1 Change log file for Exim from version 4.21
2 ------------------------------------------
3 This document describes *changes* to previous versions, that might
4 affect Exim's operation, with an unchanged configuration file. For new
5 options, and new features, see the NewStuff file next to this ChangeLog.
11 JH/01 OpenSSL: With debug enabled output keying information sufficient, server
12 side, to decode a TLS 1.3 packet capture.
14 JH/02 OpenSSL: Suppress the sending of (stateful) TLS1.3 session tickets.
15 Previously the default library behaviour applied, sending two, each in
18 JH/03 Debug output for ACL now gives the config file name and line number for
21 JH/04 The default received_header_text now uses the RFC 8314 tls cipher clause.
23 JH/05 DKIM: ensure that dkim_domain elements are lowercased before use.
25 JH/06 Fix buggy handling of autoreply bounce_return_size_limit, and a possible
26 buffer overrun for (non-chunking) other transports.
28 JH/07 GnuTLS: Our use of late (post-handshake) certificate verification, under
29 TLS1.3, means that a server rejecting a client certificate is not visible
30 to the client until the first read of encrypted data (typically the
31 response to EHLO). Add detection for that case and treat it as a failed
32 TLS connection attempt, so that the normal retry-in-clear can work (if
35 JB/01 Bug 2375: fix expansions of 822 addresses having comments in local-part
36 and/or domain. Found and fixed by Jason Betts.
38 JH/08 Add hardening against SRV & TLSA lookups the hit CNAMEs (a nonvalid
39 configuration). If a CNAME target was not a wellformed name pattern, a
42 JH/09 Logging: Fix initial listening-on line for multiple ports for an IP when
43 the OS reports them interleaved with other addresses.
45 JH/10 OpenSSL: Fix aggregation of messages. Previously, when PIPELINING was
46 used both for input and for a verify callout, both encrypted, SMTP
47 responses being sent by the server could be lost. This resulted in
48 dropped connections and sometimes bounces generated by a peer sending
51 JH/11 Harden plaintext authenticator against a badly misconfigured client-send
52 string. Previously it was possible to cause undefined behaviour in a
53 library routine (usually a crash). Found by "zerons".
55 JH/12 Bug 2384: fix "-bP smtp_receive_timeout". Previously it returned no
58 JH/13 Bug 2386: Fix builds with Dane under LibreSSL 2.9.0 onward. Some old
59 API was removed, so update to use the newer ones.
61 JH/14 Bug 1891: Close the log file if receiving a non-smtp message, without
62 any timeout set, is taking a long time. Previously we would hang on to a
63 rotated logfile "forever" if the input was arriving with long gaps
64 (a previous attempt to fix addressed lack, for a long time, of initial
67 HS/01 Bug 2390: Use message_id for tempfile creation to avoid races in a
68 shared (NFS) environment. The length of the tempfile name is now
69 4 + 16 ("hdr.$message_exim_id") which might break on file
70 systems which restrict the file name length to lower values.
73 HS/01 Bug 2390: Use message_id for tempfile creation to avoid races in a
74 shared (NFS) environment.
76 HS/02 Bug 2392: exigrep does case sensitive *option* processing (as it
77 did for all versions <4.90). Notably -M, -m, --invert, -I may be
80 JH/15 Use unsigned when creating bitmasks in macros, to avoid build errors
81 on some platforms for bit 31.
83 JH/16 GnuTLS: rework ciphersuite strings under recent library versions. Thanks
84 to changes apparently associated with TLS1.3 handling some of the APIs
85 previously used were either nonfunctional or inappropriate. Strings
86 like TLS1.3:ECDHE_SECP256R1__RSA_PSS_RSAE_SHA256__AES_256_GCM__AEAD:256
87 and TLS1.2:ECDHE_SECP256R1__RSA_SHA256__AES_128_CBC__SHA256:128 replace
88 the previous TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256 .
89 This affects log line X= elements, the $tls_{in,out}_cipher variables,
90 and the use of specific cipher names in the encrypted= ACL condition.
92 JH/17 OpenSSL: the default openssl_options now disables ssl_v3.
94 JH/18 GnuTLS: fix $tls_out_ocsp under hosts_request_ocsp. Previously the
95 verification result was not updated unless hosts_require_ocsp applied.
97 JH/19 Bug 2398: fix listing of a named-queue. Previously, even with the option
98 queue_list_requires_admin set to false, non-admin users were denied the
101 JH/20 Bug 2389: fix server advertising of usable certificates, under GnuTLS in
102 directory-of-certs mode. Previously they were advertised despite the
105 JH/21 The smtp transport option "hosts_noproxy_tls" is now unset by default.
106 A single TCP connection by a client will now hold a TLS connection open
107 for multiple message deliveries, by default. Previoud the default was to
110 JH/22 The smtp transport option "hosts_try_dane" now enables all hosts by
111 default. If built with the facility, DANE will be used. The facility
112 SUPPORT_DANE is now enabled in the prototype build Makefile "EDITME".
114 JH/23 The build default is now for TLS to be included; the SUPPORT_TLS define
115 is replaced with DISABLE_TLS. Either USE_GNUTLS or (the new) USE_OPENSSL
116 must be defined and you must still, unless you define DISABLE_TLS, manage
117 the the include-dir and library-file requirements that go with that
118 choice. Non-TLS builds are still supported.
120 JH/24 Fix duplicated logging of peer name/address, on a transport connection-
123 JH/25 The smtp transport option "hosts_try_fastopen" now enables all hosts by
124 default. If the platform supports and has the facility enabled, it will
125 be requested on all coneections.
127 JH/26 The PIPE_CONNECT facility is promoted from experimental status and is now
128 controlled by the build-time option SUPPORT_PIPE_CONNECT.
130 PP/01 Unbreak heimdal_gssapi, broken in 4.92.
132 JH/27 Bug 2404: Use the main-section configuration option "dsn_from" for
133 success-DSN messages. Previously the From: header was always the default
134 one for these; the option was ignored.
136 JH/28 Fix the timeout on smtp response to apply to the whole response.
137 Previously it was reset for every read, so a teergrubing peer sending
138 single bytes within the time limit could extend the connection for a
139 long time. Credit to Qualsys Security Advisory Team for the discovery.
141 JH/29 Fix DSN Final-Recipient: field. Previously it was the post-routing
142 delivery address, which leaked information of the results of local
143 forwarding. Change to the original envelope recipient address, per
150 JH/01 Remove code calling the customisable local_scan function, unless a new
151 definition "HAVE_LOCAL_SCAN=yes" is present in the Local/Makefile.
153 JH/02 Bug 1007: Avoid doing logging from signal-handlers, as that can result in
154 non-signal-safe functions being used.
156 JH/03 Bug 2269: When presented with a received message having a stupidly large
157 number of DKIM-Signature headers, disable DKIM verification to avoid
158 a resource-consumption attack. The limit is set at twenty.
160 JH/04 Add variables $arc_domains, $arc_oldest_pass for ARC verify. Fix the
161 report of oldest_pass in ${authres } in consequence, and separate out
162 some descriptions of reasons for verification fail.
164 JH/05 Bug 2273: Cutthrough delivery left a window where the received messsage
165 files in the spool were present and unlocked. A queue-runner could spot
166 them, resulting in a duplicate delivery. Fix that by doing the unlock
167 after the unlink. Investigation by Tim Stewart. Take the opportunity to
168 add more error-checking on spoolfile handling while that code is being
171 PP/01 Refuse to open a spool data file (*-D) if it's a symlink.
172 No known attacks, no CVE, this is defensive hardening.
174 JH/06 Bug 2275: The MIME ACL unlocked the received message files early, and
175 a queue-runner could start a delivery while other operations were ongoing.
176 Cutthrough delivery was a common victim, resulting in duplicate delivery.
177 Found and investigated by Tim Stewart. Fix by using the open message data
178 file handle rather than opening another, and not locally closing it (which
179 releases a lock) for that case, while creating the temporary .eml format
180 file for the MIME ACL. Also applies to "regex" and "spam" ACL conditions.
182 JH/07 Bug 177: Make a random-recipient callout success visible in ACL, by setting
183 $sender_verify_failure/$recipient_verify_failure to "random".
185 JH/08 When generating a selfsigned cert, use serial number 1 since zero is not
188 JH/09 Bug 2274: Fix logging of cmdline args when starting in an unlinked cwd.
189 Previously this would segfault.
191 JH/10 Fix ARC signing for case when DKIM signing failed. Previously this would
194 JH/11 Bug 2264: Exim now only follows CNAME chains one step by default. We'd
195 like zero, since the resolver should be doing this for us, But we need one
196 as a CNAME but no MX presence gets the CNAME returned; we need to check
197 that doesn't point to an MX to declare it "no MX returned" rather than
198 "error, loop". A new main option is added so the older capability of
199 following some limited number of chain links is maintained.
201 JH/12 Add client-ip info to non-pass iprev ${authres } lines.
203 JH/13 For receent Openssl versions (1.1 onward) use modern generic protocol
204 methods. These should support TLS 1.3; they arrived with TLS 1.3 and the
205 now-deprecated earlier definitions used only specified the range up to TLS
206 1.2 (in the older-version library docs).
208 JH/14 Bug 2284: Fix DKIM signing for body lines starting with a pair of dots.
210 JH/15 Rework TLS client-side context management. Stop using a global, and
211 explicitly pass a context around. This enables future use of TLS for
212 connections to service-daemons (eg. malware scanning) while a client smtp
213 connection is using TLS; with cutthrough connections this is quite likely.
215 JH/16 Fix ARC verification to do AS checks in reverse order.
217 JH/17 Support a "tls" option on the ${readsocket } expansion item.
219 JH/18 Bug 2287: Fix the protocol name (eg utf8esmtp) for multiple messages
220 using the SMTPUTF8 option on their MAIL FROM commands, in one connection.
221 Previously the "utf8" would be re-prepended for every additional message.
223 JH/19 Reject MAIL FROM commands with SMTPUTF8 when the facility was not advertised.
224 Previously thery were accepted, resulting in issues when attempting to
225 forward messages to a non-supporting MTA.
227 PP/02 Let -n work with printing macros too, not just options.
229 JH/20 Bug 2296: Fix cutthrough for >1 address redirection. Previously only
230 one parent address was copied, and bogus data was used at delivery-logging
231 time. Either a crash (after delivery) or bogus log data could result.
232 Discovery and analysis by Tim Stewart.
234 PP/03 Make ${utf8clean:} expansion operator detect incomplete final character.
235 Previously if the string ended mid-character, we did not insert the
236 promised '?' replacement.
238 PP/04 Documentation: current string operators work on bytes, not codepoints.
240 JH/21 Change as many as possible of the global flags into one-bit bitfields; these
241 should pack well giving a smaller memory footprint so better caching and
242 therefore performance. Group the declarations where this can't be done so
243 that the byte-sized flag variables are not interspersed among pointer
244 variables, giving a better chance of good packing by the compiler.
246 JH/22 Bug 1896: Fix the envelope from for DMARC forensic reports to be possibly
247 non-null, to avoid issues with sites running BATV. Previously reports were
248 sent with an empty envelope sender so looked like bounces.
250 JH/23 Bug 2318: Fix the noerror command within filters. It wasn't working.
251 The ignore_error flag wasn't being returned from the filter subprocess so
252 was not set for later routers. Investigation and fix by Matthias Kurz.
254 JH/24 Bug 2310: Raise a msg:fail:internal event for each undelivered recipient,
255 and a msg:complete for the whole, when a message is manually removed using
256 -Mrm. Developement by Matthias Kurz, hacked on by JH.
258 JH/25 Avoid fixed-size buffers for pathnames in DB access. This required using
259 a "Gnu special" function, asprintf() in the DB utility binary builds; I
260 hope that is portable enough.
262 JH/26 Bug 2311: Fix DANE-TA verification under GnuTLS. Previously it was also
263 requiring a known-CA anchor certificate; make it now rely entirely on the
264 TLSA as an anchor. Checking the name on the leaf cert against the name
265 on the A-record for the host is still done for TA (but not for EE mode).
267 JH/27 Fix logging of proxy address. Previously, a pointless "PRX=[]:0" would be
268 included in delivery lines for non-proxied connections, when compiled with
269 SUPPORT_SOCKS and running with proxy logging enabled.
271 JH/28 Bug 2314: Fire msg:fail:delivery event even when error is being ignored.
272 Developement by Matthias Kurz, tweaked by JH. While in that bit of code,
273 move the existing event to fire before the normal logging of message
274 failure so that custom logging is bracketed by normal logging.
276 JH/29 Bug 2322: A "fail" command in a non-system filter (file) now fires the
277 msg:fail:internal event. Developement by Matthias Kurz.
279 JH/30 Bug 2329: Increase buffer size used for dns lookup from 2k, which was
280 far too small for todays use of crypto signatures stored there. Go all
281 the way to the max DNS message size of 64kB, even though this might be
282 overmuch for IOT constrained device use.
284 JH/31 Fix a bad use of a copy function, which could be used to pointlessly
285 copy a string over itself. The library routine is documented as not
286 supporting overlapping copies, and on MacOS it actually raised a SIGABRT.
288 JH/32 For main options check_spool_space and check_inode_space, where the
289 platform supports 64b integers, support more than the previous 2^31 kB
290 (i.e. more than 2 TB). Accept E, P and T multipliers in addition to
291 the previous G, M, k.
293 JH/33 Bug 2338: Fix the cyrus-sasl authenticator to fill in the
294 $authenticated_fail_id variable on authentication failure. Previously
297 JH/34 Increase RSA keysize of autogen selfsign cert from 1024 to 2048. RHEL 8.0
298 OpenSSL didn't want to use such a weak key. Do for GnuTLS also, and for
299 more-modern GnuTLS move from GNUTLS_SEC_PARAM_LOW to
300 GNUTLS_SEC_PARAM_MEDIUM.
302 JH/35 OpenSSL: fail the handshake when SNI processing hits a problem, server
303 side. Previously we would continue as if no SNI had been received.
305 JH/36 Harden the handling of string-lists. When a list consisted of a sole
306 "<" character, which should be a list-separator specification, we walked
307 off past the nul-terimation.
309 JH/37 Bug 2341: Send "message delayed" warning MDNs (restricted to external
310 causes) even when the retry time is not yet met. Previously they were
311 not, meaning that when (say) an account was over-quota and temp-rejecting,
312 and multiple senders' messages were queued, only one sender would get
313 notified on each configured delay_warning cycle.
315 JH/38 Bug 2351: Log failures to extract envelope addresses from message headers.
317 JH/39 OpenSSL: clear the error stack after an SSL_accept(). With anon-auth
318 cipher-suites, an error can be left on the stack even for a succeeding
319 accept; this results in impossible error messages when a later operation
322 AM/01 Bug 2359: GnuTLS: repeat lowlevel read and write operations while they
323 return error codes indicating retry. Under TLS1.3 this becomes required.
325 JH/40 Fix the feature-cache refresh for EXPERIMENTAL_PIPE_CONNECT. Previously
326 it only wrote the new authenticators, resulting in a lack of tracking of
327 peer changes of ESMTP extensions until the next cache flush.
329 JH/41 Fix the loop reading a message header line to check for integer overflow,
330 and more-often against header_maxsize. Previously a crafted message could
331 induce a crash of the recive process; now the message is cleanly rejected.
333 JH/42 Bug 2366: Fix the behaviour of the dkim_verify_signers option. It had
334 been totally disabled for all of 4.91. Discovery and fix by "Mad Alex".
340 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
341 When redis_servers is set to a list of > 1 element, and the Redis servers
342 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
343 case of MOVED into a DEFER case instead, thus moving the query onto the
344 next server in the list. For a cluster of N elements, all N servers must
345 be defined in redis_servers.
347 GF/02 Catch and remove uninitialized value warning in exiqsumm
348 Check for existence of @ARGV before looking at $ARGV[0]
350 JH/01 Replace the store_release() internal interface with store_newblock(),
351 which internalises the check required to safely use the old one, plus
352 the allocate and data copy operations duplicated in both (!) of the
353 extant use locations.
355 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
356 modifier. This matches the restriction on the commandline.
358 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
359 Previously only the last row was returned.
361 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
362 we assumed that tags in the header were well-formed, and parsed the
363 element content after inspecting only the first char of the tag.
364 Assumptions at that stage could crash the receive process on malformed
367 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
368 While running the DKIM ACL we operate on the Permanent memory pool so that
369 variables created with "set" persist to the DATA ACL. Also (at any time)
370 DNS lookups that fail create cache records using the Permanent pool. But
371 expansions release any allocations made on the current pool - so a dnsdb
372 lookup expansion done in the DKIM ACL releases the memory used for the
373 DNS negative-cache, and bad things result. Solution is to switch to the
374 Main pool for expansions.
375 While we're in that code, add checks on the DNS cache during store_reset,
376 active in the testsuite.
377 Problem spotted, and debugging aided, by Wolfgang Breyha.
379 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
380 When none of the hosts presented to a transport match an already-open
381 connection, close it and proceed with the list. Previously we would
382 queue the message. Spotted by Lena with Yahoo, probably involving
385 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
386 Previously a spurious "250 OK id=" response was appended to the proper
389 JH/08 The "support for" informational output now, which built with Content
390 Scanning support, has a line for the malware scanner interfaces compiled
391 in. Interface can be individually included or not at build time.
393 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
394 by the template makefile "src/EDITME". The "STREAM" support for an older
395 ClamAV interface method is removed.
397 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
398 rows affected is given instead).
