwritten at the end of the current message (or the start of a
5864 new message if it was set in a HELO ACL). The value is now preserved for
5865 the duration of the SMTP connection.
5867 12. If a transport had a headers_rewrite setting, and a matching header line
5868 contained an unqualified address, that address was qualified, even if it
5869 did not match any rewriting rules. The underlying bug was that the values
5870 of the flags that permit the existence of unqualified sender and recipient
5871 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5872 non-local messages, and by -bnq for local messages) were not being
5873 preserved with the message after it was received.
5875 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5876 "next input=" as part of the text comprising the host identity instead of
5877 the correct text. The code was using the same buffer for two different
5878 strings. However, depending on which order the printing function evaluated
5879 its arguments, the bug did not always show up. Under Linux, for example, my
5880 test suite worked just fine.
5882 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5883 doesn't work with some older versions of Perl. It has been changed to "my",
5884 which in any case is probably the better facility to use.
5886 15. A really picky compiler found some instances of statements for creating
5887 error messages that either had too many or two few arguments for the format
5890 16. The size of the buffer for calls to the DNS resolver has been increased
5891 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5892 for addresses that have a lot of PTR records. This alleviates a problem; it
5893 does not fully solve it.
5895 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5896 buffer now truncates the list and logs the incident, which is the same
5897 action as happens when Exim is looking up a host name and its aliases.
5898 Previously in this situation something unpredictable would happen;
5899 sometimes it was "internal error: store_reset failed".
5901 18. If a server dropped the connection unexpectedly when an Exim client was
5902 using GnuTLS and trying to read a response, the client delivery process
5903 crashed while trying to generate an error log message.
5905 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5906 string, for example:
5908 warn message = H1: something\nH2: something
5910 the text was added as a single header line from Exim's point of view
5911 though it ended up OK in the delivered message. However, searching for the
5912 second and subsequent header lines using $h_h2: did not work. This has been
5913 fixed. Similarly, if a system filter added multiple headers in this way,
5914 the routers could not see them.
5916 20. Expanded the error message when iplsearch is called with an invalid key to
5917 suggest using net-iplsearch in a host list.
5919 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5920 ACLs are no longer actually imposed (and a message to that effect is
5923 22. If a "gecos" field in a passwd entry contained escaped characters, in
5924 particular, if it contained a \" sequence, Exim got it wrong when building
5925 a From: or a Sender: header from that name. A second bug also caused
5926 incorrect handling when an unquoted " was present following a character
5927 that needed quoting.
5929 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5930 was not being matched caselessly.
5932 24. Arranged for all hyphens in the exim.8 source to be escaped with
5935 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5936 properly. Recipient callout cache records were still being keyed to include
5937 the sender, even when use_sender was set false. This led to far more
5938 callouts that were necessary. The sender is no longer included in the key
5939 when use_sender is false.
5941 26. Added "control = submission" modifier to ACLs.
5943 27. Added the ${base62d: operator to decode base 62 numbers.
5945 28. dnsdb lookups can now access SRV records.
5947 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5948 the configuration file.
5950 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5951 (-v) mode. This makes the output for a verbose queue run more intelligible.
5953 31. Added a use_postmaster feature to recipient callouts.
5955 32. Added the $body_zerocount variable, containing the number of binary zero
5956 bytes in the message body.
5958 33. The time of last modification of the "new" subdirectory is now used as the
5959 "mailbox time last read" when there is a quota error for a maildir
5962 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5964 35. Added +ignore_unknown as a special item in host lists.
5966 36. Code for decoding IPv6 addresses in host lists is now included, even if
5967 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5968 address was recognized as an IP address, but was then not correctly decoded
5969 into binary, causing unexpected and incorrect effects when compared with
5976 1. Very minor rewording of debugging text in manualroute to say "list of
5977 hosts" instead of "hostlist".
5979 2. If verify=header_syntax was set, and a header line with an unqualified
5980 address (no domain) and a large number of spaces between the end of the
5981 name and the colon was received, the reception process suffered a buffer
5982 overflow, and (when I tested it) crashed. This was caused by some obsolete
5983 code that should have been removed. The fix is to remove it!
5985 3. When running in the test harness, delay a bit after writing a bounce
5986 message to get a bit more predictability in the log output.
5988 4. Added a call to search_tidyup() just before forking a reception process. In
5989 theory, someone could use a lookup in the expansion of smtp_accept_max_
5990 per_host which, without the tidyup, could leave open a database connection.
5992 5. Added the variables $recipient_data and $sender_data which get set from a
5993 lookup success in an ACL "recipients" or "senders" condition, or a router
5994 "senders" option, similar to $domain_data and $local_part_data.
5996 6. Moved the writing of debug_print from before to after the "senders" test
5999 7. Change 4.31/66 (moving the time when the Received: is generated) caused
6000 problems for message scanning, either using a data ACL, or using
6001 local_scan() because the Received: header was not generated till after they
6002 were called (in order to set the time as the time of reception completion).
6003 I have revised the way this works. The header is now generated after the
6004 body is received, but before the ACL or local_scan() are called. After they
6005 are run, the timestamp in the header is updated.
6011 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
6012 before starting a queue runner without re-exec. This happened only when
6013 deliver_drop_privilege was set or when the Exim user was set to root. The
6014 effect of the bug was that timeouts during subsequent deliveries caused
6015 crashes instead of being properly handled. The handler is now left at its
6016 default (and expected) setting.
6018 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
6019 message, again when deliver_drop_privilege is set or Exim is run as root.
6020 The bug described in (1) was not present in this case, but the tidying up
6021 of the other signals was missing. I have made the two cases consistent.
6023 3. The ignore_target_hosts setting on a manualroute router was being ignored
6024 for hosts that were looked up using the /MX notation.
6026 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
6029 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
6030 operated on the sender address. After changing the $sender_address to <>
6031 for the sender address verify, Exim was re-instated it as the original
6032 (before rewriting) address, but remembering that it had rewritten it, so it
6033 wasn't rewriting it again. This bug also had the effect of breaking the
6034 sender address verification caching when the sender address was rewritten.
6036 6. The ignore_target_hosts option was being ignored by the ipliteral router.
6037 This has been changed so that if the ip literal address matches
6038 ignore_target_hosts, the router declines.
6040 7. Added expansion conditions match_domain, match_address, and match_local_
6041 part (NOT match_host).
6043 8. The placeholder for the Received: header didn't have a length field set.
6045 9. Added code to Exim itself and to exim_lock to test for a specific race
6046 condition that could lead to file corruption when using MBX delivery. The
6047 issue is with the lockfile that is created in /tmp. If this file is removed
6048 after a process has opened it but before that process has acquired a lock,
6049 there is the potential for a second process to recreate the file and also
6050 acquire a lock. This could lead to two Exim processes writing to the file
6051 at the same time. The added code performs the same test as UW imapd; it
6052 checks after acquiring the lock that its file descriptor still refers to
6053 the same named file.
6055 10. The buffer for building added header lines was of fixed size, 8192 bytes.
6056 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
6059 11. Added the smtp_active_hostname option. If used, this will typically be made
6060 to depend on the incoming interface address. Because $interface_address is
6061 not set up until the daemon has forked a reception process, error responses
6062 that can happen earlier (such as "too many connections") no longer contain
6065 12. If an expansion in a condition on a "warn" statement fails because a lookup
6066 defers, the "warn" statement is abandoned, and the next ACL statement is
6067 processed. Previously this caused the whole ACL to be aborted.
6069 13. Added the iplsearch lookup type.
6071 14. Added ident_timeout as a log selector.
6073 15. Added tls_certificate_verified as a log selector.
6075 16. Added a global option tls_require_ciphers (compare the smtp transport
6076 option of the same name). This controls incoming TLS connections.
6078 17. I finally figured out how to make tls_require_ciphers do a similar thing
6079 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
6080 before starting the TLS session.
6082 18. Tabs are now shown as \t in -bP output.
6084 19. If the log selector return_path_on_delivery was set, Exim crashed when
6085 bouncing a message because it had too many Received: header lines.
6087 20. If two routers both had headers_remove settings, and the first one included
6088 a superfluous trailing colon, the final name in the first list and the
6089 first name in the second list were incorrectly joined into one item (with a
6090 colon in the middle).
6096 1. Added -C and -D options to the exinext utility, mainly to make it easier
6097 to include in the automated testing, but these could be helpful when
6098 multiple configurations are in use.
6100 2. The exinext utility was not formatting the output nicely when there was
6101 an alternate port involved in the retry record key, nor when there was a
6102 message id as well (for retries that were specific to a specific message
6103 and a specific host). It was also confused by IPv6 addresses, because of
6104 the additional colons they contain. I have fixed the IPv4 problem, and
6105 patched it up to do a reasonable job for IPv6.
6107 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
6108 delivery, the log line now contains "pipelined" if PIPELINING was used.
6110 4. An SMTP transport process used to panic and die if the bind() call to set
6111 an explicit outgoing interface failed. This has been changed; it is now
6112 treated in the same way as a connect() failure.
6114 5. A reference to $sender_host_name in the part of a conditional expansion
6115 that was being skipped was still causing a DNS lookup. This no longer
6118 6. The def: expansion condition was not recognizing references to header lines
6119 that used bh_ and bheader_.
6121 7. Added the _cache feature to named lists.
6123 8. The code for checking quota_filecount in the appendfile transport was
6124 allowing one more file than it should have been.
6126 9. For compatibility with Sendmail, the command line option
6134 and sets the incoming protocol and host name (for trusted callers). The
6135 host name and its colon can be omitted when only the protocol is to be set.
6136 Note the Exim already has two private options, -pd and -ps, that refer to
6137 embedded Perl. It is therefore impossible to set a protocol value of "d" or
6138 "s", but I don't think that's a major issue.
6140 10. A number of refactoring changes to the code, none of which should affect
6143 (a) The number of logging options was getting close to filling up the
6144 32-bit word that was used as a bit map. I have split them into two classes:
6145 those that are passed in the argument to log_write(), and those that are
6146 only ever tested independently outside of that function. These are now in
6147 separate 32-bit words, so there is plenty of room for expansion again.
6148 There is no change in the user interface or the logging behaviour.
6150 (b) When building, for example, log lines, the code previously used a
6151 macro that called string_cat() twice, in order to add two strings. This is
6152 not really sufficiently general. Furthermore, there was one instance where
6153 it was actually wrong because one of the argument was used twice, and in
6154 one call a function was used. (As it happened, calling the function twice
6155 did not affect the overall behaviour.) The macro has been replaced by a
6156 function that can join an arbitrary number of extra strings onto a growing
6159 (c) The code for expansion conditions now uses a table and a binary chop
6160 instead of a serial search (which was left over from when there were very
6161 few conditions). Also, it now recognizes conditions like "pam" even when
6162 the relevant support is not compiled in: a suitably worded error message is
6163 given if an attempt is made to use such a condition.
6165 11. Added ${time_interval:xxxxx}.
6167 12. A bug was causing one of the ddress fields not to be passed back correctly
6168 from remote delivery subprocesses. The field in question was not being
6169 subsequently used, so this caused to problems in practice.
6171 13. Added new log selectors queue_time and deliver_time.
6173 14. Might have fixed a bug in maildirsizefile handling that threw up
6174 "unexpected character" debug warnings, and recalculated the data
6175 unnecessarily. In any case, I expanded the warning message to give more
6178 15. Added the message "Restricted characters in address" to the statements in
6179 the default ACL that block characters like @ and % in local parts.
6181 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
6182 Three changes have been made:
6184 (a) There was a serious bug; a negative response to MAIL caused the whole
6185 recipient domain to be cached as invalid, thereby blocking all messages
6186 to all local parts at the same domain, from all senders. This bug has
6187 been fixed. The domain is no longer cached after a negative response to
6188 MAIL if the sender used is not empty.
6190 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
6193 (c) A new callout option, "use_sender" has been added for people who want
6194 the modified behaviour.
6200 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
6203 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
6204 indeed breaks things for older releases.
6206 3. Added additional logging to the case where there is a problem reading data
6207 from a filter that is running in a subprocess using a pipe, in order to
6208 try to track down a specific problem.
6210 4. Testing facility fudge: when running in the test harness and attempting
6211 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
6212 getting "No route to host". Convert this to a timeout.
6214 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
6217 6. Some OS don't have socklen_t but use size_t instead. This affects the
6218 fifth argument of getsockopt() amongst other things. This is now
6219 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
6220 can be set for individual OS. I have set it for SunOS5, OSF1, and
6221 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
6222 some earlier ones do not.
6224 7. Change 4.30/15 was not doing the test caselessly.
6226 8. The standard form for an IPv6 address literal was being rejected by address
6227 parsing in, for example, MAIL and RCPT commands. An example of this kind of
6228 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6229 this, as well as the form without the "IPv6" on the front (but only when
6230 address literals are enabled, of course).
6232 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
6234 10. Exim crashed if a message with an empty sender address specified by -f
6235 encountered a router with an errors_to setting. This could be provoked only
6236 by a command such as
6240 where an empty string was supplied; "<>" did not hit this bug.
6242 11. Installed PCRE release 4.5.
6244 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
6245 remained set. It is now erased.
6247 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
6248 times from message ids (which are base 36 rather than the normal 62).
6250 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
6251 were being counted as actual protocol errors, and logged if the log
6252 selector +smtp_protocol_error was set. One cannot be perfect in this test,
6253 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
6254 and DATA following a set of rejected RCPTs do not count as protocol errors.
6255 In other words, Exim assumes they were pipelined, though this may not
6256 actually be the case. Of course, in all cases the client gets an
6257 appropriate error code.
6259 15. If a lookup fails in an ACL condition, a message about the failure may
6260 be available; it is used if testing the ACL cannot continue, because most
6261 such messages specify what the cause of the deferral is. However, some
6262 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
6263 that caused an old message to be retained and used if a later statement
6264 caused a defer, replacing the real cause of the deferral.
6266 16. If an IP address had so many PTR records that the DNS lookup buffer
6267 was not large enough to hold them, Exim could crash while trying to process
6268 the truncated data. It now detects and logs this case.
6270 17. Further to 4.21/58, another change has been made: if (and only if) the
6271 first line of a message (the first header line) ends with CRLF, a bare LF
6272 in a subsequent header line has a space inserted after it, so as not to
6273 terminate the header.
6275 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
6276 unnecessarily, used atoi() instead of strtol(), and didn't check the
6277 termination when getting file sizes from file names by regex.
6279 19. Completely re-implemented the support for maildirsize files, in the light
6280 of a number of problems with the previous contributed implementation
6281 (4.30/29). In particular:
6283 . If the quota is zero, the maildirsize file is maintained, but no quota is
6286 . If the maildir directory does not exist, it is created before any attempt
6287 to write a maildirsize file.
6289 . The quota value in the file is just a cache; if the quota is changed in
6290 the transport, the new value overrides.