400 JH/11 The runtime Berkeley DB library version is now additionally output by
401 "exim -d -bV". Previously only the compile-time version was shown.
403 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
404 SMTP connection. Previously, when one had more receipients than the
405 first, an abortive onward connection was made. Move to full support for
406 multiple onward connections in sequence, handling cutthrough connection
407 for all multi-message initiating connections.
409 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
410 routers. Previously, a multi-recipient message would fail to match the
411 onward-connection opened for the first recipient, and cause its closure.
413 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
414 a timeout on read on a GnuTLS initiating connection, resulting in the
415 initiating connection being dropped. This mattered most when the callout
416 was marked defer_ok. Fix to keep the two timeout-detection methods
419 JH/15 Relax results from ACL control request to enable cutthrough, in
420 unsupported situations, from error to silently (except under debug)
421 ignoring. This covers use with PRDR, frozen messages, queue-only and
424 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
426 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
427 metadata, resulting in a crash in free().
429 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
430 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
431 Broken also in d185889f4, with init system revamp.
433 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
434 Previously we abruptly closed the connection after reading a malware-
435 found indication; now we go on to read the "scan ok" response line,
438 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
439 ACL. Previously, a crash would result.
441 JH/19 Speed up macro lookups during configuration file read, by skipping non-
442 macro text after a replacement (previously it was only once per line) and
443 by skipping builtin macros when searching for an uppercase lead character.
445 JH/20 DANE support moved from Experimental to mainline. The Makefile control
446 for the build is renamed.
448 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
449 was allocated for every new TLS startup, meaning one per message. Fix
450 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
452 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
453 reported the original. Fix to report (as far as possible) the ACL
454 result replacing the original.
456 JH/23 Fix memory leak during multi-message connections using STARTTLS under
457 OpenSSL. Certificate information is loaded for every new TLS startup,
458 and the resources needed to be freed.
460 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
462 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
465 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
466 DATA response info to the (existing) per-recipient response info for
467 the "C=" log element. It can have useful tracking info from the
468 destination system. Patch from Simon Arlott.
470 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
471 length value. Previously this would segfault.
473 HS/02 Support Avast multiline protoocol, this allows passing flags to
474 newer versions of the scanner.
476 JH/28 Ensure that variables possibly set during message acceptance are marked
477 dead before release of memory in the daemon loop. This stops complaints
478 about them when the debug_store option is enabled. Discovered specifically
479 for sender_rate_period, but applies to a whole set of variables.
480 Do the same for the queue-runner and queue-list loops, for variables set
481 from spool message files. Do the same for the SMTP per-message loop, for
482 certain variables indirectly set in ACL operations.
484 JH/29 Bug 2250: Fix a longstanding bug in heavily-pipelined SMTP input (such
485 as a multi-recipient message from a mailinglist manager). The coding had
486 an arbitrary cutoff number of characters while checking for more input;
487 enforced by writing a NUL into the buffer. This corrupted long / fast
488 input. The problem was exposed more widely when more pipelineing of SMTP
489 responses was introduced, and one Exim system was feeding another.
490 The symptom is log complaints of SMTP syntax error (NUL chars) on the
491 receiving system, and refused recipients seen by the sending system
492 (propating to people being dropped from mailing lists).
493 Discovered and pinpointed by David Carter.
495 JH/30 The (EXPERIMENTAL_DMARC) variable $dmarc_ar_header is withdrawn, being
496 replaced by the ${authresults } expansion.
498 JH/31 Bug 2257: Fix pipe transport to not use a socket-only syscall.
500 HS/03 Set a handler for SIGTERM and call exit(3) if running as PID 1. This
501 allows proper process termination in container environments.
503 JH/32 Bug 2258: Fix spool_wireformat in combination with LMTP transport.
504 Previously the "final dot" had a newline after it; ensure it is CR,LF.
506 JH/33 SPF: remove support for the "spf" ACL condition outcome values "err_temp"
507 and "err_perm", deprecated since 4.83 when the RFC-defined words
508 "temperror" and "permerror" were introduced.
510 JH/34 Re-introduce enforcement of no cutthrough delivery on transports having
511 transport-filters or DKIM-signing. The restriction was lost in the
512 consolidation of verify-callout and delivery SMTP handling.
513 Extend the restriction to also cover ARC-signing.
515 JH/35 Cutthrough: for a final-dot response timeout (and nonunderstood responses)
516 in defer=pass mode supply a 450 to the initiator. Previously the message
519 PP/02 DANE: add dane_require_tls_ciphers SMTP Transport option; if unset,
520 tls_require_ciphers is used as before.
522 HS/03 Malware Avast: Better match the Avast multiline protocol. Add
523 "pass_unscanned". Only tmpfails from the scanner are written to
524 the paniclog, as they may require admin intervention (permission
525 denied, license issues). Other scanner errors (like decompression
526 bombs) do not cause a paniclog entry.
528 JH/36 Fix reinitialisation of DKIM logging variable between messages.
529 Previously it was possible to log spurious information in receive log
532 JH/37 Bug 2255: Revert the disable of the OpenSSL session caching. This
533 triggered odd behaviour from Outlook Express clients.
535 PP/03 Add util/renew-opendmarc-tlds.sh script for safe renewal of public
538 JH/38 DKIM: accept Ed25519 pubkeys in SubjectPublicKeyInfo-wrapped form,
539 since the IETF WG has not yet settled on that versus the original
540 "bare" representation.
542 JH/39 Fix syslog logging for syslog_timestamp=no and log_selector +millisec.
543 Previously the millisecond value corrupted the output.
544 Fix also for syslog_pid=no and log_selector +pid, for which the pid
545 corrupted the output.
551 JH/01 Rework error string handling in TLS interface so that the caller in
552 more cases is responsible for logging. This permits library-sourced
553 string to be attached to addresses during delivery, and collapses
554 pairs of long lines into single ones.
556 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
557 during configuration. Wildcards are allowed and expanded.
559 JH/02 Rework error string handling in DKIM to pass more info back to callers.
560 This permits better logging.
562 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
563 do not close it down and have the child transport start it up again on
564 the passed-on TCP connection. Instead, proxy the child (and any
565 subsequent ones) for TLS via a unix-domain socket channel. Logging is
566 affected: the continued delivery log lines do not have any DNSSEC, TLS
567 Certificate or OCSP information. TLS cipher information is still logged.
569 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
570 identical IP addresses on different listening ports. Will also affect
573 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
574 add noisy ifdef guards to special-case this sillyness.
575 Patch from Bernd Kuhls.
577 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
578 than 255 are no longer allowed.
580 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
581 Disable the session-cache too, which might reduce our load. Since we
582 currrectly use a new context for every connection, both as server and
583 client, there is no benefit for these.
584 GnuTLS appears to not support tickets server-side by default (we don't
585 call gnutls_session_ticket_enable_server()) but client side is enabled
586 by default on recent versions (3.1.3 +) unless the PFS priority string
589 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
590 <https://reproducible-builds.org/specs/source-date-epoch/>.
592 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
593 the check for any unsuccessful recipients did not notice the limit, and
594 erroneously found still-pending ones.
596 JH/08 Pipeline CHUNKING command and data together, on kernels that support
597 MSG_MORE. Only in-clear (not on TLS connections).
599 JH/09 Avoid using a temporary file during transport using dkim. Unless a
600 transport-filter is involved we can buffer the headers in memory for
601 creating the signature, and read the spool data file once for the
602 signature and again for transmission.
604 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
605 4.77 as the kernel support then wasn't solid, having issues in 64bit
606 mode. Now, it's been long enough. Add support for FreeBSD also.
608 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
609 case where the routing stage had gathered several addresses to send to
610 a host before calling the transport for the first, we previously failed
611 to close down TLS in the old transport process before passing the TCP
612 connection to the new process. The new one sent a STARTTLS command
613 which naturally failed, giving a failed delivery and bloating the retry
614 database. Investigation and fix prototype from Wolfgang Breyha.
616 JH/12 Fix check on SMTP command input synchronisation. Previously there were
617 false-negatives in the check that the sender had not preempted a response
618 or prompt from Exim (running as a server), due to that code's lack of
619 awareness of the SMTP input buffering.
621 PP/04 Add commandline_checks_require_admin option.
622 Exim drops privileges sanely, various checks such as -be aren't a
623 security problem, as long as you trust local users with access to their
624 own account. When invoked by services which pass untrusted data to
625 Exim, this might be an issue. Set this option in main configuration
626 AND make fixes to the calling application, such as using `--` to stop
629 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
630 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
631 responses to those, into a single TLS record each way (this usually means
632 a single packet). As a side issue, smtp_enforce_sync now works on TLS
635 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
636 affects you only if you're dancing at the edge of the param size limits.
637 If you are, and this message makes sense to you, then: raise the
638 configured limit or use OpenSSL 1.1. Nothing we can do for older
641 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
642 cmdline element to get the documented default one. Previously it was
645 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
646 are unsuable for non-message-specific reasons - eg. connection timeout,
647 banner-time rejection.
649 JH/16 Fix logging of delivery remote port, when specified by router, under
652 PP/06 Repair manualroute's ability to take options in any order, even if one
653 is the name of a transport.
656 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
658 JH/17 Change the list-building routines interface to use the expanding-string
659 triplet model, for better allocation and copying behaviour.
661 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
662 Previously it was constructed the first time a possibly-matching string
663 was met in the configuration file input during startup; now it is done
666 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
667 compatible one, to avoid the (poorly documented) possibility of a config
668 file in the working directory redirecting the DB files, possibly correpting
669 some existing file. CVE-2017-10140 assigned for BDB.
671 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
672 cache-hot. Previously, although the result was properly cached, the
673 initial verify call returned a defer.
675 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
676 the main verify for receipient in uncached-mode.
678 JH/22 Retire historical build files to an "unsupported" subdir. These are
679 defined as "ones for which we have no current evidence of testing".
681 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
682 if present. Previously it was ignored.
684 JH/24 Start using specified-initialisers in C structure init coding. This is
685 a C99 feature (it's 2017, so now considered safe).
687 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
688 if was a fixed-sized field and bitmask ops via macros; it is now more
691 PP/07 GitHub PR 56: Apply MariaDB build fix.
692 Patch provided by Jaroslav Škarvada.
694 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
695 during Coverity cleanups [4.87 JH/47]
696 Diagnosis and fix provided by Michael Fischer v. Mollard.
698 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
699 the right size to place the terminating semicolon on its own folded
700 line, the header hash was calculated to an incorrect value thanks to
701 the (relaxed) space the fold became.
703 HS/02 Fix Bug 2130: large writes from the transport subprocess were chunked
704 and confused the parent.
706 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
707 which could crash as a result. This could lead to undeliverable messages.
709 JH/28 Logging: "next input sent too soon" now shows where input was truncated
712 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
713 matters on fast-turnover and PID-randomising systems, which were getting
714 out-of-order delivery.
716 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
717 a possibly-overlapping copy. The symptom was that "Remote host closed
718 connection in response to HELO" was logged instead of the actual 4xx
721 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
722 Previously only that bufferd was discarded, resulting in SYMTP command
725 JH/32 DKIM: when a message has multiple signatures matching an identity given
726 in dkim_verify_signers, run the dkim acl once for each. Previously only
727 one run was done. Bug 2189.
729 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
730 "panic the current process" to "deliberately defer". The panic log is
731 still written with the problem list name; the mail and reject logs now
732 get a temp-reject line for the message that was being handled, saying
733 something like "domains check lookup or other defer". The SMTP 451
734 message is still "Temporary local problem".
736 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
737 A crafted sequence of BDAT commands could result in in-use memory beeing
738 freed. CVE-2017-16943.
740 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
741 from SMTP input. Previously it was always done; now only done for DATA
742 and not BDAT commands. CVE-2017-16944.
744 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
745 to the message (such as an overlong header line). Previously this was
746 not done and we did not exit BDAT mode. Followon from the previous item
747 though a different problem.
753 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
754 than -2003 did; needs libidn2 in addition to libidn.
756 JH/02 The path option on a pipe transport is now expanded before use.
758 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
759 Patch provided by "Björn", documentation fix added too.
761 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
762 missing a wire-to-host endian conversion.
764 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
765 close after a BDAT command line could be taken as a following command,
766 giving a synch failure. Fix by only checking for synch immediately
767 before acknowledging the chunk.
769 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
770 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
771 macro. Patches provided by Josh Soref.
773 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
774 Previously we did not; the RFC seems ambiguous and VRFY is not listed
775 by IANA as a service extension. However, John Klensin suggests that we
778 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
779 the dkim code may be unix-mode line endings rather than smtp wire-format
780 CRLF, so prepend a CR to any bare LF.
782 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
783 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
785 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
786 the first were themselves being wrongly included in the feed into dkim
787 processing; with most chunk sizes in use this resulted in an incorrect
788 body hash calculated value.
790 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
791 DKIM signature block, for verification. Although advised against by
792 standards it is specifically not ruled illegal.
794 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
796 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
797 missing a body hash (the bh= tag).
799 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
800 It seems that HAProxy sends the Proxy Protocol information in clear and
801 only then does a TLS startup, so do the same.
803 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
804 TCP connections (such as for Spamd) unless the daemon successfully set
805 Fast Open mode on its listening sockets. This fixes breakage seen on
806 too-old kernels or those not configured for Fast Open, at the cost of
807 requiring both directions being enabled for TFO, and TFO never being used
808 by non-daemon-related Exim processes.
810 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
811 endings, at least on the first header line. Try to canonify any that get
812 past that check, despite the cost.
814 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
815 now limited to an arbitrary five deep, while parsing addresses with the
816 strip_excess_angle_brackets option enabled.
818 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
819 instead leave the unprompted TLS handshake in socket buffer for the
820 TLS library to consume.
822 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
824 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
826 JH/16 Drop variables when they go out of scope. Memory management drops a whole
827 region in one operation, for speed, and this leaves assigned pointers
828 dangling. Add checks run only under the testsuite which checks all
829 variables at a store-reset and panics on a dangling pointer; add code
830 explicitly nulling out all the variables discovered. Fixes one known
831 bug: a transport crash, where a dangling pointer for $sending_ip_address
832 originally assigned in a verify callout, is re-used.
834 PP/06 Drop '.' from @INC in various Perl scripts.
836 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
838 PP/08 Reduce a number of compilation warnings under clang; building with
839 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
840 should be warning-free.
842 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
844 HS/01 Fix portability problems introduced by PP/08 for platforms where
845 realloc(NULL) is not equivalent to malloc() [SunOS et al].
847 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
848 chunk. This allows us to accept broken chunked messages. We need a more
849 general solution here.
851 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
852 already-broken messages in the queue.
854 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
856 JH/19 Fix reference counting bug in routing-generated-address tracking.
862 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
863 supports it and a size is available (ie. the sending peer gave us one).
865 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
866 years of being deprecated). The replacements are the ACLs
867 acl_smtp_mime and acl_not_smtp_mime.
869 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
870 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
871 or even an in-clear connection were permitted. Now, if the host lookup
872 was dnssec and dane was requested then the host is only used if the
873 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
874 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
875 if one fails this test.
876 This means that a poorly-configured remote DNS will make it incommunicado;
877 but it protects against a DNS-interception attack on it.
879 JH/04 Bug 1810: make continued-use of an open smtp transport connection
880 non-noisy when a race steals the message being considered.
882 JH/05 If main configuration option tls_certificate is unset, generate a
883 self-signed certificate for inbound TLS connections.
885 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
886 in rewrites and routers.
888 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
889 and logged a warning sing 4.83; now they are a configuration file error.
891 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
892 (lacking @domain). Apply the same qualification processing as RCPT.
894 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
896 JH/10 Support ${sha256:} applied to a string (as well as the previous
899 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
900 a cutthrough deliver is pending, as we always want to make a connection.
901 This also avoids re-routing the message when later placing the cutthrough
902 connection after a verify cache hit.
903 Do not update it with the verify result either.
905 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
906 when routing results in more than one destination address.
908 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
909 signing (which inhibits the cutthrough capability). Previously only
910 the presence of an option was tested; now an expansion evaluating as
911 empty is permissible (obviously it should depend only on data available
912 when the cutthrough connection is made).
914 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
915 the relevant preceding SMTP command did not note the pipelining mode.
917 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
918 Previously they were not counted.
920 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
921 as one having no matching records. Previously we deferred the message
922 that needed the lookup.
924 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
925 distinguished as "(=".
927 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
928 for missing MX records. Previously it only worked for missing A records.
930 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
932 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
933 after the data-go-ahead and data-ack. Patch from Jason Betts.
935 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
936 even for a "none" policy. Patch from Tony Meyer.
938 JH/22 Fix continued use of a connection for further deliveries. If a port was
939 specified by a router, it must also match for the delivery to be
942 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
943 When one of the recipients of a message was unsuitable for the connection
944 (has no matching addresses), we lost track of needing to mark it
945 deferred. As a result mail would be lost.
947 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
949 JH/25 Decoding ACL controls is now done using a binary search; the source code
950 takes up less space and should be simpler to maintain. Merge the ACL
951 condition decode tables also, with similar effect.
953 JH/26 Fix problem with one_time used on a redirect router which returned the
954 parent address unchanged. A retry would see the parent address marked as
955 delivered, so not attempt the (identical) child. As a result mail would
958 JH/27 Fix a possible security hole, wherein a process operating with the Exim
959 UID can gain a root shell. Credit to http://www.halfdog.net/ for
960 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
963 JH/28 Enable {spool,log} filesystem space and inode checks as default.
964 Main config options check_{log,spool}_{inodes,space} are now
965 100 inodes, 10MB unless set otherwise in the configuration.