6292 . A regular expression is available for excluding directories from the
6295 20. The autoreply transport checks the characters in options that define the
6296 message's headers; it allows continued headers, but it was checking with
6297 isspace() after an embedded newline instead of explicitly looking for a
6300 21. If all the "regular" hosts to which an address was routed had passed their
6301 expiry times, and had not reached their retry times, the address was
6302 bounced, even if fallback hosts were defined. Now Exim should go on to try
6305 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
6306 equivalent code in the SMTP transport. Some hosts send humungous responses
6307 to HELO/EHLO, more than 1024 it seems.
6309 23. Refactoring: code in filter.c used (void *) for "any old type" but this
6310 gives compiler warnings in some environments. I've now done it "properly",
6313 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
6314 (because of problems with the built-in one) was declared to return uschar *
6315 instead of char *, causing compiler failure.
6317 25. Fixed a file descriptor leak when processing alias/forward files.
6319 26. Fixed a minor format string issue in dbfn.c.
6321 27. Typo in exim.c: ("dmbnz" for "dbmnz").
6323 28. If a filter file refered to $h_xxx or $message_headers, and the headers
6324 contained RFC 2047 "words", Exim's memory could, under certain conditions,
6327 29. When a sender address is verified, it is cached, to save repeating the test
6328 when there is more than one recipient in a message. However, when the
6329 verification involves a callout, it is possible for different callout
6330 options to be set for different recipients. It is too complicated to keep
6331 track of this in the cache, so now Exim always runs a verification when a
6332 callout is required, relying on the callout cache for the optimization.
6333 The overhead is duplication of the address routing, but this should not be
6336 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
6337 to be verified with callout=postmaster, and the main callout worked but the
6338 postmaster check failed, the verification correctly failed. However, if a
6339 subsequent RCPT command asked for sender verification *without* the
6340 postmaster check, incorrect caching caused this verification also to fail,
6343 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
6344 it was not caching the DNS options (qualify_single, search_parents) that
6345 were used when the lookup failed. A subsequent lookup with different
6346 options therefore always gave the same answer, though there were cases
6347 where it should not have. (Example: a "domains = !$mx_any" option on a
6348 dnslookup router: the "domains" option is always processed without any
6349 widening, but the router might have qualify_single set.) Now Exim uses the
6350 cached value only when the same options are set.
6352 32. Added John Jetmore's "exipick" utility to the distribution.
6354 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
6355 than a timeout (e.g. a certificate is required, and is not provided), an
6356 Exim server now closes the connection immediately. Previously it waited for
6357 the client to close - but if the client is SSL, it seems that they each
6358 wait for each other, leading to a delay before one of them times out.
6360 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
6361 maintained 0.8.x compatibility because I don't think many are using it, and
6362 it is clearly obsolete.
6364 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
6367 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
6368 client certificate was expired. A simple patch fixes this, though I don't
6369 understand the full logic of why the verify callback is called multiple
6372 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
6373 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
6374 which causes problems with some clients (such as the Certicom SSL Plus
6375 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
6376 disables the coutermeasure allowing Eudora to connect."
6378 38. Exim was not checking that a write() to a log file succeeded. This could
6379 lead to Bad Things if a log got too big, in particular if it hit a file
6380 size limit. Exim now panics and dies if it cannot write to a log file, just
6381 as it does if it cannot open a log file.
6383 39. Modified OS/Makefile-Linux so that it now contains
6385 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
6387 The two -D definitions ensure that Exim is compiled with large file
6388 support, which makes it possible to handle log files that are bigger than
6391 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
6392 instance) a domain was checked against a named list that involved a lookup,
6393 causing $domain_data to be set, then another domain was checked against the
6394 same list, then the first domain was re-checked, the value of $domain_data
6395 after the final check could be wrong. In particular, if the second check
6396 failed, it could be set empty. This bug probably also applied to
6399 41. The strip_trailing_dot option was not being applied to the address given
6400 with the -f command-line option.
6402 42. The code for reading a message's header from the spool was incrementing
6403 $received_count, but never initializing it. This meant that the value was
6404 incorrect (doubled) while delivering a message in the same process in which
6405 it was received. In the most common configuration of Exim, this never
6406 happens - a fresh exec is done - but it can happen when
6407 deliver_drop_privilege is set.
6409 43. When Exim logs an SMTP synchronization error - client data sent too soon -
6410 it now includes up to 150 characters of the unexpected data in the log
6413 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
6414 and building data strings. The size of both of these buffers was 10 000
6415 bytes - far larger than anybody would *ever* want, thought I. Needless to
6416 say, somebody hit the limit. I have increased the maximum line length to
6417 20 000 and the maximum data length of concatenated lines to 100 000. I have
6418 also fixed two bugs, because there was no checking on these buffers. Tsk,
6419 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
6420 buffer is too small.
6422 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6423 lsearch lookups. Now it does.
6425 46. When parsing a route_list item in a manualroute router, a fixed-length
6426 buffer was used for the list of hosts. I made this 1024 bytes long,
6427 thinking that nobody would ever have a list of hosts that long. Wrong.
6428 Somebody had a whole pile of complicated expansion conditions, and the
6429 string was silently truncated, leading to an expansion error. It turns out
6430 that it is easier to change to an unlimited length (owing to other changes
6431 that have happened since this code was originally written) than to build
6432 structure for giving a limitation error. The length of the item that
6433 expands into the list of hosts is now unlimited.
6435 47. The lsearch lookup could not handle data where the length of text line was
6436 more than 4095 characters. Such lines were truncated, leading to shortened
6437 data being returned. It should now handle lines of any length.
6439 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6440 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6443 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6444 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6445 no longer happens in comment lines. A list of replacements is now placed
6446 at the head of all of the source files, except those whose only change is
6447 to replace PERL_COMMAND in the very first #! line.
6449 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6450 messages on the queue, with a bottom-up merge sort, using code contributed
6451 by Michael Haardt. This should make operations like -bp somewhat faster on
6452 large queues. It won't affect queue runners, except when queue_run_in_order
6455 51. Installed eximstats 1.31 in the distribution.
6457 52. Added support for SRV lookups to the dnslookup router.
6459 53. If an ACL referred to $message_body or $message_body_end, the value was not
6460 reset for any messages that followed in the same SMTP session.
6462 54. The store-handling optimization for building very long strings was not
6463 differentiating between the different store pools. I don't think this
6464 actually made any difference in practice, but I've tidied it.
6466 55. While running the routers to verify a sender address, $sender_address
6467 was still set to the sender address. This is wrong, because when routing to
6468 send a bounce to the sender, it would be empty. Therefore, I have changed
6469 it so that, while verifying a sender address, $sender_address is set to <>.
6470 (There is no change to what happens when verifying a recipient address.)
6472 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6473 target A or AAAA records (if not already returned) without resetting the
6474 qualify_single or search_parents options of the DNS resolver. These are
6475 inappropriate in this case because the targets of MX and SRV records must
6476 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6477 target that, when qualified, matched something in the local domain. These
6478 two options are now turned off when doing these lookups.
6480 57. It seems that at least some releases of Reiserfs (which does not have the
6481 concept of a fixed number of inodes) returns zero and not -1 for the
6482 number of available inodes. This interacted badly with check_spool_inodes,
6483 which assumed that -1 was the "no such thing" setting. What I have done is
6484 to check that the total number of inodes is greater than zero before doing
6485 the test of how many are available.
6487 58. When a "warn" ACL statement has a log_message modifier, the message is
6488 remembered, and not repeated. This is to avoid a lot of repetition when a
6489 message has many recipients that cause the same warning to be written.
6490 However, Exim was preserving the list of already written lines for an
6491 entire SMTP session, which doesn't seem right. The memory is now reset if a
6492 new message is started.
6494 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6495 debugging output unless log_rewrite was also set.
6497 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6498 of (int)(handle) when we know that handle contains (void *)(-1).
6500 61. The Exim daemon panic-logs an error return when it closes the incoming
6501 connection. However "connection reset by peer" seems to be common, and
6502 isn't really an error worthy of noting specially, so that particular error
6505 62. When Exim is trying to find all the local interfaces, it used to panic and
6506 die if the ioctl to get the interface flags failed. However, it seems that
6507 on at least one OS (Solaris 9) it is possible to have an interface that is
6508 included in the list of interfaces, but for which you get a failure error
6509 for this call. This happens when the interface is not "plumbed" into a
6510 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6511 failure of the "get flags" call assumes that the interface is down.
6513 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6514 makes life easier for people who are doing arithmetic on fields extracted
6515 from dates, where you often get leading zeros that should not be
6516 interpreted as octal.
6518 64. Added qualify_domain to the redirect router, to override the global
6521 65. If a pathologically long header line contained very many addresses (the
6522 report of this problem mentioned 10 000) and each of them was rewritten,
6523 Exim could use up a very large amount of memory. (It kept on making new
6524 copies of the header line as it rewrote, and never released the old ones.)
6525 At the expense of a bit more processing, the header rewriting function has
6526 been changed so that it no longer eats memory in this way.
6528 66. The generation of the Received: header has been moved from the time that a
6529 message starts to be received, to the time that it finishes. The timestamp
6530 in the Received: header should now be very close to that of the <= log
6531 line. There are two side-effects of this change:
6533 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6534 logged header lines no longer include the local Received: line, because
6535 it has not yet been created. The same applies to a copy of the message
6536 that is returned to a non-SMTP sender when a message is rejected.
6538 (b) When a filter file is tested using -bf, no additional Received: header
6539 is added to the test message. After some thought, I decided that this
6542 This change does not affect the value of $received_for. It is still set
6543 after address rewriting, but before local_scan() is called.
6545 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6547 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6548 gave an unhelpful panic error message, and a defer error. I have managed to
6549 change this behaviour so that it now rejects any supplied certificate,
6550 which seems right, as the list of acceptable certificates is empty.
6552 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6553 gave an unhelpful defer error. I have not managed to make this reject any
6554 supplied certificates, but the error message it gives is "no certificate
6555 supplied", which is not helpful.
6557 70. exigrep's output now also includes lines that are not associated with any
6558 message, but which match the given pattern. Implemented by a patch from
6559 Martin Sluka, which also tidied up the Perl a bit.
6561 71. Recipient callout verification, like sender verification, was using <> in
6562 the MAIL FROM command. This isn't really the right thing, since the actual
6563 sender may affect whether the remote host accepts the recipient or not. I
6564 have changed it to use the actual sender in the callout; this means that
6565 the cache record is now keyed on a recipient/sender pair, not just the
6566 recipient address. There doesn't seem to be a real danger of callout loops,
6567 since a callout by the remote host to check the sender would use <>.
6568 [SEE ABOVE: changed after hitting problems.]
6570 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6571 temporary errors. However, in the case of such a code being given after
6572 the end of a data transmission (i.e. after ".") Exim was failing to write
6573 a retry record for the message. (Yes, there was some broken host that was
6574 actually sending 8xx at this point.)
6576 73. An unknown lookup type in a host list could cause Exim to panic-die when
6577 the list was checked. (An example that provoked this was putting <; in the
6578 middle of a list instead of at the start.) If this happened during a DATA
6579 ACL check, a -D file could be left lying around. This kind of configuration
6580 error no longer causes Exim to die; instead it causes a defer error. The
6581 incident is still logged to the main and panic logs.
6583 74. Buglet left over from Exim 3 conversion. The message "too many messages
6584 in one connection" was written to the rejectlog but not the mainlog, except
6585 when address rewriting (yes!) was being logged.
6587 75. Added write_rejectlog option.
6589 76. When a system filter was run not as root (that is, when system_filter_user
6590 was set), the values of the $n variables were not being returned to the
6591 main process; thus, they were not subsequently available in the $sn
6594 77. Added +return_path_on_delivery log selector.
6596 78. A connection timeout was being treated differently from recipients deferred
6597 when testing hosts_max_try with a message that was older than the host's
6598 retry timeout. (The host should not be counted, thus allowing all hosts to
6599 be tried at least once before bouncing.) This may have been the cause of an
6600 occasionally reported bug whereby a message would remain on the queue
6601 longer than the retry timeout, but would be bounced if a delivery was
6602 forced. I say "may" because I never totally pinned down the problem;
6603 setting up timeout/retry tests is difficult. See also the next item.
6605 79. The ultimate address timeout was not being applied to errors that involved
6606 a combination of host plus message (for example, a timeout on a MAIL
6607 command). When an address resolved to a number of possible hosts, and they
6608 were not all tried for each delivery (e.g. because of hosts_max_try), a
6609 message could remain on the queue longer than the retry timeout.
6611 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6614 81. Fixed an obscure SMTP outgoing bug which required at least the following
6615 conditions: (a) there was another message waiting for the same server;
6616 (b) the server returned 5xx to all RCPT commands in the first message so
6617 that the message was not completed; (c) the server dropped the connection
6618 or gave a negative response to the RSET that Exim sends to abort the
6619 transaction. The observed case was a dropped connection after DATA that had
6620 been sent in pipelining mode. That is, the server had advertised PIPELINING
6621 but was not implementing it correctly. The effect of the bug was incorrect
6622 behaviour, such as trying another host, and this could lead to a crash.
6628 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6629 and daemon.c were passed as pointers to ints; they should have been
6630 pointers to socklen_t variables (which are typically unsigned ints).
6632 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6635 3. Fixed a really odd bug that affected only the testing scheme; patching a
6636 certain fixed string in the binary changed the value of another string that
6637 happened to be identical to the end of the original first string.
6639 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6640 name", it returns that address as the IP address. On some operating
6641 systems (e.g. Solaris), it also passes back the IP address string as the
6642 "host name". However, on others (e.g. Linux), it passes back an empty
6643 string. Exim wasn't checking for this, and was changing the host name to an
6644 empty string, assuming it had been canonicalized.
6646 5. Although rare, it is permitted to have more than one PTR record for a given
6647 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6648 all the names associated with an address, because they do in Solaris.
6649 However, it seems that they do not in Linux for data that comes from the
6650 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6651 I found this out when I moved to a new Linux workstation and tried to run
6652 the Exim test suite.
6654 To get round this problem I have changed the code so that it now does its
6655 own call to the DNS to look up PTR records when searching for a host name.
6656 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6657 addresses that are only in /etc/hosts are still found.
6659 This behaviour is, however, controlled by an option called host_lookup_
6660 order, which defaults to "bydns:byaddr". If people want to use the other
6661 order, or indeed, just use one or the other means of lookup, they can
6662 specify it in this variable.
6664 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6665 some operating systems, this comes back from gethostbyaddr() as an empty
6666 string, and this is what Exim used to test for. However, it seems that in
6667 other systems, "." is yielded. Exim now tests for this case too.
6669 7. The values of check_spool_space and check_log_space are now held internally
6670 as a number of kilobytes instead of an absolute number of bytes. If a
6671 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6672 kilobyte. This means that much larger values can be stored.
6674 8. Exim monitor: an attempt to get the action menu when not actually pointing
6675 at a message produces an empty menu entitled "No message selected". This
6676 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6677 no entries in it ("Shell widget menu has zero width and/or height"). So I
6678 have added a single, blank menu entry in this case.
6680 9. Added ${quote_local_part.
6682 10. MIME decoding is now applied to the contents of Subject: header lines when
6685 11. Now that a reference to $sender_host_address automatically causes a reverse
6686 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6687 host lookup before query-style lookups in lists that might use this
6688 variable. This has therefore been abolished, and the "net-" prefix is no
6689 longer necessary for query-style lookups.