967 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
968 Previously it only applied to the main-section connection policy
971 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
973 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
974 by me. Added RFC7919 DH primes as an alternative.
976 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
977 are not in the system include path.
979 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
980 GnuTLS, when a session startup failed (eg because the client disconnected)
981 Exim did stdio operations after fclose. This was exposed by a recent
982 change which nulled out the file handle after the fclose.
984 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
985 signed directly by the cert-signing cert, rather than an intermediate
986 OCSP-signing cert. This is the model used by LetsEncrypt.
988 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
990 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
991 an incoming connection.
993 HS/02 Bug 1802: Do not half-close the connection after sending a request
996 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
997 fallback to "prime256v1".
999 JH/34 SECURITY: Use proper copy of DATA command in error message.
1000 Could leak key material. Remotely exploitable. CVE-2016-9963.
1006 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
1007 and 3.4.4 - once the server is enabled to respond to an OCSP request
1008 it does even when not requested, resulting in a stapling non-aware
1009 client dropping the TLS connection.
1011 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
1012 support variable-length bit vectors. No functional change.
1014 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
1015 The I= interface field on outgoing lines is now after the H= remote
1016 host field, same as incoming lines. There is a separate
1017 outgoing_interface log selector which allows you to disable the
1020 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
1021 If not running log_selector +smtp_connection the mainlog would be held
1022 open indefinitely after a "too many connections" event, including to a
1023 deleted file after a log rotate. Leave the per net connection logging
1024 leaving it open for efficiency as that will be quickly detected by the
1025 check on the next write.
1027 HS/01 Bug 1671: Fix post transport crash.
1028 Processing the wait-<transport> messages could crash the delivery
1029 process if the message IDs didn't exist for some reason. When
1030 using 'split_spool_directory=yes' the construction of the spool
1031 file name failed already, exposing the same netto behaviour.
1033 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
1034 mime_regex ACL conditions.
1036 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
1037 to DSN fail messages (bounces): remote IP, remote greeting, remote response
1038 to HELO, local diagnostic string.
1040 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
1041 log line to debug. Even when configured with a tls authenticator many
1042 client connections are expected to not authenticate in this way, so
1043 an authenticate fail is not an error.
1045 HS/02 Add the Exim version string to the process info. This way exiwhat
1046 gives some more detail about the running daemon.
1048 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
1049 matter for fast-change records such as DNSBLs.
1051 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
1052 retry record, even if constant. There may be multiple transports with
1053 different interface settings and the retry behaviour needs to be kept
1056 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
1058 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
1060 JH/10 Bug 840: fix log_defer_output option of pipe transport
1062 JH/11 Bug 830: use same host for all RCPTS of a message, even under
1063 hosts_randomize. This matters a lot when combined with mua_wrapper.
1065 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
1066 ${quote_pgsql:<string>} operator.
1068 JH/13 Bug 1708: avoid misaligned access in cached lookup.
1070 JH/14 Change header file name for freeradius-client. Relevant if compiling
1071 with Radius support; from the Gentoo tree and checked under Fedora.
1073 JH/15 Bug 1712: Introduce $prdr_requested flag variable
1075 JH/16 Bug 1714: Permit an empty string as expansion result for transport
1076 option transport_filter, meaning no filtering.
1078 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
1080 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
1081 defaults to "*" (all hosts). The variable is now available when not built
1082 with TLS, default unset, mainly to enable keeping the testsuite sane.
1083 If a server certificate is not supplied (via tls_certificate) an error is
1084 logged, and clients will find TLS connections fail on startup. Presumably
1085 they will retry in-clear.
1086 Packagers of Exim are strongly encouraged to create a server certificate
1087 at installation time.
1089 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
1090 with the $config_file variable.
1092 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
1093 in transport context, after the attempt, and per-recipient. The latter type
1094 is per host attempted. The event data is the error message, and the errno
1095 information encodes the lookup type (A vs. MX) used for the (first) host,
1096 and the trailing two digits of the smtp 4xx response.
1098 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
1099 to write to mainlog (or rejectlog, paniclog) in the window between file
1100 creation and permissions/ownership being changed. Particularly affects
1101 installations where exicyclog is run as root, rather than exim user;
1102 result is that the running daemon panics and dies.
1104 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
1106 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
1107 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
1108 "pri" and "weight". Note that the previous implicit priority given by the
1109 list order is no longer honoured.
1111 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
1112 for DKIM processing.
1114 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
1115 by defining SUPPORT_SOCKS.
1117 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
1118 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
1119 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
1120 variables are renamed to proxy_{local,external}_{address,port}.
1122 JH/27 Move Internationalisation support from Experimental to mainline, enabled
1123 for a build by defining SUPPORT_I18N
1125 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
1126 of the query string, and make ${quote_redis:} do that quoting.
1128 JH/29 Move Events support from Experimental to mainline, enabled by default
1129 and removable for a build by defining DISABLE_EVENT.
1131 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
1133 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
1134 cached by the daemon.
1136 JH/32 Move Redis support from Experimental to mainline, enabled for a build
1137 by defining LOOKUP_REDIS. The libhiredis library is required.
1139 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
1140 keys are given for lookup.
1142 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
1143 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
1144 only supported when built with TLS support. The PolarSSL SHA routines
1145 are still used when the TLS library is too old for convenient support.
1147 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
1148 openssl_options), for security. OpenSSL forces this from version 1.1.0
1149 server-side so match that on older versions.
1151 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
1152 allocation for $value could be released as the expansion processing
1153 concluded, but leaving the global pointer active for it.
1155 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
1156 and to use the domains and local_parts ACL conditions.
1158 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
1159 incorrectly not doubled on cutthrough transmission, hence seen as a
1160 body-termination at the receiving system - resulting in truncated mails.
1161 Commonly the sender saw a TCP-level error, and retransmitted the message
1162 via the normal store-and-forward channel. This could result in duplicates
1163 received - but deduplicating mailstores were liable to retain only the
1164 initial truncated version.
1166 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
1168 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
1170 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
1171 we're in there, support oversigning also; bug 1309.
1173 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
1175 HS/04 Add support for keep_environment and add_environment options.
1177 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
1178 either intentional arithmetic overflow during PRNG, or testing config-
1181 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
1182 delivery resulted in actual delivery. Cancel cutthrough before DATA
1185 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
1186 rejects a recipient: pass the reject to the originator.
1188 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
1189 Many were false-positives and ignorable, but it's worth fixing the
1192 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
1193 for the new environment-manipulation done at startup. Move the routines
1194 from being local to tls.c to being global via the os.c file.
1196 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
1197 an extract embedded as result-arg for a map, the first arg for extract
1198 is unavailable so we cannot tell if this is a numbered or keyed
1199 extraction. Accept either.
1205 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
1208 JH/02 The smtp transport option "multi_domain" is now expanded.
1210 JH/03 The smtp transport now requests PRDR by default, if the server offers
1213 JH/04 Certificate name checking on server certificates, when exim is a client,
1214 is now done by default. The transport option tls_verify_cert_hostnames
1215 can be used to disable this per-host. The build option
1216 EXPERIMENTAL_CERTNAMES is withdrawn.
1218 JH/05 The value of the tls_verify_certificates smtp transport and main options
1219 default to the word "system" to access the system default CA bundle.
1220 For GnuTLS, only version 3.0.20 or later.
1222 JH/06 Verification of the server certificate for a TLS connection is now tried
1223 (but not required) by default. The verification status is now logged by
1224 default, for both outbound TLS and client-certificate supplying inbound
1227 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
1230 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
1231 Status Notification (bounce) messages are now MIME format per RFC 3464.
1232 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
1233 under the control of the dsn_advertise_hosts option, and routers may
1234 have a dsn_lasthop option.
1236 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
1237 default, modifiable by a malware= option. The list separator for
1238 the options can now be changed in the usual way. Bug 68.
1240 JH/10 The smtp_receive_timeout main option is now expanded before use.
1242 JH/11 The incoming_interface log option now also enables logging of the
1243 local interface on delivery outgoing connections.
1245 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
1246 if the interface and destination host and port all match.
1248 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
1251 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
1252 Patch from Andrew Lewis.
1254 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
1255 now supports optional time-restrictions, weighting, and priority
1256 modifiers per server. Patch originally by <rommer@active.by>.
1258 JH/16 The spamd_address main option now supports a mixed list of local
1259 and remote servers. Remote servers can be IPv6 addresses, and
1260 specify a port-range.
1262 JH/17 Bug 68: The spamd_address main option now supports an optional
1263 timeout value per server.
1265 JH/18 Bug 1581: Router and transport options headers_add/remove can
1266 now have the list separator specified.
1268 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
1271 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
1274 JH/21 Support for the A6 type of dns record is withdrawn.
1276 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
1277 rather than the verbs used.
1279 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
1280 from 255 to 1024 chars.
1282 JH/24 Verification callouts now attempt to use TLS by default.
1284 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
1285 are generic router options now. The defaults didn't change.
1287 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
1288 Original patch from Alexander Shikoff, worked over by JH.
1290 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
1291 files if ZCAT_COMMAND is not executable.
1293 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
1295 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
1297 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
1298 Normally benign, it bites when the pair was led to by a CNAME;
1299 modern usage is to not canonicalize the domain to a CNAME target
1300 (and we were inconsistent anyway for A-only vs AAAA+A).
1302 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
1304 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
1305 when evaluating $sender_host_dnssec.
1307 JH/31 Check the HELO verification lookup for DNSSEC, adding new
1308 $sender_helo_dnssec variable.
1310 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
1312 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
1314 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
1316 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
1317 documented as working, but never had. Support all but $spam_report.
1319 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
1320 added for tls authenticator.
1322 HS/03 Add perl_taintmode main config option
1328 TL/01 When running the test suite, the README says that variables such as
1329 no_msglog_check are global and can be placed anywhere in a specific
1330 test's script, however it was observed that placement needed to be near
1331 the beginning for it to behave that way. Changed the runtest perl
1332 script to read through the entire script once to detect and set these
1333 variables, reset to the beginning of the script, and then run through
1334 the script parsing/test process like normal.
1336 TL/02 The BSD's have an arc4random API. One of the functions to induce
1337 adding randomness was arc4random_stir(), but it has been removed in
1338 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
1339 function when detected.
1341 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
1342 cause callback expansion.
1344 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
1345 syntax errors in an expansion can be treated as a string instead of
1346 logging or causing an error, due to the internal use of bool_lax
1347 instead of bool when processing it.
1349 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
1350 server certificates when making smtp deliveries.
1352 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
1354 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
1356 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
1358 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
1359 Merged patch from Sebastian Wiedenroth.
1361 JH/05 Fix results-pipe from transport process. Several recipients, combined
1362 with certificate use, exposed issues where response data items split
1363 over buffer boundaries were not parsed properly. This eventually
1364 resulted in duplicates being sent. This issue only became common enough
1365 to notice due to the introduction of connection certificate information,
1366 the item size being so much larger. Found and fixed by Wolfgang Breyha.
1368 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
1369 size buffer was used, resulting in syntax errors when an expansion
1372 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1373 version 3.3.6 or later.
1375 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1376 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1377 both become "event_action", the variables become $event_name, $event_data
1378 and $event_defer_errno. There is a new variable $verify_mode, usable in
1379 routers, transports and related events. The tls:cert event is now also
1380 raised for inbound connections, if the main configuration event_action
1383 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1384 early OCSP support, but no stapling (appears to be less than 1.0.0).
1386 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1387 server certificate names available under the smtp transport option
1388 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1391 JH/10 Time-related extraction expansions from certificates now use the main
1392 option "timezone" setting for output formatting, and are consistent
1393 between OpenSSL and GnuTLS compilations. Bug 1541.
1395 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1396 encoded parameter in the incoming message. Bug 1558.
1398 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1399 include certificate info, eximon was claiming there were spoolfile
1402 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1404 JH/14 Log delivery-related information more consistently, using the sequence
1405 "H=<name> [<ip>]" wherever possible.
1407 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1408 are problematic for Debian distribution, omit them from the release
1411 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1413 JH/16 Fix string representation of time values on 64bit time_t architectures.
1416 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1417 output list separator was used.
1422 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1423 checkers that were complaining about end of non-void function with no
1426 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1427 This was a regression introduced in 4.83 by another bugfix.
1429 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1431 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1432 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1438 TF/01 Correctly close the server side of TLS when forking for delivery.
1440 When a message was received over SMTP with TLS, Exim failed to clear up
1441 the incoming connection properly after forking off the child process to
1442 deliver the message. In some situations the subsequent outgoing
1443 delivery connection happened to have the same fd number as the incoming
1444 connection previously had. Exim would try to use TLS and fail, logging
1445 a "Bad file descriptor" error.
1447 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1448 utilities have not been installed.
1450 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1451 temporary space as the ACL may create new global variables.
1453 TL/01 LDAP support uses per connection or global context settings, depending
1454 upon the detected version of the libraries at build time.
1456 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1457 to extract and use the src ip:port in logging and expansions as if it
1458 were a direct connection from the outside internet. PPv2 support was
1459 updated based on HAProxy spec change in May 2014.
1461 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1463 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1464 Properly escape header and check for NULL return.
1466 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1469 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1471 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1472 characters in header names, implemented as a verify condition.
1473 Contributed by Michael Fischer v. Mollard.
1475 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1476 results permerror and temperror. Previous values are deprecated but
1477 still accepted. In a future release, err_perm and err_temp will be
1478 completely removed, which will be a backward incompatibility if the
1479 ACL tests for either of these two old results. Patch contributed by
1480 user bes-internal on the mailing list.
1482 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1484 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1485 selectors, in both main and reject logs.
1487 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1490 JH/07 Add malware type "sock" for talking to simple daemon.
1492 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1494 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1495 routers/transports under cutthrough routing.
1497 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1498 numbers. Touch up "bool" conditional to keep the same definition.
1500 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1502 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1504 JH/12 Expand items in router/transport headers_add or headers_remove lists
1505 individually rather than the list as a whole. Bug 1452.
1507 Required for reasonable handling of multiple headers_ options when
1508 they may be empty; requires that headers_remove items with embedded
1509 colons must have them doubled (or the list-separator changed).
1511 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1512 view the policy declared in the DMARC record. Currently, $dmarc_status
1513 is a combined value of both the record presence and the result of the
1516 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1518 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1519 dnslookup router and the smtp transport (applying to the forward
1522 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1523 of ldap servers used for a specific lookup. Patch provided by Heiko
1526 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1527 New variable $lookup_dnssec_authenticated for observability.
1529 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1530 Patch submitted by Lars Timman.
1532 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1534 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1535 Requires trusted mode and valid format message id, aborts otherwise.
1536 Patch contributed by Heiko Schlichting.
1538 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1539 certextract with support for various fields. Bug 1358.
1541 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1542 is requested by default, modifiable by smtp transport option
1545 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1546 operate on certificate variables to give certificate fingerprints
1547 Also new ${sha256:cert_variable}.
1549 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1551 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1552 Christian Aistleitner.
1554 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1556 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1557 file. Patch from Wolfgang Breyha.
1559 JH/25 Expand the coverage of the delivery $host and $host_address to
1560 client authenticators run in verify callout. Bug 1476.
1562 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1563 align with daemon_smtp_ports. Bug 72.
1565 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1566 support and error reporting did not work properly.
1568 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1569 and is readable. Patch from Andrew Colin Kissa.
1571 TL/14 Enhance documentation of ${run expansion and how it parses the
1572 commandline after expansion, particularly in the case when an
1573 unquoted variable expansion results in an empty value.
1575 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1577 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1578 by renaming away from it. Addresses GH issue 15, reported by
1581 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1583 TL/15 SECURITY: prevent double expansion in math comparison functions
1584 (can expand unsanitized data). Not remotely exploitable.
1591 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1593 PP/02 Make -n do something, by making it not do something.
1594 When combined with -bP, the name of an option is not output.
1596 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1599 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1600 $sender_host_name and config options to manage this, and basic check
1603 PP/05 DSCP support for outbound connections and control modifier for inbound.
1605 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1606 (Only plugin which currently uses this is kerberos4, which nobody should
1607 be using, but we should make it available and other future plugins might
1608 conceivably use it, even though it would break NAT; stuff *should* be
1609 using channel bindings instead).
1611 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1612 name; added for Sendmail compatibility; requires admin caller.
1613 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1614 just ignore it); requires trusted caller.
1615 Also parse but ignore: -Ac -Am -X<logfile>
1618 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1620 TL/02 Add +smtp_confirmation as a default logging option.
1622 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1623 Patch by Magnus Holmgren from 2007-02-20.
1625 TL/04 Bugzilla 1281 - Spec typo.
1626 Bugzilla 1283 - Spec typo.
1627 Bugzilla 1290 - Spec grammar fixes.
1629 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1631 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1633 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1634 mailing list by by Dmitry Isaikin.
1636 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1638 JH/02 Support "G" suffix to numbers in ${if comparisons.
1640 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1642 NM/01 Bugzilla 1197 - Spec typo
1643 Bugzilla 1196 - Spec examples corrections
1645 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1647 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1648 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1651 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1652 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1654 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1655 "acl {{name}{arg}...}", and optional args on acl condition
1658 JH/05 Permit multiple router/transport headers_add/remove lines.
1660 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1662 JH/07 Avoid using a waiting database for a single-message-only transport.
1663 Performance patch from Paul Fisher. Bugzilla 1262.
1665 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1668 JH/09 Add $headers_added variable, with content from use of ACL modifier
1669 add_header (but not yet added to the message). Bugzilla 199.