6691 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6692 have been a typo for LFLAGS, so it has been changed.
6694 13. The install script calls Exim with "-C /dev/null" in order to find the
6695 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6696 to be output. However, since Exim outputs its version number before the
6697 error, it didn't break the script. It just looked ugly. I fixed this by
6698 always allowing "-C /dev/null" if the caller is root.
6700 14. Ignore overlarge ACL variable number when reading spool file - insurance
6701 against a later release with more variables having written the file.
6703 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6704 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6705 this, as well as the form without the "IPv6" on the front.
6707 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6708 OS/Makefile-Darwin file.
6710 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6711 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6713 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6714 variable. However, some OS use ints for inodes. I've added cast to long int
6715 to get rid of the compiler warning.
6717 19. I had forgotten to lock out "/../" in configuration file names when
6718 ALT_CONFIG_PREFIX was set.
6720 20. Routers used for verification do not need to specify transports. However,
6721 if such a router generated a host list, and callout was configured, Exim
6722 crashed, because it could not find a port number from the (non-existent)
6723 transport. It now assumes port 25 in this circumstance.
6725 21. Added the -t option to exigrep.
6727 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6728 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6729 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6730 them is set, LOOKUP_LSEARCH is forced.
6732 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6734 24. Added sender and host information to the "rejected by local_scan()" log
6735 line; previously there was no indication of these.
6737 25. Added .include_if_exists.
6739 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6740 Linux. It turns out that not all file systems support this. Apparently some
6741 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6742 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6743 not supported on the file descriptor, is now ignored when Exim is trying to
6744 sync a directory. This applies only to Linux.
6746 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6748 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6749 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6752 29. Added support for maildirsize files from supplied patch (modified a bit).
6754 30. The use of :fail: followed by an empty string could lead Exim to respond to
6755 sender verification failures with (e.g.):
6757 550 Verification failed for <xxx>
6758 550 Sender verify failed
6760 where the first response line was missing the '-' that indicates it is not
6761 the final line of the response.
6763 31. The loop for finding the name of the user that called Exim had a hardwired
6764 limit of 10; it now uses the value of finduser_retries, which is used for
6765 all other user lookups.
6767 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6770 33. Exim was neglecting to zero errno before one call of strtol() when
6771 expanding a string and expecting an integer value. On some systems this
6772 resulted in spurious "integer overflow" errors. Also, it was casting the
6773 result into an int without checking.
6775 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6776 did not work. The code looks as if it has *never* worked, though it appears
6777 to have been documented since at least release 1.62. I have made it work.
6779 35. The "timeout_DNS" error in retry rules, also documented since at least
6780 1.62, also never worked. As it isn't clear exactly what this means, and
6781 clearly it isn't a major issue, I have abolished the feature by treating it
6782 as "timeout", and writing a warning to the main and panic logs.
6784 36. The display of retry rules for -brt wasn't always showing the error code
6787 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6788 timeout_connect_A, timeout_connect_MX.
6790 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6791 to the empty sender.
6793 39. The daemon was not analysing the content of -oX till after it had closed
6794 stderr and disconnected from the controlling terminal. This meant that any
6795 syntax errors were only noted on the panic log, and the return code from
6796 the command was 0. By re-arranging the code a little, I've made the
6797 decoding happen first, so such errors now appear on stderr, and the return
6798 code is 1. However, the actual setting up of the sockets still happens in
6799 the disconnected process, so errors there are still only recorded on the
6802 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6803 connections (as happens on some IP stacks) was logged at start up time as
6804 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6805 it from "IPv6 and IPv4", which means that two separate sockets are being
6808 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6809 says whether AF_INET or AF_INET6 was passed as an argument.
6811 42. Exiwhat output was messed up when time zones were included in log
6814 43. Exiwhat now gives more information about the daemon's listening ports,
6815 and whether -tls-on-connect was used.
6817 44. The "port" option of the smtp transport is now expanded.
6819 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6820 silently ignored. Now an error message is written to the main and panic
6823 46. There's a new ACL modifier called "logwrite" which writes to a log file
6824 as soon as it is encountered.
6826 47. Added $local_user_uid and $local_user_gid at routing time.
6828 48. Exim crashed when trying to verify a sender address that was being
6831 49. Exim was recognizing only a space character after ".include". It now also
6832 recognizes a tab character.
6834 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6835 extracting the relevant information from the specification. The man page no
6836 longer contains scrambled data for the -d option, and I've added a section
6837 at the front about calling Exim under different names.
6839 51. Added "extra_headers" argument to the "mail" command in filter files.
6841 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6844 53. Installed eximstats 1.29.
6846 54. Added transport_filter_timeout as a generic transport option.
6848 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6849 Cc: header lines. This was required by RFC 822, but it not required by RFC
6852 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6853 incoming messages that did not have them. Now it does so only if the
6854 message originates locally, that is, if there is no associated remote host
6855 address. When Resent- header lines are present, this applies to the Resent-
6856 lines rather than the non-Resent- lines.
6858 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6859 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6861 58. Messages for configuration errors now include the name of the main
6862 configuration file - useful now that there may be more than one file in a
6863 list (.included file names were always shown).
6865 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6866 for those rare installations that do not start the daemon as root or run it
6867 setuid root. I've cut out the call to initgroups() if the daemon is not
6870 60. The Exim user and group can now be bound into the binary as text strings
6871 that are looked up at the start of Exim's processing.
6873 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6875 62. Added $mailstore_basename variable.
6877 63. Installed patch to sieve.c from Michael Haardt.
6879 64. When Exim failed to open the panic log after failing to open the main log,
6880 the original message it was trying to log was written to stderr and debug
6881 output, but if they were not available (the usual case in production), it
6882 was lost. Now it is written to syslog before the two lines that record the
6883 failures to open the logs.
6885 65. Users' Exim filters run in subprocesses under the user's uid. It is
6886 possible for a "deliver" command or an alias in a "personal" command to
6887 provoke an address rewrite. If logging of address rewriting is configured,
6888 this fails because the process is not running as root or exim. There may be
6889 a better way of dealing with this, but for the moment (because 4.30 needs
6890 to be released), I have disabled address rewrite logging when running a
6891 filter in a non-root, non-exim process.
6897 1. The buildconfig auxiliary program wasn't quoting the value set for
6898 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6899 not defined. This bug was masked in 4.22 by the effect that was fixed in
6902 2. Some messages that were rejected after a message id was allocated were
6903 shown as "incomplete" by exigrep. It no longer does this for messages that
6904 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6906 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6907 have allow_domain_literals set, the ID did not get logged in the <= line.
6908 Domain literals are now always recognized in Message-ID: header lines.
6910 4. The first argument for a ${extract expansion item is the key name or field
6911 number. Leading and trailing spaces in this item were not being ignored,
6912 causing some misleading effects.
6914 5. When deliver_drop_privilege was set, single queue runner processes started
6915 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6916 same command in the process it spins off) were not dropping privilege.
6918 6. When the daemon running as "exim" started a queue runner, it always
6919 re-executed Exim in the spun-off process. This is a waste of effort when
6920 deliver_drop_privilege is set. The new process now just calls the
6921 queue-runner function directly.
6927 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6930 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6931 ignored. Though the use of -oP was forcing the writing of a pid file, it
6932 was always written to the default place.
6934 3. If the message "no IP address found for host xxxx" is generated during
6935 incoming verification, it is now followed by identification of the incoming
6936 connection (so you can more easily find what provoked it).
6938 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6940 5. Added some features to "harden" Exim a bit more against certain attacks:
6942 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6943 be put in Local/Makefile. This is like the never_users runtime option,
6944 but it cannot be overridden. The default setting is "root".
6946 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6947 prefix string with which any file named in a -C command line option
6950 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6951 is retained for -C and -D only if the caller of Exim is root. Without
6952 it, the exim user may also use -C and -D and retain privilege.
6954 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6955 command line option is disabled.
6957 6. Macro names set by the -D option must start with an upper case letter, just
6958 like macro names defined in the configuration file.
6960 7. Added "dereference=" facility to LDAP.
6962 8. Two instances of the typo "uknown" in the source files are fixed.
6964 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6965 the Configure-Makefile script screwed up while processing it.
6967 10. Incorporated PCRE 4.4.
6969 11. The SMTP synchronization check was not operating right at the start of an
6970 SMTP session. For example, it could not catch a HELO sent before the client
6971 waited for the greeting. There is now a check for outstanding input at the
6972 point when the greeting is written. Because of the duplex, asynchronous
6973 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6974 way, but not yet received, when the check is performed.
6976 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6977 on TCP/IP sockets, because this apparently causes some broken clients to
6980 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6981 unchanged) from the Cygwin maintainer.
6983 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6984 was supported instead of testing for mbx. Effectively a typo.
6986 15. The spa authenticator server code was not checking that the input it
6987 received was valid base64.
6989 16. The debug output line for the "set" modifier in ACLs was not showing the
6990 name of the variable that was being set.
6992 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6994 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6995 lookup on its own. Something else was needed to trigger the lookup. For
6996 example, a match in host_lookup or the need for a host name in a host list.
6997 Now, if $sender_host_name is referenced and the host name has not yet been
6998 looked up, a lookup is performed. If the lookup fails, the variable remains
6999 empty, and $host_lookup_failed is set to "1".
7001 19. Added "eqi" as a case-independent comparison operator.
7003 20. The saslauthd authentication condition could segfault if neither service
7004 nor realm was specified.
7006 21. If an overflowing value such as "2048M" was set for message_size_limit, the
7007 error message that was logged was misleading, and incoming SMTP
7008 connections were dropped. The message is now more accurate, and temporary
7009 errors are given to SMTP connections.
7011 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
7012 (except RSET) with a 421 error, until QUIT is received. However, it was
7013 failing to send a response to QUIT.
7015 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
7016 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
7017 ACL is now run after the helo_try_verify_hosts code.
7019 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
7020 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
7021 case-independent, so other case variants are also recognized). Apparently
7022 some systems use these upper case variants.
7024 25. If more than two messages were waiting for the same host, and a transport
7025 filter was specified for the transport, Exim sent two messages over the
7026 same TCP/IP connection, and then failed with "socket operation on non-
7027 socket" when it tried to send the third.
7029 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
7031 27. The extern definition of crypt16() in expand.c was not being excluded when
7032 the OS had its own crypt16() function.
7034 28. Added bounce_return_body as a new option, and bounce_return_size_limit
7035 as a preferred synonym for return_size_limit, both as an option and as an
7038 29. Added LIBS=-liconv to OS/Makefile-OSF1.
7040 30. Changed the default configuration ACL to relax the local part checking rule
7041 for addresses that are not in any local domains. For these addresses,
7042 slashes and pipe symbols are allowed within local parts, but the sequence
7043 /../ is explicitly forbidden.
7045 31. SPA server authentication was not clearing the challenge buffer before
7048 32. log_message in a "warn" ACL statement was writing to the reject log as
7049 well as to the main log, which contradicts the documentation and doesn't
7050 seem right (because no rejection is happening). So I have stopped it.
7052 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
7053 However, I am unable to do any testing of this.
7055 34. Fixed an infelicity in the appendfile transport. When checking directories
7056 for a mailbox, to see if any needed to be created, it was accidentally
7057 using path names with one or more superfluous leading slashes; tracing
7058 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
7060 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
7061 contents are added to the log line that is written for every discarded
7062 recipient. (Previously a log_message setting was ignored.)
7064 36. The ${quote: operator now quotes the string if it is empty.
7066 37. The install script runs exim in order to find its version number. If for
7067 some reason other than non-existence or emptiness, which it checks, it
7068 could not run './exim', it was installing it with an empty version number,
7069 i.e. as "exim-". This error state is now caught, and the installation is
7072 38. An argument was missing from the function that creates an error message
7073 when Exim fails to connect to the socket for saslauthd authentication.
7074 This could cause Exim to crash, or give a corrupted message.
7076 39. Added isip, isip4, and isip6 to ${if conditions.
7078 40. The ACL variables $acl_xx are now saved with the message, and can be
7079 accessed later in routers, transports, and filters.
7081 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
7082 strings in the file are not string-expanded.
7084 42. If a MAIL command specified a SIZE value that was too large to fit into an
7085 int variable, the check against message_size_limit failed. Such values are
7086 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
7087 day this will have to be increased, but I don't think I want to be around
7088 when emails are that large.
7095 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
7096 iconv() is not standard in FreeBSD.
7098 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
7099 IPv6 enabled. The observed symptom was a segmentation fault on return from
7100 the function os_common_find_running_interfaces() in src/os.c.
7102 3. In the check_special_case() function in daemon.c I had used "errno" as an
7103 argument name, which causes warnings on some systems. This was basically a
7104 typo, since it was named "eno" in the comments!
7106 4. The code that waits for the clock to tick (at a resolution of some fraction
7107 of a second) so as to ensure message-id uniqueness was always waiting for
7108 at least one whole tick, when it could have waited for less. [This is
7109 almost certainly not relevant at current processor speeds, where it is
7110 unlikely to ever wait at all. But we try to future-proof.]
7112 5. The function that sleeps for a time interval that includes fractions of a
7113 second contained a race. It did not block SIGALRM between setting the
7114 timer, and suspending (a couple of lines later). If the interval was short
7115 and the sigsuspend() was delayed until after it had expired, the suspension
7116 never ended. On busy systems this could lead to processes getting stuck for
7119 6. Some uncommon configurations may cause a lookup to happen in a queue runner
7120 process, before it forks any delivery processes. The open lookup caching
7121 mechanism meant that the open file or database connection was passed into
7122 the delivery process. The problem was that delivery processes always tidy
7123 up cached lookup data. This could cause a problem for the next delivery
7124 process started by the queue runner, because the external queue runner
7125 process does not know about the closure. So the next delivery process
7126 still has data in the lookup cache. In the case of a file lookup, there was
7127 no problem because closing a file descriptor in a subprocess doesn't affect
7128 the parent. However, if the lookup was caching a connection to a database,
7129 the connection was closed, and the second delivery process was likely to
7130 see errors such as "PGSQL: query failed: server closed the connection
7131 unexpectedly". The problem has been fixed by closing all cached lookups
7132 in a queue runner before running a delivery process.
7134 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
7135 seem to have the "const" qualifier which it has on other OS. I've
7138 8. Change 4.21/2 was too strict. It is only if there are two authenticators
7139 *of the same type* (client or server) with the same public name that an
7140 error should be diagnosed.
7142 9. When Exim looked up a host name for an IP address, but failed to find the
7143 original IP address when looking up the host name (a safety check), it
7144 output the message "<ip address> does not match any IP for NULL", which was
7145 confusing, to say the least. The bug was that the host name should have
7146 appeared instead of "NULL".
7148 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
7149 user that is built into the binary, and the -C or -D options is used, root
7150 privilege is dropped before the configuration file is read. In addition,
7151 logging is switched to stderr instead of the normal log files. If the
7152 configuration then re-defines the Exim user, the unprivileged environment
7153 is probably not what is expected, so Exim logs a panic warning message (but
7156 However, if deliver_drop_privilege is set, the unprivileged state may well
7157 be exactly what is intended, so the warning has been cut out in that case,
7158 and Exim is allowed to try to write to its normal log files.