1671 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1672 Pulled from Bugzilla 817 by Wolfgang Breyha.
1674 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1676 (nb: this is the same fix as in Exim 4.80.1)
1678 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1681 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1684 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1686 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1687 advertise SMTP AUTH mechanism to us, instead of a generic
1688 protocol violation error. Also, make Exim more robust to bad
1689 data from the Dovecot auth socket.
1691 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1693 When a queue runner is handling a message, Exim first routes the
1694 recipient addresses, during which it prunes them based on the retry
1695 hints database. After that it attempts to deliver the message to
1696 any remaining recipients. It then updates the hints database using
1699 So if a recipient address works intermittently, it can get repeatedly
1700 deferred at routing time. The retry hints record remains fresh so the
1701 address never reaches the final cutoff time.
1703 This is a fairly common occurrence when a user is bumping up against
1704 their storage quota. Exim had some logic in its local delivery code
1705 to deal with this. However it did not apply to per-recipient defers
1706 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1708 This change adds a proper retry rule check during routing so that the
1709 final cutoff time is checked against the message's age. We only do
1710 this check if there is an address retry record and there is not a
1711 domain retry record; this implies that previous attempts to handle
1712 the address had the retry_use_local_parts option turned on. We use
1713 this as an approximation for the destination being like a local
1714 delivery, as in LMTP.
1716 I suspect this new check makes the old local delivery cutoff check
1717 redundant, but I have not verified this so I left the code in place.
1719 TF/02 Correct gecos expansion when From: is a prefix of the username.
1721 Test 0254 submits a message to Exim with the header
1725 When I ran the test suite under the user fanf2, Exim expanded
1726 the header to contain my full name, whereas it should have added
1727 a Resent-Sender: header. It erroneously treats any prefix of the
1728 username as equal to the username.
1730 This change corrects that bug.
1732 GF/01 DCC debug and logging tidyup
1733 Error conditions log to paniclog rather than rejectlog.
1734 Debug lines prefixed by "DCC: " to remove any ambiguity.
1736 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1738 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1739 Bug spotted by Jeremy Harris; was flawed since initial commit.
1740 Would have resulted in OCSP responses post-SNI triggering an Exim
1741 NULL dereference and crash.
1743 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1745 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1746 Bug detection, analysis and fix by Samuel Thibault.
1747 Bugzilla 1331, Debian bug #698092.
1749 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1751 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1752 Server implementation by Todd Lyons, client by JH.
1753 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1754 config variable "prdr_enable" controls whether the server
1755 advertises the facility. If the client requests PRDR a new
1756 acl_data_smtp_prdr ACL is called once for each recipient, after
1757 the body content is received and before the acl_smtp_data ACL.
1758 The client is controlled by both of: a hosts_try_prdr option
1759 on the smtp transport, and the server advertisement.
1760 Default client logging of deliveries and rejections involving
1761 PRDR are flagged with the string "PRDR".
1763 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1764 fclose(). Diagnosis by Todd Lyons.
1766 PP/17 Update configure.default to handle IPv6 localhost better.
1767 Patch by Alain Williams (plus minor tweaks).
1770 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1771 This is now consistent with GnuTLS, and is now documented: the
1772 previous undocumented portable approach to treating the option as
1773 unset was to force an expansion failure. That still works, and
1774 an empty string is now equivalent.
1776 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1777 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1778 not performing validation itself.
1780 PP/20 Added force_command boolean option to pipe transport.
1781 Patch from Nick Koston, of cPanel Inc.
1783 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1786 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1788 PP/21 Fix eximon continuous updating with timestamped log-files.
1789 Broken in a format-string cleanup in 4.80, missed when I repaired the
1790 other false fix of the same issue.
1791 Report and fix from Heiko Schlichting.
1794 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1795 Report from Prashanth Katuri.
1797 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1798 It's SecureTransport, so affects any MacOS clients which use the
1799 system-integrated TLS libraries, including email clients.
1801 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1802 using a MIME ACL for non-SMTP local injection.
1803 Report and assistance in diagnosis by Warren Baker.
1805 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1807 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1809 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1810 last id that failed so it may be referenced in subsequent ACL's.
1812 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1815 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1816 ldap library initialization, allowing self-signed CA's to be
1817 used. Also properly sets require_cert option later in code by
1818 using NULL (global ldap config) instead of ldap handle (per
1819 session). Bug diagnosis and testing by alxgomz.
1821 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1822 the src/util/ subdirectory.
1824 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1825 renamed to Transport Post Delivery Action by Jeremy Harris, as
1828 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1829 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1830 redis_servers = needs to be configured which will be used by the redis
1831 lookup. Patch from Warren Baker, of The Packet Hub.
1833 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1835 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1836 hostname or reverse DNS when processing a host list. Used suggestions
1837 from multiple comments on this bug.
1839 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1841 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1842 Missed a few lines, added it to make the runtest require no keyboard
1845 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1846 contains upper case chars. Make router use caseful_local_part.
1848 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1849 support when GnuTLS has been built with p11-kit.
1855 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1857 This, or similar/improved, will also be change PP/11 of 4.82.
1863 PP/01 Handle short writes when writing local log-files.
1864 In practice, only affects FreeBSD (8 onwards).
1865 Bugzilla 1053, with thanks to Dmitry Isaikin.
1867 NM/01 Bugzilla 949 - Documentation tweak
1869 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1872 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1874 PP/02 Implemented gsasl authenticator.
1876 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1878 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1879 `pkg-config foo` for cflags/libs.
1881 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1882 with rest of GSASL and with heimdal_gssapi.
1884 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1885 `pkg-config foo` for cflags/libs for the TLS implementation.
1887 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1888 properties get this fed in as external SSF. A number of robustness
1889 and debugging improvements to the cyrus_sasl authenticator.
1891 PP/08 cyrus_sasl server now expands the server_realm option.
1893 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1894 Patch by Jeremy Harris.
1896 PP/10 Added dbmjz lookup type.
1898 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1900 PP/12 MAIL args handles TAB as well as SP, for better interop with
1901 non-compliant senders.
1902 Analysis and variant patch by Todd Lyons.
1904 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1905 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1906 Patch from Dirk Mueller <dmueller@suse.com>
1908 PP/13 tls_peerdn now print-escaped for spool files.
1909 Observed some $tls_peerdn in wild which contained \n, which resulted
1910 in spool file corruption.
1912 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1913 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1914 or write after TLS renegotiation, which otherwise led to messages
1917 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1918 as a tracking header (ie: a signed header comes before the signature).
1919 Patch from Wolfgang Breyha.
1921 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1922 comma-sep list; embedded commas doubled.
1924 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1926 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1928 Report and patch from Dmitry Banschikov.
1930 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1931 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1932 failures. We appear to now support TLS1.1+ with Exim.
1934 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1935 lets Exim select keys and certificates based upon TLS SNI from client.
1936 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1937 before an outbound SMTP session. New log_selector, +tls_sni.
1939 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1940 NULL dereference. Report and patch from Alun Jones.
1942 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1943 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1944 Not seeing resolver debug output on NetBSD, but suspect this is a
1945 resolver implementation change.
1947 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1948 Left warnings. Added "eximon gdb" invocation mode.
1950 PP/21 Defaulting "accept_8bitmime" to true, not false.
1952 PP/22 Added -bw for inetd wait mode support.
1954 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1955 locate the relevant includes and libraries. Made this the default.
1957 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1958 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1960 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1961 This may cause build issues on older platforms.
1963 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1964 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1965 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1966 Added SNI support via GnuTLS too.
1967 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1969 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1971 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1972 Applied second patch from Janne, implementing suggestion to default
1973 multiple-strings-in-record handling to match SPF spec.
1975 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1977 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1978 read-only, out of scope).
1979 Patch from Wolfgang Breyha, report from Stuart Northfield.
1981 PP/29 Fix three issues highlighted by clang analyser static analysis.
1982 Only crash-plausible issue would require the Cambridge-specific
1983 iplookup router and a misconfiguration.
1984 Report from Marcin Mirosław.
1986 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1988 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1989 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1990 As part of this, removing so much warning spew let me fix some minor
1991 real issues in debug logging.
1993 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1994 assignment on my part. Fixed.
1996 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1997 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1998 Janne Snabb (who went above and beyond: thank you).
2000 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
2001 string otherwise requires a connection and a bunch more work and it's
2002 relatively easy to get wrong. Should also expose TLS library linkage
2005 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
2006 64-bit ${eval} (JH/03).
2008 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
2009 GNU libc to support some of the 64-bit stuff, should not lead to
2010 conflicts. Defined before os.h is pulled in, so if a given platform
2011 needs to override this, it can.
2013 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
2014 protection layer was required, which is not implemented.
2015 Bugzilla 1254, patch from Wolfgang Breyha.
2017 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
2018 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
2019 tls_dhparam take prime identifiers. Also unbreak combination of
2020 OpenSSL+DH_params+TLSSNI.
2022 PP/39 Disable SSLv2 by default in OpenSSL support.
2028 PP/01 Solaris build fix for Oracle's LDAP libraries.
2029 Bugzilla 1109, patch from Stephen Usher.
2031 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
2033 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
2036 TF/02 Fix a couple more cases where we did not log the error message
2037 when unlink() failed. See also change 4.74-TF/03.
2039 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
2040 lock up or crash if it happened to be inside a call to libc when it
2041 got a SIGUSR1 from exiwhat.
2043 The SIGUSR1 handler appends the current process status to the process
2044 log which is later printed by exiwhat. It used to use the general
2045 purpose logging code to do this, but several functions it calls are
2046 not safe for signals.
2048 The new output code in the SIGUSR1 handler is specific to the process
2049 log, and simple enough that it's easy to inspect for signal safety.
2050 Removing some special cases also simplifies the general logging code.
2051 Removing the spurious timestamps from the process log simplifies
2054 TF/04 Improved ratelimit ACL condition.
2056 The /noupdate option has been deprecated in favour of /readonly which
2057 has clearer semantics. The /leaky, /strict, and /readonly update modes
2058 are mutually exclusive. The update mode is no longer included in the
2059 database key; it just determines when the database is updated. (This
2060 means that when you upgrade Exim will forget old rate measurements.)
2062 Exim now checks that the per_* options are used with an update mode that
2063 makes sense for the current ACL. For example, when Exim is processing a
2064 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
2065 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
2066 must specify per_mail/readonly. If you omit the update mode it defaults to
2067 /leaky where that makes sense (as before) or /readonly where required.
2069 The /noupdate option is now undocumented but still supported for
2070 backwards compatibility. It is equivalent to /readonly except that in
2071 ACLs where /readonly is required you may specify /leaky/noupdate or
2072 /strict/noupdate which are treated the same as /readonly.
2074 A useful new feature is the /count= option. This is a generalization
2075 of the per_byte option, so that you can measure the throughput of other
2076 aggregate values. For example, the per_byte option is now equivalent
2077 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
2079 The per_rcpt option has been generalized using the /count= mechanism
2080 (though it's more complicated than the per_byte equivalence). When it is
2081 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
2082 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
2083 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
2084 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
2085 Note that using per_rcpt with a non-readonly update mode in more than
2086 one ACL will cause the recipients to be double-counted. (The per_mail
2087 and per_byte options don't have this problem.)
2089 The handling of very low rates has changed slightly. If the computed rate
2090 is less than the event's count (usually one) then this event is the first
2091 after a long gap. In this case the rate is set to the same as this event's
2092 count, so that the first message of a spam run is counted properly.
2094 The major new feature is a mechanism for counting the rate of unique
2095 events. The new per_addr option counts the number of different
2096 recipients that someone has sent messages to in the last time period. It
2097 behaves like per_rcpt if all the recipient addresses are different, but
2098 duplicate recipient addresses do not increase the measured rate. Like
2099 the /count= option this is a general mechanism, so the per_addr option
2100 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
2101 example, measure the rate that a client uses different sender addresses
2102 with the options per_mail/unique=$sender_address. There are further
2103 details in the main documentation.
2105 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
2107 TF/06 Removed a few PCRE remnants.
2109 TF/07 Automatically extract Exim's version number from tags in the git
2110 repository when doing development or release builds.
2112 PP/02 Raise smtp_cmd_buffer_size to 16kB.
2113 Bugzilla 879. Patch from Paul Fisher.
2115 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
2116 Heavily based on revision 40f9a89a from Simon Arlott's tree.
2119 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
2121 PP/05 Variable $av_failed, true if the AV scanner deferred.
2122 Bugzilla 1078. Patch from John Horne.
2124 PP/06 Stop make process more reliably on build failure.
2125 Bugzilla 1087. Patch from Heiko Schlittermann.
2127 PP/07 Make maildir_use_size_file an _expandable_ boolean.
2128 Bugzilla 1089. Patch from Heiko Schlittermann.
2130 PP/08 Handle ${run} returning more data than OS pipe buffer size.
2131 Bugzilla 1131. Patch from Holger Weiß.
2133 PP/09 Handle IPv6 addresses with SPF.
2134 Bugzilla 860. Patch from Wolfgang Breyha.
2136 PP/10 GnuTLS: support TLS 1.2 & 1.1.
2138 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
2141 PP/11 match_* no longer expand right-hand-side by default.
2142 New compile-time build option, EXPAND_LISTMATCH_RHS.
2143 New expansion conditions, "inlist", "inlisti".
2145 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
2147 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
2149 PP/14 fix log_write() format string regression from TF/03.
2150 Bugzilla 1152. Patch from Dmitry Isaikin.
2156 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
2158 PP/02 Harmonised TLS library version reporting; only show if debugging.
2159 Layout now matches that introduced for other libraries in 4.74 PP/03.
2161 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
2163 PP/04 New "dns_use_edns0" global option.
2165 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
2168 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
2169 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
2171 TK/01 Updated PolarSSL code to 0.14.2.
2172 Bugzilla 1097. Patch from Andreas Metzler.
2174 PP/07 Catch divide-by-zero in ${eval:...}.
2175 Fixes bugzilla 1102.
2177 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
2180 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
2181 format-string attack -- SECURITY: remote arbitrary code execution.
2183 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
2184 time unintentionally subject to list matching rules, letting the header
2185 cause arbitrary Exim lookups (of items which can occur in lists, *not*
2186 arbitrary string expansion). This allowed for information disclosure.
2188 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
2189 INT_MIN/-1 -- value coerced to INT_MAX.
2195 NM/01 Workaround for PCRE version dependency in version reporting
2198 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
2199 This fixes portability to compilers other than gcc, notably
2200 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
2202 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
2203 makefiles for portability to HP-UX and POSIX correctness.
2205 PP/01 Permit LOOKUP_foo enabling on the make command-line.
2206 Also via indented variable definition in the Makefile.
2207 (Debugging by Oliver Heesakkers).
2209 PP/02 Restore caching of spamd results with expanded spamd_address.
2210 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
2212 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
2213 Improves build reliability. Fix from: Frank Elsner
2215 NM/02 Fix wide character breakage in the rfc2047 coding
2216 Fixes bug 1064. Patch from Andrey N. Oktyabrski
2218 NM/03 Allow underscore in dnslist lookups
2219 Fixes bug 1026. Patch from Graeme Fowler
2221 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
2222 Code patches from Adam Ciarcinski of NetBSD.
2224 NM/04 Fixed exiqgrep to cope with mailq missing size issue
2227 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
2228 is logged, to avoid truncation. Patch from John Horne.
2230 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
2231 Patch from Jakob Hirsch.
2233 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
2234 SQL string expansion failure details.
2235 Patch from Andrey Oktyabrski.
2237 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
2238 Patch from Simon Arlott.
2240 PP/09 New lookups functionality failed to compile on old gcc which rejects
2241 extern declarations in function scope.
2242 Patch from Oliver Fleischmann
2244 PP/10 Use sig_atomic_t for flags set from signal handlers.
2245 Check getgroups() return and improve debugging.
2246 Fixed developed for diagnosis in bug 927 (which turned out to be
2249 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
2250 Patch from Mark Zealey.
2252 PP/12 Bugzilla 1056: Improved spamd server selection.
2253 Patch from Mark Zealey.
2255 PP/13 Bugzilla 1086: Deal with maildir quota file races.
2256 Based on patch from Heiko Schlittermann.
2258 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
2259 Patch from Uwe Doering, sign-off by Michael Haardt.
2261 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
2262 variable declaration deep within a block. Bug and patch from
2265 PP/15 lookups-Makefile IRIX compatibility coercion.
2267 PP/16 Make DISABLE_DKIM build knob functional.
2269 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
2270 Patch by Simon Arlott
2272 TF/03 Fix valgrind.h portability to C89 compilers that do not support
2273 variable argument macros. Our copy now differs from upstream.
2279 TF/01 Failure to get a lock on a hints database can have serious
2280 consequences so log it to the panic log.
2282 TF/02 Log LMTP confirmation messages in the same way as SMTP,
2283 controlled using the smtp_confirmation log selector.
2285 TF/03 Include the error message when we fail to unlink a spool file.
2287 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
2288 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
2289 for maintaining out-of-tree patches for some time.
2291 PP/01 Bugzilla 139: Documentation and portability issues.
2292 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
2293 Handle per-OS dynamic-module compilation flags.
2295 PP/02 Let /dev/null have normal permissions.
2296 The 4.73 fixes were a little too stringent and complained about the
2297 permissions on /dev/null. Exempt it from some checks.
2298 Reported by Andreas M. Kirchwitz.
2300 PP/03 Report version information for many libraries, including
2301 Exim version information for dynamically loaded libraries. Created
2302 version.h, now support a version extension string for distributors
2303 who patch heavily. Dynamic module ABI change.