7164 1. smtp_return_error_details was not giving details for temporary sender
7165 or receiver verification errors.
7167 2. Diagnose a configuration error if two authenticators have the same public
7170 3. Exim used not to create the message log file for a message until the first
7171 delivery attempt. This could be confusing when incoming messages were held
7172 for policy or load reasons. The message log file is now created at the time
7173 the message is received, and an initial "Received" line is written to it.
7175 4. The automatically generated man page for command line options had a minor
7176 bug that caused no ill effects; however, a more serious problem was that
7177 the procedure for building the man page automatically didn't always
7178 operate. Consequently, release 4.20 contains an out-of-date version. This
7179 shouldn't happen again.
7181 5. When building Exim with embedded Perl support, the script that builds the
7182 Makefile was calling 'perl' to find its compile-time parameters, ignoring
7183 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
7185 6. The freeze_tell option was not being used for messages that were frozen on
7186 arrival, either by an ACL or by local_scan().
7188 7. Added the smtp_incomplete_transaction log selector.
7190 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
7191 was accepting AUTH without a new EHLO.
7193 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
7194 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
7197 10. Logging of TCP/IP connections (when configured) now happens in the main
7198 daemon process instead of the child process, so that the TCP/IP connection
7199 count is more accurate (but it can never be perfect).
7201 11. The use of "drop" in a nested ACL was not being handled correctly in the
7202 outer ACL. Now, if condition failure induced by the nested "drop" causes
7203 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
7204 or "require"), the connection is dropped.
7206 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
7207 that yield "discard" can now be used with an "accept" or a "discard" verb,
7208 but an error is generated for any others (because I can't see a useful way
7209 to define what should happen).
7211 13. When an ACL is read dynamically from a file (or anywhere else), the lines
7212 are now processed in the same way as lines in the Exim configuration file.
7213 In particular, continuation lines are supported.
7215 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
7217 15. Added -ti meaning -t -i.
7219 16. Check for letters, digits, hyphens, and dots in the names of dnslist
7220 domains, and warn by logging if others are found.
7222 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
7223 returned from GIFCONF when Exim is trying to find the list of interfaces on
7224 a host. The code in os.c has been modified to copy each ifreq to an aligned
7225 structure in all cases.
7227 Also, in some cases, the returned ifreq's were being copied to a 'struct
7228 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
7229 means the last couple of bytes of an IPv6 address could be chopped if the
7230 ifreq contained only a normal sockaddr (14 bytes storage).
7232 18. Named domain lists were not supported in the hosts_treat_as_local option.
7233 An entry such as +xxxx was not recognized, and was treated as a literal
7236 19. Ensure that header lines added by a DATA ACL are included in the reject log
7237 if the ACL subsequently rejects the message.
7239 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
7240 MD5 (which is deprecated).
7242 21. When testing a filter file using -bf, Exim was writing a message when it
7243 took the sender from a "From " line in the message, but it was not doing so
7244 when it took $return_path from a Return-Path: header line. It now does.
7246 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
7247 with a valid header line field name (a series of printing characters
7248 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
7250 23. Changed "disc" in the source to "disk" to conform to the documentation and
7251 the book and for uniformity.
7253 24. Ignore Sendmail's -Ooption=value command line item.
7255 25. When execve() failed while trying to run a command in a pipe transport,
7256 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
7257 could be confused with a return value of 69 from the command itself. This
7258 has been changed to 127, the value the shell returns if it is asked to run
7259 a non-existent command. The wording for the related log line suggests a
7260 non-existent command as the problem.
7262 26. If received_header_text expands to an empty string, do not add a Received:
7263 header line to the message. (Well, it adds a token one on the spool, but
7264 marks it "old" so that it doesn't get used or transmitted.)
7266 27. Installed eximstats 1.28 (addition of -nt option).
7268 28. There was no check for failure on the call to getsockname() in the daemon
7269 code. This can fail if there is a shortage of resources on the system, with
7270 ENOMEM, for example. A temporary error is now given on failure.
7272 29. Contrary to the C standard, it seems that in some environments, the
7273 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
7274 program. Exim now does this explicitly; it affects the formatting of
7275 timestamps using strftime().
7277 30. If exiqsumm was given junk data, it threw up some uninitialized variable
7278 complaints. I've now initialized all the variables, to avoid this.
7280 32. Header lines added by a system filter were not being "seen" during
7281 transport-time rewrites.
7283 33. The info_callback() function passed to OpenSSL is set up with type void
7284 (*)(SSL *, int, int), as described somewhere. However, when calling the
7285 function (actually a macro) that sets it up, the type void(*)() is
7286 expected. I've put in a cast to prevent warnings from picky compilers.
7288 34. If a DNS black list lookup found a CNAME record, but there were no A
7289 records associated with the domain it pointed at, Exim crashed.
7291 35. If a DNS black list lookup returned more than one A record, Exim ignored
7292 all but the first. It now scans all returned addresses if a particular IP
7293 value is being sought. In this situation, the contents of the
7294 $dnslist_value variable are a list of all the addresses, separated by a
7297 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
7298 to accept a host name and all its aliases if the forward lookup for any of
7299 them yielded the IP address of the incoming connection. Now it accepts only
7300 those names whose forward lookup yields the correct IP address. Any other
7301 names are discarded. This closes a loophole whereby a rogue DNS
7302 administrator could create reverse DNS records to break through a
7303 wildcarded host restriction in an ACL.
7305 37. If a user filter or a system filter that ran in a subprocess used any of
7306 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
7307 the wrong values were passed to the pipe command ($thisaddress had the
7308 value of $0, $0 had the value of $1, etc). This bug was introduced by
7309 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
7311 38. Improved the line breaking for long SMTP error messages from ACLs.
7312 Previously, if there was no break point between 40 and 75 characters, Exim
7313 left the rest of the message alone. Two changes have been made: (a) I've
7314 reduced the minimum length to 35 characters; (b) if it can't find a break
7315 point between 35 and 75 characters, it looks ahead and uses the first one
7316 that it finds. This may give the occasional overlong line, but at least the
7317 remaining text gets split now.
7319 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
7320 file descriptors might be low, and that setting 1000 would always raise it.
7321 It turns out that in some environments, the limit is already over 1000 and
7322 that lowering it causes trouble. So now Exim takes care not to decrease it.
7324 40. When delivering a message, the value of $return_path is set to $sender_
7325 address at the start of routing (routers may change the value). By an
7326 oversight, this default was not being set up when an address was tested by
7327 -bt or -bv, which affected the outcome if any router or filter referred to
7330 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
7331 log only when "message" or "log_message" are set. However, if one of the
7332 conditions was an address verification, or a call to a nested ACL, the
7333 messages generated by the underlying test were being passed through. This
7334 no longer happens. The underlying message is available in $acl_verify_
7335 message for both "message" and "log_message" expansions, so it can be
7336 passed through if needed.
7338 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
7339 new expansion $bh_ to give the encoded byte string without charset
7340 translation. Translation happens only if iconv() is available; HAVE_ICONV
7341 indicates this at build time. HEADERS_CHARSET gives the charset to
7342 translate to; headers_charset can change it in the configuration, and
7343 "headers charset" can change it in an individual filter file.
7345 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
7346 that creates RFC 2047 encoded "words" labels them as that charset instead
7347 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
7348 expansion operator; (ii) when Exim creates a From: line for a local
7349 message; (iii) when a header line is rewritten to include a "phrase" part.