2305 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
2306 privilege escalation vulnerability whereby the Exim run-time user
2307 can cause root to append content of the attacker's choosing to
2310 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
2313 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
2314 If dropping privileges for untrusted macros, we disabled normal logging
2315 on the basis that it would fail; for the Exim run-time user, this is not
2316 the case, and it resulted in successful deliveries going unlogged.
2317 Fixed. Reported by Andreas Metzler.
2323 PP/01 Date: & Message-Id: revert to normally being appended to a message,
2324 only prepend for the Resent-* case. Fixes regression introduced in
2325 Exim 4.70 by NM/22 for Bugzilla 607.
2327 PP/02 Include check_rfc2047_length in configure.default because we're seeing
2328 increasing numbers of administrators be bitten by this.
2330 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
2332 PP/03 Bugzilla 994: added openssl_options main configuration option.
2334 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
2336 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
2338 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
2340 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
2341 without a peer certificate, leading to a segfault because of an
2342 assumption that peers always have certificates. Be a little more
2343 paranoid. Problem reported by Martin Tscholak.
2345 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
2346 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
2347 NB: ClamAV planning to remove STREAM in "middle of 2010".
2348 CL also introduces -bmalware, various -d+acl logging additions and
2349 more caution in buffer sizes.
2351 PP/09 Implemented reverse_ip expansion operator.
2353 PP/10 Bugzilla 937: provide a "debug" ACL control.
2355 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
2357 PP/12 Bugzilla 973: Implement --version.
2359 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
2361 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
2363 PP/15 Bugzilla 816: support multiple condition rules on Routers.
2365 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
2366 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
2367 ignore trailing whitespace.
2369 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2371 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2374 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2375 Notification from Dr Andrew Aitchison.
2377 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2378 ExtendedDetectionInfo response format.
2379 Notification from John Horne.
2381 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2384 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2385 XSL and documented dependency on system catalogs, with examples of how
2388 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2391 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2392 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2393 configuration file which is writeable by the Exim user or group.
2395 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2396 of configuration files to cover files specified with the -C option if
2397 they are going to be used with root privileges, not just the default
2400 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2401 option (effectively making it always true).
2403 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2404 files to be used while preserving root privileges.
2406 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2407 that rogue child processes cannot use them.
2409 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2410 run-time user, instead of root.
2412 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2413 Exim run-time user without dropping privileges.
2415 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2416 result string, instead of calling string_vformat() twice with the same
2419 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2420 for other users. Others should always drop root privileges if they use
2421 -C on the command line, even for a whitelisted configure file.
2423 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2425 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2431 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2432 $data_path, and $header_path variables; fixed documentation bugs and
2435 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2436 exipick to access non-standard spools, including the "frozen" queue
2439 NM/01 Bugzilla 965: Support mysql stored procedures.
2440 Patch from Alain Williams
2442 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2444 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2445 Patch from Andreas Metzler
2447 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2448 Patch from Kirill Miazine
2450 NM/05 Bugzilla 671: Added umask to procmail example.
2452 JJ/03 installed exipick 20100323.0, fixing doc bug
2454 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2455 directory. Notification and patch from Dan Rosenberg.
2457 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2459 TK/02 Improve log output when DKIM signing operation fails.
2461 MH/01 Treat the transport option dkim_domain as a colon separated
2462 list, not as a single string, and sign the message with each element,
2463 omitting multiple occurences of the same signer.
2465 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2466 Bugzilla 985, 986. Patch by Simon Arlott
2468 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2469 Patch by Simon Arlott
2471 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2472 MBX locking. Notification from Dan Rosenberg.
2478 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2480 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2482 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2484 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2486 NM/04 Bugzilla 915: Flag broken perl installation during build.
2492 TK/01 Added patch by Johannes Berg that expands the main option
2493 "spamd_address" if it starts with a dollar sign.
2495 TK/02 Write list of recipients to X-Envelope-Sender header when building
2496 the mbox-format spool file for content scanning (suggested by Jakob
2499 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2500 (http://www.dcc-servers.net/) support via dccifd. Activated by
2501 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2503 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2504 by Mark Daniel Reidel <mr@df.eu>.
2506 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2507 When building exim an external PCRE library is now needed -
2508 PCRE is a system library on the majority of modern systems.
2509 See entry on PCRE_LIBS in EDITME file.
2511 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2512 conversation. Added nologin parameter to request.
2513 Patch contributed by Kirill Miazine.
2515 TF/01 Do not log submission mode rewrites if they do not change the address.
2517 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2519 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2520 log files in place. Contributed by Roberto Lima.
2522 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2524 TF/03 Bugzilla 615: When checking the local_parts router precondition
2525 after a local_part_suffix or local_part_prefix option, Exim now
2526 does not use the address's named list lookup cache, since this
2527 contains cached lookups for the whole local part.
2529 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2530 Robert Millan. Documentation is in experimental-spec.txt.
2532 TF/04 Bugzilla 668: Fix parallel build (make -j).
2534 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2536 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2537 Patch provided by Jan Srzednicki.
2539 TF/05 Leading white space used to be stripped from $spam_report which
2540 wrecked the formatting. Now it is preserved.
2542 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2543 that they are available at delivery time.
2545 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2547 TF/08 TLS error reporting now respects the incoming_interface and
2548 incoming_port log selectors.
2550 TF/09 Produce a more useful error message if an SMTP transport's hosts
2551 setting expands to an empty string.
2553 NM/06 Bugzilla 744: EXPN did not work under TLS.
2554 Patch provided by Phil Pennock.
2556 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2557 Patch provided by Richard Godbee.
2559 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2560 acl_smtp_notquit, added index entry.
2562 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2563 Patch provided by Eugene Bujak.
2565 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2566 accept(). Patch provided by Maxim Dounin.
2568 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2569 Patch provided by Phil Pennock.
2571 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2573 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2574 Patch provided by Brad "anomie" Jorsch.
2576 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2577 Patch provided by Dean Brooks.
2579 TK/05 Add native DKIM support (does not depend on external libraries).
2581 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2582 Patch provided by Graeme Fowler.
2584 NM/16 Bugzilla 851: Documentation example syntax fix.
2586 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2588 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2591 NM/19 Bugzilla 745: TLS version reporting.
2592 Patch provided by Phil Pennock.
2594 NM/20 Bugzilla 167: bool: condition support.
2595 Patch provided by Phil Pennock.
2597 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2598 clients. Patch provided by Phil Pennock.
2600 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2601 Patch provided by Brad "anomie" Jorsch.
2603 NM/23 Bugzilla 687: Fix misparses in eximstats.
2604 Patch provided by Heiko Schlittermann.
2606 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2607 Patch provided by Heiko Schlittermann.
2609 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2610 plus update to original patch.
2612 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2614 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2615 Patch provided by David Brownlee.
2617 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2619 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2621 NM/30 Bugzilla 888: TLS documentation bugfixes.
2623 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2625 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2626 Unlike the original bugzilla I have changed all shell scripts in src tree.
2628 NM/33 Bugzilla 898: Transport filter timeout fix.
2629 Patch by Todd Rinaldo.
2631 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2632 Patch by Serge Demonchaux.
2634 NM/35 Bugzilla 39: Base64 decode bug fixes.
2635 Patch by Jakob Hirsch.
2637 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2639 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2641 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2643 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2649 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2650 ALT-N's libdkim that I have put here:
2651 http://duncanthrax.net/exim-experimental/
2653 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2654 were called 'true' and it seems that C99 defines that as a reserved
2655 keyword to be used with 'bool' variable types. That means you could
2656 not include C99-style headers which use bools without triggering
2657 build errors in sieve.c.
2659 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2660 as mailq or other aliases. Changed the --help handling significantly
2661 to do whats expected. exim_usage() emits usage/help information.
2663 SC/01 Added the -bylocaldomain option to eximstats.
2665 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2667 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2669 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2675 PH/01 Another patch from the Sieve maintainer.
2677 PH/02 When an IPv6 address is converted to a string for single-key lookup
2678 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2679 dots are used instead of colons so that keys in lsearch files need not
2680 contain colons. This was done some time before quoting was made available
2681 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2682 (notated using the quote facility) so as to distinguish them from IPv4
2683 keys. This meant that lookups for IP addresses in host lists did not work
2684 for iplsearch lookups.
2686 This has been fixed by arranging for IPv6 addresses to be expressed with
2687 colons if the lookup type is iplsearch. This is not incompatible, because
2688 previously such lookups could never work.
2690 The situation is now rather anomalous, since one *can* have colons in
2691 ordinary lsearch keys. However, making the change in all cases is
2692 incompatible and would probably break a number of configurations.
2694 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2697 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2698 conversion specification without a maximum field width, thereby enabling
2699 a rogue spamd server to cause a buffer overflow. While nobody in their
2700 right mind would setup Exim to query an untrusted spamd server, an
2701 attacker that gains access to a server running spamd could potentially
2702 exploit this vulnerability to run arbitrary code as the Exim user.
2704 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2705 $primary_hostname instead of what libspf2 thinks the hosts name is.
2707 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2708 a directory entry by the name of the lookup key. Previously, if a
2709 symlink pointed to a non-existing file or a file in a directory that
2710 Exim lacked permissions to read, a lookup for a key matching that
2711 symlink would fail. Now it is enough that a matching directory entry
2712 exists, symlink or not. (Bugzilla 503.)
2714 PH/03 The body_linecount and body_zerocount variables are now exported in the
2717 PH/04 Added the $dnslist_matched variable.
2719 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2720 This means they are set thereafter only if the connection becomes
2723 PH/06 Added the client_condition to authenticators so that some can be skipped
2724 by clients under certain conditions.
2726 PH/07 The error message for a badly-placed control=no_multiline_responses left
2727 "_responses" off the end of the name.
2729 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2731 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2732 (without spaces) instead of just copying the configuration text.
2734 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2736 PH/11 Added $max_received_linelength.
2738 PH/12 Added +ignore_defer and +include_defer to host lists.
2740 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2741 way in which PCRE > 7.0 is built.
2743 PH/14 Implemented queue_only_load_latch.
2745 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2746 MAIL command. The effect was to mangle the value on 64-bit systems.
2748 PH/16 Another patch from the Sieve maintainer.
2750 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2752 PH/18 If a system quota error occurred while trying to create the file for
2753 a maildir delivery, the message "Mailbox is full" was not appended to the
2754 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2755 only to a quota excession during the actual writing of the file.
2757 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2758 characters?) which causes problems in log lines. The DN values are now
2759 passed through string_printing() before being added to log lines.
2761 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2762 and InterBase are left for another time.)
2764 PH/21 Added message_body_newlines option.
2766 PH/22 Guard against possible overflow in moan_check_errorcopy().
2768 PH/23 POSIX allows open() to be a macro; guard against that.
2770 PH/24 If the recipient of an error message contained an @ in the local part
2771 (suitably quoted, of course), incorrect values were put in $domain and
2772 $local_part during the evaluation of errors_copy.
2778 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2779 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2782 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2783 issue a MAIL command.
2785 PH/02 In an ACL statement such as
2787 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2789 if a client was not listed at all, or was listed with a value other than
2790 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2791 the condition was not true (as it should be), so access was not denied.
2792 The bug was that the ! inversion was incorrectly passed on to the second
2793 item. This has been fixed.
2795 PH/03 Added additional dnslists conditions == and =& which are different from
2796 = and & when the dns lookup returns more than one IP address.
2798 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2799 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2801 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2802 FSYNC, which compiles an option called disable_fsync that allows for
2803 bypassing fsync(). The documentation is heavily laced with warnings.
2805 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2807 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2808 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2809 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2810 including adding "make clean"; (3) Added -fPIC when compiling the test
2811 dynamically loaded module, to get rid of a warning.
2813 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2814 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2815 = 0s. The bug is otherwise harmless.
2817 PH/07 There was a bug in the dovecot authenticator such that the value of
2818 $auth1 could be overwritten, and so not correctly preserved, after a
2819 successful authentication. This usually meant that the value preserved by
2820 the server_setid option was incorrect.
2822 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2824 PH/09 Installed PCRE release 7.0.
2826 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2827 run for batched SMTP input. It is now run at the start of every message
2828 in the batch. While fixing this I discovered that the process information
2829 (output by running exiwhat) was not always getting set for -bs and -bS
2830 input. This is fixed, and it now also says "batched" for BSMTP.
2832 PH/11 Added control=no_pipelining.
2834 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2835 patch, slightly modified), and move the expansion of helo_data till after
2836 the connection is made in the smtp transport (so it can use these
2839 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2841 PH/14 Added log_selector = +pid.
2843 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2845 PH/16 Add ${if forany and ${if forall.
2847 PH/17 Added dsn_from option to vary the From: line in DSNs.
2849 PH/18 Flush SMTP output before performing a callout, unless control =
2850 no_callout_flush is set.
2852 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2853 was true (the default) a successful delivery failed to delete the retry
2854 item, thus causing premature timeout of the address. The bug is now
2857 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2859 PH/21 Long custom messages for fakedefer and fakereject are now split up
2860 into multiline responses in the same way that messages for "deny" and
2861 other ACL rejections are.
2863 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2864 with slight modification.
2866 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2867 draft, changing the syntax and factoring some duplicate code".
2869 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2870 for deliveries of the second and subsequent messages over the same SMTP
2873 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2874 ${reduce, with only minor "tidies".
2876 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2878 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2879 expansion side effects.
2881 PH/27 When a message times out after an over-quota error from an Exim-imposed
2882 quota, the bounce message says "mailbox is full". This message was not
2883 being given when it was a system quota that was exceeded. It now should
2886 MH/03 Made $recipients available in local_scan(). local_scan() already has
2887 better access to the recipient list through recipients_list[], but
2888 $recipients can be useful in postmaster-provided expansion strings.
2890 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2891 in the case of a MAIL command with additional options following the
2892 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2893 were accidentally chopped off.
2895 PH/29 SMTP synchronization checks are implemented when a command is read -
2896 there is a check that no more input is waiting when there shouldn't be
2897 any. However, for some commands, a delay in an ACL can mean that it is
2898 some time before the response is written. In this time, more input might
2899 arrive, invalidly. So now there are extra checks after an ACL has run for
2900 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2901 pipelining has not been advertised.
2903 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2905 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2906 returned in the SMTP response when smtp_return_error_details was false.
2907 This has been fixed.
2909 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2910 instead of the C I/O that was originally supplied, because problems were
2911 reported on Solaris.
2913 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2914 Exim which did not show up earlier: it was assuming that a call to
2915 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2916 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2917 was a macro that became an assignment, so it seemed to work. This has
2918 changed to a proper function call with a void return, hence the compile
2919 error. Exim's code has been fixed.
2921 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2924 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2926 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2928 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2929 because it turns out to be too big or there is a timeout) there is no
2930 "Completed" line in the log. When some messages of this type were
2931 selected by exigrep, they were listed as "not completed". Others were
2932 picked up by some special patterns. I have improved the selection
2933 criteria to be more general.
2935 PH/38 The host_find_failed option in the manualroute router can now be set
2936 to "ignore", to completely ignore a host whose IP address cannot be
2937 found. If all hosts are ignored, the behaviour is controlled by the new
2938 host_all_ignored option.
2940 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2941 homing or because of multiple MX records with /mx) generated more than
2942 one IP address, and the following item turned out to be the local host,
2943 all the secondary addresses of the first item were incorrectly removed
2944 from the list, along with the local host and any following hosts (which
2945 is what is supposed to happen).
2947 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2948 whoever called Exim into the -H file. In the case of the daemon it was
2949 behaving confusingly. When first started, it used values for whoever
2950 started the daemon, but after a SIGHUP it used the Exim user (because it
2951 calls itself on a restart). I have changed the code so that it now always
2954 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2955 message are rejected with the same error (e.g. no authentication or bad
2956 sender address), and a DATA command is nevertheless sent (as can happen
2957 with PIPELINING or a stupid MUA), the error message that was given to the
2958 RCPT commands is included in the rejection of the DATA command. This is
2959 intended to be helpful for MUAs that show only the final error to their
2962 PH/42 Another patch from the Sieve maintainer.
2964 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2965 Eximstats - Fixed some broken HTML links and added missing column headers
2967 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2968 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2970 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2972 PH/43 Yet another patch from the Sieve maintainer.
2974 PH/44 I found a way to check for a TCP/IP connection going away before sending
2975 the response to the final '.' that terminates a message, but only in the
2976 case where the client has not sent further data following the '.'
2977 (unfortunately, this is allowed). However, in many cases there won't be
2978 any further data because there won't be any more messages to send. A call
2979 to select() can be used: if it shows that the input is "ready", there is
2980 either input waiting, or the socket has been closed. An attempt to read
2981 the next input character can distinguish the two cases. Previously, Exim
2982 would have sent an OK response which the client would never have see.
2983 This could lead to message repetition. This fix should cure that, at
2984 least in a lot of common cases.
2986 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2987 advertised in response to EHLO.
2993 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2994 fixed by 4.65/MH/01 (is this a record?) are fixed:
2996 (i) An empty string was always treated as zero by the numeric comparison
2997 operators. This behaviour has been restored.
2999 (ii) It is documented that the numeric comparison operators always treat
3000 their arguments as decimal numbers. This was broken in that numbers
3001 starting with 0 were being interpreted as octal.
3003 While fixing these problems I realized that there was another issue that
3004 hadn't been noticed. Values of message_size_limit (both the global option
3005 and the transport option) were treated as octal if they started with 0.
3006 The documentation was vague. These values are now always treated as
3007 decimal, and I will make that clear in the documentation.
3013 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
3014 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
3017 MH/01 Don't check that the operands of numeric comparison operators are
3018 integers when their expansion is in "skipping" mode (fixes bug
3019 introduced by 4.64-PH/07).
3021 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
3022 child addresses, Exim now panics and dies. Previously, because the count
3023 is held in a short int, deliveries were likely to be lost. As such a
3024 large number of recipients for a single message is ridiculous
3025 (performance will be very, very poor), I have chosen to impose a limit
3026 rather than extend the field.
3032 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
3033 leftover -K file (the existence of which was triggered by #402).
3034 While we were at it, introduced process PID as part of the -K
3035 filename. This should rule out race conditions when creating
3038 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
3039 processing considerably. Previous code took too long for large mails,
3040 triggering a timeout which in turn triggers #401.
3042 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
3043 in the DK code in transports.c. sendfile() is not really portable,
3044 hence the _LINUX specificness.
3046 TF/01 In the add_headers option to the mail command in an Exim filter,
3047 there was a bug that Exim would claim a syntax error in any
3048 header after the first one which had an odd number of characters
3051 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
3052 callout verification, Exim cached a "reject" for the entire domain. This
3053 is correct for most verifications, but it is not correct for a recipient
3054 verification with use_sender or use_postmaster set, because in that case
3055 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
3056 case of MAIL FROM:<> rejection from other early rejections (e.g.
3057 rejection of HELO). When verifying a recipient using a non-null MAIL
3058 address, the cache is ignored if it shows MAIL FROM:<> rejection.
3059 Whatever the result of the callout, the value of the domain cache is
3060 left unchanged (for any other kind of callout, getting as far as trying
3061 RCPT means that the domain itself is ok).
3063 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
3066 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
3067 manifest itself as EPIPE rather than ECONNECT. When tidying away a
3068 session, the daemon ignores ECONNECT errors and logs others; it now
3069 ignores EPIPE as well.
3071 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
3072 (quoted-printable decoding).
3074 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
3075 later the small subsequent patch to fix an introduced bug.
3077 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
3079 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
3081 PH/08 An error is now given if message_size_limit is specified negative.
3083 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
3084 to be given (somewhat) arbitrary names.
3086 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
3089 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
3090 miscellaneous code fixes
3092 PH/10 Added the log_reject_target ACL modifier to specify where to log
3095 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
3096 hostname. This is wrong, because it relates to the incoming message (and
3097 probably the interface on which it is arriving) and not to the outgoing
3098 callout (which could be using a different interface). This has been
3099 changed to use the value of the helo_data option from the smtp transport
3100 instead - this is what is used when a message is actually being sent. If
3101 there is no remote transport (possible with a router that sets up host
3102 addresses), $smtp_active_hostname is used.
3104 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
3105 tweaks were necessary in order to get it to work (see also 21 below):
3106 (a) The code assumed that strncpy() returns a negative number on buffer
3107 overflow, which isn't the case. Replaced with Exim's string_format()
3109 (b) There were several signed/unsigned issues. I just did the minimum
3110 hacking in of casts. There is scope for a larger refactoring.
3111 (c) The code used strcasecmp() which is not a standard C function.
3112 Replaced with Exim's strcmpic() function.
3113 (d) The code set only $1; it now sets $auth1 as well.
3114 (e) A simple test gave the error "authentication client didn't specify
3115 service in request". It would seem that Dovecot has changed its
3116 interface. Fortunately there's a specification; I followed it and
3117 changed what the client sends and it appears to be working now.
3119 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
3122 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
3123 address A is aliased to B and C, where B exists and C does not. Without
3124 -v the output is "A verified" because verification stops after a
3125 successful redirection if more than one address is generated. However,
3126 with -v the child addresses are also verified. Exim was outputting "A
3127 failed to verify" and then showing the successful verification for C,
3128 with its parentage. It now outputs "B failed to verify", showing B's
3129 parentage before showing the successful verification of C.
3131 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
3132 look up a TXT record in a specific list after matching in a combined
3135 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
3136 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
3137 they consult the DNS. I had assumed they would set it the way they
3138 wanted; and indeed my experiments on Linux seem to show that in some
3139 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
3140 To be on the safe side, however, I have now made the interface to
3141 host_find_byname() similar to host_find_bydns(), with an argument
3142 containing the DNS resolver options. The host_find_byname() function now
3143 sets these options at its start, just as host_find_bydns() does. The smtp
3144 transport options dns_qualify_single and dns_search_parents are passed to
3145 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
3146 of host_find_byname() use the default settings of RES_DEFNAMES
3147 (qualify_single) but not RES_DNSRCH (search_parents).
3149 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
3150 spool_read_header() do less string testing, by means of a preliminary
3151 switch on the second character of optional "-foo" lines. (This is
3152 overdue, caused by the large number of possibilities that now exist.
3153 Originally there were few.) While I was there, I also converted the
3154 str(n)cmp tests so they don't re-test the leading "-" and the first
3155 character, in the hope this might squeeze out yet more improvement.
3157 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
3158 flag allowing group syntax was set by the header_syntax check but not
3159 turned off, possible causing trouble later; (2) The flag was not being
3160 set at all for the header_verify test, causing "group"-style headers to
3161 be rejected. I have now set it in this case, and also caused header_
3162 verify to ignore an empty address taken from a group. While doing this, I
3163 came across some other cases where the code for allowing group syntax
3164 while scanning a header line wasn't quite right (mostly, not resetting
3165 the flag correctly in the right place). These bugs could have caused
3166 trouble for malformed header lines. I hope it is now all correct.
3168 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
3169 with the "reply" argument non-NULL. The code, however (which originally
3170 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
3171 but it didn't always do it. This confused somebody who was copying the
3172 code for some other use. I have removed all the tests.
3174 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
3175 feature that was used to support insecure browsers during the U.S. crypto
3176 embargo. It requires special client support, and Exim is probably the
3177 only MTA that supported it -- and would never use it because real RSA is
3178 always available. This code has been removed, because it had the bad
3179 effect of slowing Exim down by computing (never used) parameters for the
3180 RSA_EXPORT functionality.
3182 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
3183 authenticator to fail if there's a tab character in the incoming data
3184 (there should never be unless someone is messing about, as it's supposed
3185 to be base64-encoded). Also added, on Timo's advice, the "secured" option
3186 if the connection is using TLS or if the remote IP is the same as the
3187 local IP, and the "valid-client-cert option" if a client certificate has
3190 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
3191 authenticators. This can be used for authorization after authentication
3192 succeeds. (In the case of plaintext, it servers for both authentication
3195 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
3196 if any retry times were supplied.
3198 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
3199 connection, where there is no client IP address to check. In this
3200 situation, the verify now always succeeds.
3202 PH/25 Applied John Jetmore's -Mset patch.
3204 PH/26 Added -bem to be like -Mset, but loading a message from a file.
3206 PH/27 In a string expansion for a processed (not raw) header when multiple
3207 headers of the same name were present, leading whitespace was being
3208 removed from all of them, but trailing whitespace was being removed only
3209 from the last one. Now trailing whitespace is removed from each header
3210 before concatenation. Completely empty headers in a concatenation (as
3211 before) are ignored.
3213 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
3214 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
3216 PH/29 [Removed. This was a change that I later backed out, and forgot to
3217 correct the ChangeLog entry (that I had efficiently created) before
3218 committing the later change.]
3220 PH/30 Exim was sometimes attempting to deliver messages that had suffered
3221 address errors (4xx response to RCPT) over the same connection as other
3222 messages routed to the same hosts. Such deliveries are always "forced",
3223 so retry times are not inspected. This resulted in far too many retries
3224 for the affected addresses. The effect occurred only when there were more
3225 hosts than the hosts_max_try setting in the smtp transport when it had
3226 the 4xx errors. Those hosts that it had tried were not added to the list
3227 of hosts for which the message was waiting, so if all were tried, there
3228 was no problem. Two fixes have been applied:
3230 (i) If there are any address or message errors in an SMTP delivery, none
3231 of the hosts (tried or untried) are now added to the list of hosts
3232 for which the message is waiting, so the message should not be a
3233 candidate for sending over the same connection that was used for a
3234 successful delivery of some other message. This seems entirely
3235 reasonable: after all the message is NOT "waiting for some host".
3236 This is so "obvious" that I'm not sure why it wasn't done
3237 previously. Hope I haven't missed anything, but it can't do any
3238 harm, as the worst effect is to miss an optimization.
3240 (ii) If, despite (i), such a delivery is accidentally attempted, the
3241 routing retry time is respected, so at least it doesn't keep
3242 hammering the server.
3244 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
3245 in ${readsocket because some servers need this prod.
3247 PH/32 Added some extra debug output when updating a wait-xxx database.
3249 PH/33 The hint "could be header name not terminated by colon", which has been
3250 given for certain expansion errors for a long time, was not being given
3251 for the ${if def:h_colon_omitted{... case.
3253 PH/34 The spec says: "With one important exception, whenever a domain list is
3254 being scanned, $domain contains the subject domain." There was at least
3255 one case where this was not true.
3257 PH/35 The error "getsockname() failed: connection reset by peer" was being
3258 written to the panic log as well as the main log, but it isn't really
3259 panic-worthy as it just means the connection died rather early on. I have
3260 removed the panic log writing for the ECONNRESET error when getsockname()
3263 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
3264 runs only) independently of the message's sender address. This meant
3265 that, if the 4xx error was in fact related to the sender, a different
3266 message to the same recipient with a different sender could confuse
3267 things. In particular, this can happen when sending to a greylisting
3268 server, but other circumstances could also provoke similar problems.
3269 I have changed the default so that the retry time for these errors is now
3270 based a combination of the sender and recipient addresses. This change
3271 can be overridden by setting address_retry_include_sender=false in the
3274 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
3275 remote server are returned as part of bounce messages. This was not
3276 happening for LMTP over a pipe (the lmtp transport), but now it is the
3277 same for both kinds of LMTP.
3279 PH/38 Despite being documented as not happening, Exim was rewriting addresses
3280 in header lines that were in fact CNAMEs. This is no longer the case.
3282 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
3283 and queue runs started by the daemon processed all messages. This has
3284 been fixed so that -R and -S can now usefully be given with -q<time>.
3286 PH/40 Import PCRE release 6.7 (fixes some bugs).
3288 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
3290 PH/42 Give an error if -q is specified more than once.
3292 PH/43 Renamed the variables $interface_address and $interface_port as
3293 $received_ip_address and $received_port, to make it clear that these
3294 values apply to message reception, and not to the outgoing interface when
3295 a message is delivered. (The old names remain recognized, of course.)
3297 PH/44 There was no timeout on the connect() call when using a Unix domain
3298 socket in the ${readsocket expansion. There now is.
3300 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
3301 be meaningful with "accept".
3303 SC/01 Eximstats V1.43
3304 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
3306 SC/02 Eximstats V1.44
3307 Use a glob alias rather than an array ref in the generated
3308 parser. This improves both readability and performance.
3310 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
3311 Collect SpamAssassin and rejection statistics.
3312 Don't display local sender or destination tables unless
3313 there is data to show.
3314 Added average volumes into the top table text output.
3316 SC/04 Eximstats V1.46
3317 Collect data on the number of addresses (recipients)
3318 as well as the number of messages.
3320 SC/05 Eximstats V1.47
3321 Added 'Message too big' to the list of mail rejection
3322 reasons (thanks to Marco Gaiarin).
3324 SC/06 Eximstats V1.48
3325 Mainlog lines which have GMT offsets and are too short to
3326 have a flag are now skipped.
3328 SC/07 Eximstats V1.49 (Alain Williams)
3329 Added the -emptyok flag.
3331 SC/08 Eximstats V1.50
3332 Fixes for obtaining the IP address from reject messages.
3334 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
3335 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
3336 whitespace changes from 4.64-PH/27
3338 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
3341 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
3342 are found, allow negative numbers in numeric criteria)
3344 JJ/06 exipick.20061117.2, added new $message_body_missing variable
3346 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
3347 to match changes made in 4.64-PH/43
3349 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
3351 PH/47 Put in an explicit test for a DNS lookup of an address record where the
3352 "domain" is actually an IP address, and force a failure. This locks out
3353 those revolvers/nameservers that support "A-for-A" lookups, in
3354 contravention of the specifications.
3356 PH/48 When a host name was looked up from an IP address, and the subsequent
3357 forward lookup of the name timed out, the host name was left in
3358 $sender_host_name, contrary to the specification.
3360 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
3361 restricted to single-key lookups, Exim was not diagnosing an error if
3362 * or *@ was used with a query-style lookup.
3364 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
3366 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
3367 long ago, but noone interested enough thought of it. Let's just say that
3368 the "1.1" means that there are some new functions that weren't there at
3369 some point in the past.
3371 PH/51 Error processing for expansion failure of helo_data from an smtp
3372 transport during callout processing was broken.
3374 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3375 tested/used via the -bh/-bhc/-bs options.
3377 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3378 bug, fixed in subsequent PCRE releases).
3380 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3381 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3383 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3389 SC/01 Use a glob alias rather than an array ref in eximstats generated
3390 parser. This improves both readability and performance.
3392 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3393 Don't display local sender or destination tables in eximstats unless
3394 there is data to show.
3395 Added average volumes into the eximstats top table text output.
3397 SC/03 Collect data on the number of addresses (recipients) as well
3398 as the number of messages in eximstats.
3400 TF/01 Correct an error in the documentation for the redirect router. Exim
3401 does (usually) call initgroups() when daemonizing.
3403 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3404 with consistent privilege compared to when running as a daemon.
3406 TF/03 Note in the spec that $authenticated_id is not set for local
3407 submissions from trusted users.
3409 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3410 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3412 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3413 by adding some example configuration directives to the default
3414 configuration file. A little bit of work is required to uncomment the
3415 directives and define how usernames and passwords are checked, but
3416 there is now a framework to start from.
3418 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3419 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3420 without this. I don't know how relevant this is to other LDAP libraries.
3422 PH/02 Add the verb name to the "unknown ACL verb" error.
3424 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3426 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3428 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3429 directory not expanded when it should be if an expanded home directory
3430 was set for the address (which is overridden by the transport).
3432 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3435 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3436 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3437 because it is too late at that time, and has no effect.
3439 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3440 security issue with \' (bugzilla #107). I could not use the
3441 PQescapeStringConn() function, because it needs a PGconn value as one of
3444 PH/08 When testing addresses using -bt, indicate those final addresses that
3445 are duplicates that would not cause an additional delivery. At least one
3446 person was confused, thinking that -bt output corresponded to deliveries.
3447 (Suppressing duplicates isn't a good idea as you lose the information
3448 about possibly different redirections that led to the duplicates.)
3450 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3451 systems where poll() doesn't work, in particular OS X.
3453 PH/10 Added more information to debugging output for retry time not reached.
3455 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3456 operations in malware.c.
3458 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3461 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3462 syslog_duplication set false, log lines that would normally be written
3463 both the the main log and to the reject log were not written to syslog at
3466 PH/14 In the default configuration, change the use of "message" in ACL warn
3467 statements to "add_header".
3469 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3470 not followed by a command (e.g. "seen endif").
3472 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3473 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3476 PH/17 Added extra conditions to the default value of delay_warning_condition
3480 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3481 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3482 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3485 The Auto-Submitted: and various List- headers are standardised, whereas I
3486 don't think Precedence: ever was.
3488 PH/18 Refactored debugging code in route_finduser() to show more information,
3489 in particular, the error code if getpwnam() issues one.
3491 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3492 This is apparently needed in addition to the PH/07 change above to avoid
3493 any possible encoding problems.
3495 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3496 but not after initializing Perl.
3498 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3499 output them only if debugging. By default they are written stderr,
3500 apparently, which is not desirable.
3502 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3505 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3508 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3510 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3511 authenticated or an ident call has been made. Suppress the default
3512 values for $authenticated_id and $authenticated_sender (but permit -oMai
3513 and -oMas) when testing with -bh.
3515 PH/24 Re-jigged the order of the tests in the default configuration so that the
3516 tests for valid domains and recipients precede the DNS black list and CSA
3517 tests, on the grounds that those ones are more expensive.
3519 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3520 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3521 This bug exists in every version of Exim that I still have, right back to
3524 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3525 However, an attempt to turn on case-sensitivity in a regex key by
3526 including (?-i) didn't work because the subject string was already
3527 lowercased, and the effects were non-intuitive. It turns out that a
3528 one-line patch can be used to allow (?-i) to work as expected.
3534 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3535 other effects) broke the use of negated acl sub-conditions.
3537 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3540 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3541 "Deny" causes Exim to reject the incoming connection with a 554 error.
3542 Unfortunately, if there is a major crisis, such as a disk failure,
3543 tcp-wrappers gives "deny", whereas what one would like would be some
3544 kind of temporary error. A kludge has been added to help with this.
3545 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3546 554 error is used if errno is still zero or contains ENOENT (which occurs
3547 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3550 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3552 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3553 errors. Otherwise a message that provokes a temporary error (when other
3554 messages do not) can cause a whole host to time out.
3556 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3557 addresses were routed directly to files or pipes from a redirect router.
3558 File deliveries just didn't batch; pipe deliveries might have suffered
3561 PH/06 A failure to get a lock for a hints database would erroneously always say
3562 "Failed to get write lock", even when it was really a read lock.
3564 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3565 of 0600. This has been changed to use the value of the lockfile_mode
3566 option (which defaults to 0600).