7351 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
7352 buggy, causing it to skip the first lines of messages whose message ID
7353 ended in 'D'. This would not have bitten before Exim release 4.14, because
7354 message IDs were unlikely to end in 'D' before then. The effect was to have
7355 incorrect size information for certain domains.
7357 45. #include "config.h" was missing at the start of the crypt16.c module. This
7358 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
7361 46. If there was a timeout during a "random" callout check, Exim treated it as
7362 a failure of the random address, and carried on sending RSET and the real
7363 address. If the delay was just some slowness somewhere, the response to the
7364 original RCPT would be taken as a response to RSET and so on, causing
7365 mayhem of various kinds.
7367 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
7368 when I implemented it. It didn't allow for the fact that some option values
7369 may legitimately be negative (e.g. size_addition), and it didn't even do
7370 the right test for positive values.
7372 48. Domain names in DNS records are case-independent. Exim always looks them up
7373 in lower case. Some resolvers return domain names in exactly the case they
7374 appear in the zone file, that is, they may contain uppercase letters. Not
7375 all resolvers do this - some return always lower case. Exim was treating a
7376 change of case by a resolver as a change of domain, similar to a widening
7377 of a domain abbreviation. This triggered its re-routing code and so it was
7378 trying to route what was effectively the same domain again. This normally
7379 caused routing to fail (because the router wouldn't handle the domain
7380 twice). Now Exim checks for this case specially, and just changes the
7381 casing of the domain that it ultimately uses when it transmits the message
7384 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
7387 50. If a filter generated a file delivery with a non-absolute name (possible if
7388 no home directory exists for the router), the forbid_file option was not
7391 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
7392 the existing equality matching.
7394 52. Exim was using ints instead of ino_t variables in some places where it was
7395 dealing with inode numbers.
7397 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
7398 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
7399 called TMPDIR, and if it finds it is different, it changes its value.
7401 54. The smtp_printf() function is now made available to local_scan() so
7402 additional output lines can be written before returning. There is also an
7403 smtp_fflush() function to enable the detection of a dropped connection.
7404 The variables smtp_input and smtp_batched_input are exported to
7407 55. Changed the default runtime configuration: the message "Unknown user"
7408 has been removed from the ACL, and instead placed on the localuser router,
7409 using the cannot_route_message feature. This means that any verification
7410 failures that generate their own messages won't get overridden. Similarly,
7411 the "Unrouteable address" message that was in the ACL for unverifiable
7412 relay addresses has also been removed.
7414 56. Added hosts_avoid_esmtp to the smtp transport.
7416 57. The exicyclog script was not checking for the esoteric option
7417 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
7418 will work only if exicyclog is run under the appropriate euid.
7420 58. Following a discussion on the list, the rules by which Exim recognises line
7421 endings on incoming messages have been changed. The -dropcr and drop_cr
7422 options are now no-ops, retained only for backwards compatibility. The
7423 following line terminators are recognized: LF CRLF CR. However, special
7424 processing applies to CR:
7426 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7427 nor a local message in the state where . is a terminator.
7429 (ii) If a bare CR is encountered in a header line, an extra space is added
7430 after the line terminator so as not to end the header. The reasoning
7431 behind this is that bare CRs in header lines are most likely either
7432 to be mistakes, or people trying to play silly games.
7434 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7435 was being incorrectly given as 18 bytes larger than it should have been.
7436 This is a VOB (very old bug).
7438 60. This may never have affected anything current, but just in case it has:
7439 When the local host is found other than at the start of a list of hosts,
7440 the local host, those with the same MX, and any that follow, are discarded.
7441 When the list in question was part of a longer list of hosts, the following
7442 hosts (not currently being processed) were also being discarded. This no
7443 longer happens. I'm not sure if this situation could ever has previously
7446 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7449 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7450 header. This is something that is recommended in a new Internet Draft, and
7451 is something that is documented as being done by Sendmail. There are two
7452 possible values. For messages generated by the autoreply transport, Exim
7455 Auto-Submitted: auto-replied
7457 whereas for all other generated messages (e.g. bounces) it adds
7459 Auto-Submitted: auto-generated
7461 63. The "personal" condition in filters now includes a test for the
7462 Auto-Submitted: header. If it contains the string "auto-" the message it
7463 not considered personal.
7465 64. Added rcpt_include_affixes as a generic transport option.
7467 65. Added queue_only_override (default true).
7469 66. Added the syslog_duplication option.
7471 67. If what should have been the first header line of a message consisted of
7472 a space followed by a colon, Exim was mis-interpreting it as a header line.
7473 It isn't of course - it is syntactically invalid and should therefore be
7474 treated as the start of the message body. The misbehaviour could have
7475 caused a number of strange effects, including loss of data in subsequent
7476 header lines, and spool format errors.
7478 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7479 client host had authenticated. This control can now be exercised by an ACL
7480 for more flexibility.
7482 69. By default, callouts do not happen when testing with -bh. There is now a
7483 variant, -bhc, which does actually run the callout code, including
7484 consulting and updating the callout cache.
7486 70. Added support for saslauthd authentication, courtesy of Alexander
7489 71. If statvfs() failed on the spool or log directories while checking their
7490 size for availability, Exim confusingly gave the error "space shortage".
7491 Furthermore, in debugging mode it crashed with a floating point exception.
7492 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7493 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7494 serious problem, Exim now writes to the main and panic logs when this
7495 happens, with details of the failure. It then refuses to accept the
7496 incoming message, giving the message "spool directory problem" or "log
7497 directory problem" with a 421 code for SMTP messages.
7499 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7500 0, 1, and 2 exist, because some OS complain for execs without them (see
7501 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7502 descriptors. However, the code omitted to check that the open succeeded,
7503 causing mysterious errors if for some reason the permissions on /dev/null
7504 got screwed. Now Exim writes a message to the main and panic logs, and
7505 bombs out if it can't open /dev/null.
7507 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7508 interact so that it is all more flexible. It is supposed to remain
7509 backwards compatible. Also added extra_local_interfaces.
7511 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7512 to bomb out with an assertion failure - to the client this appears as a
7513 connection drop. This problem occurs in the part of the code that was taken
7514 from the Samba project. Fortunately, the assertion is in a very simple
7515 function, so I have fixed this by reproducing the function inline in the
7516 one place where it is called, and arranging for authentication to fail
7517 instead of killing the process with assert().
7519 75. The SPA client code was not working when the server requested OEM rather
7520 than Unicode encoding.
7522 76. Added code to make require_files with a specific uid setting more usable in
7523 the case where statting the file as root fails - usually a non-root-mounted
7524 NFS file system. When this happens and the failure is EACCES, Exim now
7525 forks a subprocess and does the per-uid checking as the relevant uid.
7527 77. Added process_log_path.
7529 78. If log_file_path was not explicitly set, a setting of check_log_space or
7530 check_log_inodes was ignored.
7532 79. If a space check for the spool or log partitions fails, the incident is now
7533 logged. Of course, in the latter case the data may get lost...
7535 80. Added the %p formatting code to string_format() so that it can be used to
7536 print addresses in debug_print(). Adjusted all the address printing in the
7537 debugging in store.c to use %p rather than %d.
7539 81. There was a concern that a line of code in smtp_in.c could overflow a
7540 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7541 initially expressed, the concern was not well-founded, because trailing
7542 spaces are removed early. However, if the trailing spaces were followed by
7543 a NULL, they did not get removed, so the overflow was possible. Two fixes
7546 (a) I re-wrote the offending code in a cleaner fashion.
7547 (b) If an incoming SMTP command contains a NULL character, it is rejected
7550 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7551 runs initgroups(), so that if the Exim user is in any additional groups,
7552 they will be used during message reception.
7558 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / heiko / exim.git / blob