3568 PH/08 Applied small patch from the Sieve maintainer.
3570 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3571 folder from quota calculations, a direct delivery into this folder messed
3572 up the contents of the maildirsize file. This was because the regex was
3573 used only to exclude .Trash (or whatever) when the size of the mailbox
3574 was calculated. There was no check that a delivery was happening into an
3575 excluded directory. This bug has been fixed by ignoring all quota
3576 processing for deliveries into excluded directories.
3578 PH/10 Added the maildirfolder_create_regex option to appendfile.
3584 PH/01 The code for finding all the local interface addresses on a FreeBSD
3585 system running IPv6 was broken. This may well have applied to all BSD
3586 systems, as well as to others that have similar system calls. The broken
3587 code found IPv4 interfaces correctly, but gave incorrect values for the
3588 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3589 that it would not match correctly against @[] and not recognize the IPv6
3592 PH/02 The ipliteral router was not recognizing addresses of the form user@
3593 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3595 PH/03 Added disable_ipv6.
3597 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3598 decoded form, because it is most often used to construct To: headers
3599 lines in autoreplies, and the decoded form may well be syntactically
3600 invalid. However, $reply_address has leading white space removed, and all
3601 newlines turned into spaces so that the autoreply transport does not
3604 PH/05 If group was specified without a user on a router, and no group or user
3605 was specified on a transport, the group from the router was ignored.
3607 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3608 for visible compile-time settings that can be used to change these
3609 numbers, for those that want even more. Backwards compatibility with old
3610 spool files has been maintained. However, going back to a previous Exim
3611 release will lost any variables that are in spool files.
3613 PH/07 Two small changes when running in the test harness: increase delay when
3614 passing a TCP/IP connection to a new process, in case the original
3615 process has to generate a bounce, and remove special handling of
3616 127.0.0.2 (sic), which is no longer necessary.
3618 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3619 be the same on different OS.
3621 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3624 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3625 whether --show-vars was specified or not
3627 JJ/02 exipick: Added support for new ACL variable spool format introduced
3630 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3631 syntactically invalid From: or Reply-to: line, and a filter used this to
3632 generate an autoreply, and therefore failed to obtain an address for the
3633 autoreply, Exim could try to deliver to a non-existent relative file
3634 name, causing unrelated and misleading errors. What now happens is that
3635 it logs this as a hard delivery error, but does not attempt to create a
3638 PH/11 The exinext utility has a -C option for testing purposes, but although
3639 the given file was scanned by exinext itself; it wasn't being passed on
3640 when Exim was called.
3642 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3643 an end-of-file indication when reading a command response.
3645 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3646 compiled. In many other places in Exim, IPv6 addresses are always
3647 recognized, so I have changed this. It also means that IPv4 domain
3648 literals of the form [IPV4:n.n.n.n] are now always recognized.
3650 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3651 used if the router is not running as root, for example, when verifying at
3652 ACL time, or when using -bh. The debugging output from this situation was
3653 non-existent - all you got was a failure to exec. I have made two
3656 (a) Failures to set uid/gid, the current directory, or a process leader
3657 in a subprocess such as that created by queryprogram now generate
3658 suitable debugging output when -d is set.
3660 (b) The queryprogram router detects when it is not running as root,
3661 outputs suitable debugging information if -d is set, and then runs
3662 the subprocess without attempting to change uid/gid.
3664 PH/15 Minor change to Makefile for building test_host (undocumented testing
3667 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3668 additional section of a DNS packet that returns MX or SRV records.
3669 Instead, it always explicitly searches for A/AAAA records. This avoids
3670 major problems that occur when a DNS server includes only records of one
3671 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3672 fixed another bug: if SRV records were looked up and the corresponding
3673 address records were *not* found in the additional section, the port
3674 values from the SRV records were lost.
3676 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3677 using the correct key (the original address) when searching the retry
3678 rules in order to find which one to use for generating the retry hint.
3680 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3681 adding the default one. Similarly, if it contains a Reply-To: header, the
3682 errors_reply_to option, if set, is not used.
3684 PH/19 When calculating a retry time, Exim used to measure the "time since
3685 failure" by looking at the "first failed" field in the retry record. Now
3686 it does not use this if it is later than than the arrival time of the
3687 message. Instead it uses the arrival time. This makes for better
3688 behaviour in cases where some deliveries succeed, thus re-setting the
3689 "first failed" field. An example is a quota failure for a huge message
3690 when small messages continue to be delivered. Without this change, the
3691 "time since failure" will always be short, possible causing more frequent
3692 delivery attempts for the huge message than are intended.
3693 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3695 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3696 $1, $2, $3) because the numerical variables can be reset during some
3697 expansion items (e.g. "match"), thereby losing the authentication data.
3699 PH/21 Make -bV show the size of off_t variables so that the test suite can
3700 decide whether to run tests for quotas > 2G.
3702 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3703 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3704 filecount value is greater than 2G or if a quota value is greater than 2G
3705 on a system where the size of off_t is not greater than 4, a panic error
3708 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3709 never match. The debug and -bh output now contains an explicit error
3710 message indicating a malformed IPv4 address or mask.
3712 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3713 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3714 PH/23 above applies.
3716 PH/25 Do not write to syslog when running in the test harness. The only
3717 occasion when this arises is a failure to open the main or panic logs
3718 (for which there is an explicit test).
3720 PH/26 Added the /no_tell option to "control=freeze".
3722 PH/27 If a host name lookup failed very early in a connection, for example, if
3723 the IP address matched host_lookup and the reverse lookup yielded a name
3724 that did not have a forward lookup, an error message of the form "no IP
3725 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3726 could be logged. Now it outputs the IP address instead of "NULL".
3728 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3729 allows the sender and the authenticated sender to be set when
3730 submitting a message from within Exim. Since child_open_exim() is
3731 documented for local_scan(), the new function should be too.
3733 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3734 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3735 results in an empty string is now treated as unset.
3737 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3739 PH/31 Added sender_verify_fail logging option.
3741 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3742 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3743 tidied the source and removed it altogether.
3745 PH/33 When a queue run was abandoned because the load average was too high, a
3746 log line was always written; now it is written only if the queue_run log
3747 selector is set. In addition, the log line for abandonment now contains
3748 information about the queue run such as the pid. This is always present
3749 in "start" and "stop" lines but was omitted from the "abandon" line.
3751 PH/34 Omit spaces between a header name and the colon in the error message that
3752 is given when verify = headers_syntax fails (if there are lots of them,
3753 the message gets confusing).
3755 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3756 names, as there are now some PTR records that contain slashes. This check
3757 is only to protect against broken name servers that fall over on strange
3758 characters, so the fact that it applies to all lookups doesn't matter.
3760 PH/36 Now that the new test suite is complete, we can remove some of the
3761 special code in Exim that was needed for the old test suite. For example,
3762 sorting DNS records because real resolvers return them in an arbitrary
3763 order. The new test suite's fake resolver always returns records in the
3766 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3767 bounces) except when queue_only is set, to avoid logging races between
3768 the different processes.
3770 PH/38 Panic-die if .include specifies a non-absolute path.
3772 PH/39 A tweak to the "H" retry rule from its user.
3774 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3775 a label. They prevented compilation on older perls.
3777 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3778 a warning to be raised on newish perls.
3780 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3781 on queue. Changes to match documented behaviour of showing count of
3782 messages matching specified criteria.
3784 PH/40 Changed the default ident timeout from 30s to 5s.
3786 PH/41 Added support for the use of login_cap features, on those BSD systems
3787 that have them, for controlling the resources used by pipe deliveries.
3789 PH/42 The content-scanning code uses fopen() to create files in which to put
3790 message data. Previously it was not paying any attention to the mode of
3791 the files. Exim runs with umask(0) because the rest of the code creates
3792 files with open(), and sets the required mode explicitly. Thus, these
3793 files were ending up world-writeable. This was not a big issue, because,
3794 being within the spool directory, they were not world-accessible. I have
3795 created a function called modefopen, which takes an additional mode
3796 argument. It sets umask(777), creates the file, chmods it to the required
3797 mode, then resets the umask. All the relevant calls to fopen() in the
3798 content scanning code have been changed to use this function.
3800 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3801 to 24 hours. This avoids potential overflow problems when processing G
3802 and H retry rules. I suspect nobody ever tinkers with this value.
3804 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3806 PH/45 When the plaintext authenticator is running as a client, the server's
3807 challenges are checked to ensure they are valid base64 strings. By
3808 default, the authentication attempt is cancelled if an invalid string is
3809 received. Setting client_ignore_invalid_base64 true ignores these errors.
3810 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3811 they are received. Thus, the responses can be made to depend on the
3812 challenges. If an invalid string is ignored, an empty string is placed in
3815 PH/46 Messages that are created by the autoreply transport now contains a
3816 References: header, in accordance with RFCs 2822 and 3834.
3818 PH/47 Added authenticated_sender_force to the smtp transport.
3820 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3822 PH/49 Installed latest patch from the Sieve maintainer.
3824 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3825 was also set, the appendfile transport was unnecessarily scanning a
3826 directory of message files (e.g. for maildir delivery) to find the count
3827 of files (along with the size), even though it did not need this
3828 information. It now does the scan only if it needs to find either the
3829 size of the count of files.
3831 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3833 PH/52 Two bugs concerned with error handling when the smtp transport is
3836 (i) Exim was not creating retry information for temporary errors given
3837 for individual recipients after the DATA command when the smtp transport
3838 was used in LMTP mode. This meant that they could be retried too
3839 frequently, and not timed out correctly.
3841 (ii) Exim was setting the flag that allows error details to be returned
3842 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3843 recipients that were returned after the DATA command.
3845 PH/53 This is related to PH/52, but is more general: for any failing address,
3846 when detailed error information was permitted to be returned to the
3847 sender, but the error was temporary, then after the final timeout, only
3848 "retry timeout exceeded" was returned. Now it returns the full error as
3849 well as "retry timeout exceeded".
3851 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3852 do this, and (what is worse) MTAs that accept it.
3854 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3855 will now be deprecated.
3857 PH/56 New os.c-cygwin from the Cygwin maintainer.
3859 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3860 formats (previously only available in exim formats via -bpr, -bpru,
3861 and -bpra. Now also available in native and exiqgrep formats)
3863 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3864 with very large, slow to parse queues
3866 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3868 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3870 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3871 responses to authentication challenges, though it was showing the
3872 challenges; (ii) I've removed the CR characters from the debug output for
3875 PH/58 Allow for the insertion of a newline as well as a space when a string
3876 is turned into more than one encoded-word during RFC 2047 encoding. The
3877 Sieve code now uses this.
3879 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3880 data_4xx, lost_connection, tls_required.
3882 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3883 message was being sent as an SMTP response.
3885 PH/61 Add -l and -k options to exicyclog.
3887 PH/62 When verifying, if an address was redirected to one new address, so that
3888 verification continued, and the new address failed or deferred after
3889 having set something in $address_data, the value of $address_data was not
3890 passed back to the ACL. This was different to the case when no
3891 redirection occurred. The value is now passed back in both cases.
3893 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3894 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3895 use login_cap.h, so on its own it isn't the distinguishing feature. The
3896 new name refers directly to the setclassresources() function.
3898 PH/65 Added configuration files for NetBSD3.
3900 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3902 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3903 is preferred over IPv4.
3905 PH/68 The bounce_return_message and bounce_return_body options were not being
3906 honoured for bounces generated during the reception of non-SMTP messages.
3907 In particular, this applied to messages rejected by the ACL. This bug has
3908 been fixed. However, if bounce_return_message is true and bounce_return_
3909 body is false, the headers that are returned for a non-SMTP message
3910 include only those that have been read before the error was detected.
3911 (In the case of an ACL rejection, they have all been read.)
3913 PH/69 The HTML version of the specification is now built in a directory called
3914 spec_html instead of spec.html, because the latter looks like a path with
3915 a MIME-type, and this confuses some software.
3917 PH/70 Catch two compiler warnings in sieve.c.
3919 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3920 function verify_get_ident() calls ip_connect() to connect a socket, but
3921 if the "connect()" function timed out, ip_connect() used to close the
3922 socket. However, verify_get_ident() also closes the socket later, and in
3923 between Exim writes to the log, which may get opened at this point. When
3924 the socket was closed in ip_connect(), the log could get the same file
3925 descriptor number as the socket. This naturally causes chaos. The fix is
3926 not to close the socket in ip_connect(); the socket should be closed by
3927 the function that creates it. There was only one place in the code where
3928 this was missing, in the iplookup router, which I don't think anybody now
3929 uses, but I've fixed it anyway.
3931 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3932 well as to direct DNS lookups. Otherwise the handling of names in host
3933 lists is inconsistent and therefore confusing.
3939 PH/01 Two changes to the default runtime configuration:
3941 (1) Move the checks for relay_from_hosts and authenticated clients from
3942 after to before the (commented out) DNS black list checks.
3944 (2) Add control=submission to the relay_from_hosts and authenticated
3945 clients checks, on the grounds that messages accepted by these
3946 statements are most likely to be submissions.
3948 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3950 (1) Generate an error if the third argument for the ${prvs expansion is
3953 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3956 (3) Reset the variables that are obtained from the first argument of
3957 ${prvscheck and used in the second argument before leaving the code,
3958 because their memory is reclaimed, so using them afterwards may do
3961 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3962 one (it's much easier than Tom thought :-).
3964 (5) Because of (4), we can now allow for the use of $prvscheck_result
3965 inside the third argument.
3967 PH/03 For some reason, the default setting of PATH when running a command from
3968 a pipe transport was just "/usr/bin". I have changed it to
3971 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3972 anything to be listed in the output from -bV.
3974 PH/05 When a filter generated an autoreply, the entire To: header line was
3975 quoted in the delivery log line, like this:
3977 => >A.N.Other <ano@some.domain> <original@ddress> ...
3979 This has been changed so that it extracts the operative address. There
3980 may be more than one such address. If so, they are comma-separated, like
3983 => >ano@some.domain,ona@other.domain <original@ddress> ...
3985 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3986 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3987 not being looked up in the rDNS to get a host name, Exim was showing the
3988 IP address twice in Received: lines, even though the IP addresses were
3989 identical. For example:
3991 Received: from [1.2.3.4] (helo=[1.2.3.4])
3993 However, if the real host name was known, it was omitting the HELO data
3994 if it matched the actual IP address. This has been tidied up so that it
3995 doesn't show the same IP address twice.
3997 PH/07 When both +timestamp and +memory debugging was on, the value given by
3998 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3999 called by the debug printing, thereby overwriting the timestamp buffer.
4000 Debugging no longer uses the tod_stamp() function when +timestamp is set.
4002 PH/08 When the original message was included in an autoreply transport, it
4003 always said "this is a copy of the message, including all the headers",
4004 even if body_only or headers_only was set. It now gives an appropriate
4007 PH/09 Applied a patch from the Sieve maintainer which:
4009 o fixes some comments
4010 o adds the (disabled) notify extension core
4011 o adds some debug output for the result of if/elsif tests
4012 o points to the current vacation draft in the documentation
4013 and documents the missing references header update
4017 o fixes a bug in processing the envelope test (when testing
4018 multiple envelope elements, the last element determined the
4021 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
4022 Electronic Mail") by including:
4024 Auto-submitted: auto-generated
4026 in the messages that it generates (bounce messages and others, such as
4027 warnings). In the case of bounce messages for non-SMTP messages, there was
4028 also a typo: it was using "Auto_submitted" (underscore instead of
4029 hyphen). Since every message generated by Exim is necessarily in response
4030 to another message, thes have all been changed to:
4032 Auto-Submitted: auto-replied
4034 in accordance with these statements in the RFC:
4036 The auto-replied keyword:
4038 - SHOULD be used on messages sent in direct response to another
4039 message by an automatic process,
4041 - MUST NOT be used on manually-generated messages,
4043 - MAY be used on Delivery Status Notifications (DSNs) and Message
4044 Disposition Notifications (MDNs),
4046 - MUST NOT be used on messages generated by automatic or periodic
4047 processes, except for messages which are automatic responses to
4050 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
4051 to the default Received: header definition.
4053 PH/12 Added log selector acl_warn_skipped (default on).
4055 PH/13 After a successful wildlsearch lookup, discard the values of numeric
4056 variables because (a) they are in the wrong storage pool and (b) even if
4057 they were copied, it wouldn't work properly because of the caching.
4059 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
4060 checking when decoding. Apparently there are clients that generate
4061 overlong encoded strings. Why am I not surprised?
4063 PH/15 If the first argument of "${if match_address" was not empty, but did not
4064 contain an "@" character, Exim crashed. Now it writes a panic log message
4065 and treats the condition as false.
4067 PH/16 In autoreply, treat an empty string for "once" the same as unset.
4069 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
4070 extension "envelope-auth". The code is finished and in agreement with
4071 other implementations, but there is no documentation so far and in fact,
4072 nobody wrote the draft yet. This extension is currently #undef'ed, thus
4073 not changing the active code.
4075 Print executed "if" and "elsif" statements when debugging is used. This
4076 helps a great deal to understand what a filter does.
4078 Document more things not specified clearly in RFC3028. I had all this
4079 sorted out, when out of a sudden new issues came to my mind. Oops."
4081 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
4084 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
4085 canonical form (as documented). However, after a host name lookup from
4086 the IP address, check_host() was doing a simple string comparison with
4087 addresses acquired from the DNS when checking that the found name did
4088 have the original IP as one of its addresses. Since any found IPv6
4089 addresses are likely to be in abbreviated form, the comparison could
4090 fail. Luckily, there already exists a function for doing the comparison
4091 by converting both addresses to binary, so now that is used instead of
4092 the text comparison.
4094 PH/20 There was another similar case to PH/19, when a complete host name was
4095 given in a host list; looking up its IP address could give an abbreviated
4096 form, whereas the current host's name might or might not be abbreviated.
4097 The same fix has been applied.
4103 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
4104 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
4107 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
4108 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
4110 PH/03 Typo: missing ".o" in src/pcre/Makefile.
4112 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
4113 header line, restrict the check to what is listed in RFCs 2369 and 2929.
4114 Also, for "Auto-Submitted", treat anything other than "no" as
4115 non-personal, in accordance with RFC 3834. (Previously it treated
4116 anything starting "auto-" as non-personal.)
4118 TF/01 The control=submission/name=... option had a problem with syntax
4119 errors if the name included a slash character. The /name= option
4120 now slurps the rest of the string, so it can include any characters
4121 but it must come last in the list of options (after /sender_retain
4124 PH/05 Some modifications to the interface to the fake nameserver for the new
4132 TK/01 Added the "success_on_redirect" address verification option. See
4133 NewStuff for rationale and an example.
4135 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
4137 PH/02 Patch to exigrep to allow it to work on syslog lines.
4139 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
4140 fread() to skip over the body file's header line, because in Cygwin the
4141 header line is locked and is inaccessible.
4143 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
4144 co-exist for some time) to make it clear that it is the Exim ID that is
4145 referenced, not the Message-ID: header line.
4147 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
4148 string_format() function, because snprintf() does not exist on all
4151 PH/06 The use of forbid_filter_existstest now also locks out the use of the
4152 ${stat: expansion item.
4154 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
4155 protocol synchronization error", to keep the pedants happy.
4157 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
4158 well as for IRIX systems, when gcc is being used. See the host.c source
4161 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
4163 PH/10 Named domain lists were not working if used in a queue_smtp_domains
4166 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
4167 transport and to the smtp transport in LMTP mode.
4169 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
4171 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
4172 run a filter in a subprocess. This could lead to confusion in subsequent
4173 lookups in the parent process. There should also be a search_tidyup() at
4174 the end of the subprocess.
4176 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
4177 only if the host matched helo_try_verify_hosts, which caused the
4178 verification to occur when the EHLO/HELO command was issued. The ACL just
4179 tested the remembered result. Now, if a previous verification attempt has
4180 not happened, "verify = helo" does it there and then.
4182 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
4184 TK/03 Fix log output including CR from clamd.
4186 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
4187 exist provoked a memory error which could cause a segfault.
4189 PH/15 Installed PCRE 6.2
4191 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
4193 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
4194 of the problem. Specifically, suggested +O2 rather than +O1 for the
4197 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
4199 PH/20 If a delivery was routed to a non-standard port by means of an SRV
4200 record, the port was not correctly logged when the outgoing_port log
4201 selector was set (it logged the transort's default port).
4203 PH/21 Added support for host-specific ports to manualroute, queryprogram,
4204 fallback_hosts, and "hosts" in the smtp transport.
4206 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
4207 host errors such as "Connection refused".
4209 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
4210 authentication with radiusclient 0.4.9:
4212 - Error returned from rc_read_config was caught wrongly
4213 - Username/password not passed on to radius server due to wrong length.
4215 The presumption is that some radiusclient API changes for 4.51/PH/17
4216 were not taken care of correctly. The code is still untested by me (my
4217 Linux distribution still has 0.3.2 of radiusclient), but it was
4218 contributed by a Radius user.
4220 PH/24 When doing a callout, the value of $domain wasn't set correctly when
4221 expanding the "port" option of the smtp transport.
4223 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
4224 while reading a MIME header. Thanks to Tom Hughes for a patch.
4226 PH/24 Include config.h inside local_scan.h so that configuration settings are
4229 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
4230 that in an ACL for RCPT (for example), you can examine exactly what was
4233 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
4234 commands, but it was not correctly comparing the address with the actual
4235 client host address. Thus, it would show the EHLO address in Received:
4236 header lines when this was not necessary.
4238 PH/27 Added the % operator to ${eval:}.
4240 PH/28 Exim tries to create and chdir to its spool directory when it starts;
4241 it should be ignoring failures (because with -C, for example, it has lost
4242 privilege). It wasn't ignoring creation failures other than "already
4245 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
4248 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
4249 because an input line was too long, either on its own, or by virtue of
4250 too many continuations, the temporary file was not being removed, and the
4251 return code was incorrect.
4253 PH/31 Missing "BOOL" in function definition in filtertest.c.
4255 PH/32 Applied Sieve patches from the maintainer.
4257 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
4259 PH/33 Added "verify = not_blind".
4261 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
4262 Local/Makefile (with some defaults set). These are used in built scripts
4263 such as exicyclog, but they have never been used in the exim_install
4264 script (though there are many overriding facilities there). I have
4265 arranged that the exim_install script now takes note of these two
4268 PH/35 Installed configuration files for Dragonfly.
4270 PH/36 When a locally submitted message by a trusted user did not contain a
4271 From: header, and the sender address was obtained from -f or from an SMTP
4272 MAIL command, and the trusted user did not use -F to supply a sender
4273 name, $originator_name was incorrectly used when constructing a From:
4274 header. Furthermore, $originator_name was used for submission mode
4275 messages from external hosts without From: headers in a similar way,
4276 which is clearly wrong.
4278 PH/37 Added control=suppress_local_fixups.
4280 PH/38 When log_selector = +received_sender was set, and the addition of the
4281 sender made the log line's construction buffer exactly full, or one byte
4282 less than full, an overflow happened when the terminating "\n" was
4285 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
4286 when the result of a list match is failure because a DNS lookup failed.
4288 PH/40 RM_COMMAND is now used in the building process.
4290 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
4291 the "build-* directories that it finds.
4293 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
4294 address in a domain literal was a prefix of an interface address.
4296 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
4297 when verifying a sender address, unless rewrite_headers is false.
4299 PH/44 Wrote a long comment about why errors_to addresses are verified as
4300 recipients, not senders.
4302 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
4303 the ratelimit ACL was added.
4305 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
4307 PH/46 Added extra information about PostgreSQL errors to the error string.
4309 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
4310 avoiding the need to install special zones in a real server. This is
4311 backwards compatible; if it can't find the fake resolver, it drops back.
4312 Thus, both old and new test suites can be run.
4314 TF/02 Added util/ratelimit.pl
4316 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
4317 clock is set back in time.
4319 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
4320 Candler <B.Candler@pobox.com>.
4322 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
4323 OK for addresses that are the result of redirections.
4325 PH/48 A number of further additions for the benefit of the new test suite,
4326 including a fake gethostbyname() that interfaces to the fake DNS resolver
4329 TF/06 The fix for widen_domains has also been applied to qualify_single and
4330 search_parents which are the other dnslookup options that can cause
4333 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
4336 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
4338 TF/07 Exim produced the error message "an SRV record indicated no SMTP
4339 service" if it encountered an MX record with an empty target hostname.
4340 The message is now "an MX or SRV record indicated no SMTP service".
4342 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
4343 if the DNS of the sending site is misconfigured. This is quite a
4344 common situation. This change restores the behaviour of treating a
4345 helo verification defer as a failure.
4347 PH/51 If self=fail was set on a router, the bounce message did not include the
4348 actual error message.
4354 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
4356 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
4357 command itself ended in error, the underlying message about the transport
4358 filter timeout was being overwritten with the pipe command error. Now the
4359 underlying error message should be appended to the second error message.
4361 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
4363 PH/02 Reduce the amount of output that "make" produces by default. Full output
4364 can still be requested.
4366 PH/03 The warning log line about a condition test deferring for a "warn" verb
4367 was being output only once per connection, rather than after each
4368 occurrence (because it was using the same function as for successful
4369 "warn" verbs). This seems wrong, so I have changed it.
4371 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4372 it should not have, which might have caused a crash in the right
4373 circumstances, but probably never did.
4375 PH/04 Installed a modified version of Tony Finch's patch to make submission
4376 mode fix the return path as well as the Sender: header line, and to
4377 add a /name= option so that you can make the user's friendly name appear
4380 TF/03 Added the control = fakedefer ACL modifier.
4382 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4383 Mark Lowes for thorough testing.
4385 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4387 TK/03 Merged latest SRS patch from Miles Wilton.
4389 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4390 with the definition in sysexits.h (which is #included earlier).
4391 Fortunately, Exim does not actually use EX_OK. The code used to try to
4392 preserve the sysexits.h value, by assuming that macro definitions were
4393 scanned for macro replacements. I have been disabused of this notion,
4394 so now the code just undefines EX_OK before #including unistd.h.
4396 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4397 in the smtp transport. When a block could not be written in a single
4398 write() function, the timeout was being re-applied to each part-write.
4399 This seems wrong - if the receiver was accepting one byte at a time it
4400 would take for ever. The timeout is now adjusted when this happens. It
4401 doesn't have to be particularly precise.
4403 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4404 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4406 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4407 without a domain if the check to <postmaster@domain> fails.
4409 SC/01 Eximstats: added -xls and the ability to specify output files
4410 (patch written by Frank Heydlauf).
4412 SC/02 Eximstats: use FileHandles for outputting results.
4414 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4416 SC/04 Eximstats: fixed display of large numbers with -nvr option
4418 SC/05 Eximstats: fixed merging of reports with empty tables.
4420 SC/06 Eximstats: added the -include_original_destination flag
4422 SC/07 Eximstats: removed tabs and trailing whitespace.
4424 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4426 TK/06 MBOX spool code: Add real "From " MBOX separator line
4427 so the .eml file is really in mbox format (even though
4428 most programs do not really care). Patch from Alex Miller.
4430 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4431 The latter is generated from $received_to and is only set if the
4432 message has one envelope recipient. SA can use these headers,
4433 obviously out-of-the-box. Patch from Alex Miller.
4435 PH/08 The ${def test on a variable was returning false if the variable's
4436 value was "0", contrary to what the specification has always said!
4437 The result should be true unless the variable is empty.
4439 PH/09 The syntax error of a character other than { following "${if
4440 def:variable_name" (after optional whitespace) was not being diagnosed.
4441 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4442 accidental colon was present, for example, could give incorrect results.
4444 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4445 result is used (not including appendfile, where other changes are about
4448 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4449 This involved changing a lot of size variables from int to off_t. It
4450 should work with maildirs and everything.
4452 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4453 spamd dying while we are connected to it.
4455 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4458 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4459 function for BDB 4.3.
4461 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4463 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4464 the sources for PCRE have been re-arranged and more files are now
4467 PH/15 The code I had for printing potentially long long variables in PH/11
4468 above was not the best (it lost precision). The length of off_t variables
4469 is now inspected at build time, and an appropriate printing format (%ld
4470 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4471 to be "long long int" or "long int". This is needed for the internal
4472 formatting function string_vformat().
4474 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4475 the configuration file to be ":syslog", then the script "guesses" where
4476 the logs files are, rather than using the compiled in default. In our
4477 case the guess is not the same as the compiled default, so the script
4478 suddenly stopped working when I started to use syslog. The patch checks
4479 to see if log_file_path is "". If so, it attempts to read it from exim
4480 with no configuration file to get the compiled in version, before it
4481 falls back to the previous guessing code."
4483 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4484 implementing BATV in an Exim configuration. See NewStuff for the gory
4487 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4488 Makefile that are specific to HP-UX.
4490 PH/18 If the "use_postmaster" option was set for a recipient callout together
4491 with the "random" option, the postmaster address was used as the MAIL
4492 FROM address for the random test, but not for the subsequent recipient
4493 test. It is now used for both.
4495 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4496 patch removes a few documentation additions to RFC 3028, because the
4497 latest draft now contains them. It adds the new en;ascii-case comparator
4498 and a new error check for 8bit text in MIME parts. Comparator and
4499 require names are now matched exactly. I enabled the subaddress
4500 extension, but it is not well tested yet (read: it works for me)."
4502 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4503 rework some of the code of TK/09 above to avoid the hardwired use of
4504 "%lld" and "long long". Replaced the call to snprintf() with a call to
4507 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4508 records point to non-existent hosts", "retry timeout exceeded", and
4509 "retry time not reached for any host after a long failure period".
4511 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4512 experimental DomainKeys support:
4514 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4515 (2) On an error such as an illegally used "control", the wrong name for
4516 the control was given.
4518 These problems did NOT occur unless DomainKeys support was compiled.
4520 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4522 PH/24 Added ${if match_ip condition.
4524 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4525 databases so that it will be absolutely obvious if a crash occurs in the
4526 DB library. This is a regular occurrence (often caused by mis-matched
4529 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4530 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4531 picked up on a user's system that detects such things. There doesn't seem
4532 to be a gcc warning option for this - only an attribute that has to be
4533 put on the function's prototype. It seems that in Fedora Core 4 they have
4534 set this on a number of new functions. No doubt there will be more in due
4537 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4538 specify a transport. However, if an address that was verified by such a
4539 router was the subject of a callout, Exim crashed because it tried to
4540 read the rcpt_include_affixes from the non-existent transport. Now it
4541 just assumes that the setting of that option is false. This bug was
4542 introduced by 4.51/PH/31.
4544 PH/28 Changed -d+all to exclude +memory, because that information is very
4545 rarely of interest, but it makes the output a lot bigger. People tend to
4546 do -d+all out of habit.
4548 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4549 code in os-type was giving problems when libc.so lives in lib64, like on
4552 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4553 aren't the modern standard, and it seems that some systems' include files
4554 don't always have them. Exim was already checking for some of the newer
4555 ones like T_AAAA, and defining it itself. I've added checks for all the
4556 record types that Exim uses.
4558 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4559 not automatically generating a new one, as it is supposed to. This
4560 prevented TLS from working. If the file did exist, but contained invalid
4561 data, a new version was generated, as expected. It was only the case of a
4562 non-existent file that was broken.
4564 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4565 with a change in libdomainkeys > 0.64.
4567 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4568 from DNS. If the selector record carries the flag, it now has
4569 precedence over the domain-wide flag.
4571 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4573 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4574 the use of a port name that isn't defined in /etc/services) occurred, the
4575 message was deferred as in a normal delivery, and thus remained on the
4576 spool, instead of being failed because of the mua_wrapper setting. This
4577 is now fixed, and I tidied up some of the mua_wrapper messages at the
4580 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4581 the messages in a hash of arrays rather than using individual hashes.
4582 This is a bit cleaner and results in dramatic memory savings, albeit
4583 at a slight CPU cost.
4585 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4586 as requested by Marc Sherman.
4588 SC/10 Eximstats: added histograms for user specified patterns as requested
4591 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4593 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4594 fopen() in the content-scanning modules that did not already have it.
4600 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4601 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4603 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4605 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4607 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4608 to test Sieve filters that use "vacation".
4610 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4611 that changes the way the GnuTLS parameters are stored in the cache file.
4612 The new format can be generated externally. For backward compatibility,
4613 if the data in the cache doesn't make sense, Exim assumes it has read an
4614 old-format file, and it generates new data and writes a new file. This
4615 means that you can't go back to an older release without removing the
4618 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4619 work if there are any delivery delays because "one_time" forces the
4620 parent to be marked "delivered", so its unseen clone is never tried
4621 again. For this reason, Exim now forbids the simultaneous setting of
4624 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4625 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4626 entry if you want to know the details. The fix, however, neglected to
4627 consider the case when local delivery batching is involved. The test for
4628 "previously delivered" was not happening when checking to see if an
4629 address could be batched with a previous (undelivered) one; under
4630 certain circumstances this could lead to multiple deliveries to the same
4633 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4634 in its include files, and this causes problems building Exim.
4636 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4637 header_syntax) but Exim was just ignoring anything given after a slash.
4638 In particular, this caused confusion with an attempt to use "verify =
4639 reverse_host_lookup/defer_ok". An error is now given when options are
4640 supplied for verify items that do not have them. (Maybe reverse_host_
4641 lookup should have a defer_ok option, but that's a different point.)
4643 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4644 defined by RFC 821) to 2048, because there were problems with some AUTH
4645 commands, and RFC 1869 says the size should be increased for extended
4646 SMTP commands that take arguments.
4648 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4651 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4652 "unknown" error; now it says that the functionality isn't in the binary.
4654 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4655 an address' error message when a string expansion fails (syntax or
4656 whatever). Otherwise the password may appear in the log. Following change
4657 PH/42 below, there is no longer a chance of it appearing in a bounce
4660 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4662 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4663 of its addresses was ever tried. (Bugzilla bug #2.)
4665 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4666 the result incorrectly in the debug output. (It correctly added a newline
4667 to what was transported.)
4669 TF/01 Added $received_time.
4671 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4672 commented out examples of how to interface to a virus scanner and to
4673 SpamAssassin. Also added commented examples of av_scanner and
4674 spamd_address settings.
4676 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4677 and controls are allowed in which ACLs. There were a couple of minor
4678 errors. Some of the entries in the conditions table (which is a table of
4679 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4680 negation of where the condition IS allowed.
4682 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4684 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4685 header file does not have a version number, so I've had to invent a new
4686 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4687 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4688 radiusclient), but it was contributed by a Radius user.
4690 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4691 files or directories, for OpenSSL.
4693 PH/19 When an Exim process that is running as root has to create an Exim log
4694 file, it does so in a subprocess that runs as exim:exim so as to get the
4695 ownership right at creation (otherwise, other Exim processes might see
4696 the file with the wrong ownership). There was no test for failure of this
4697 fork() call, which would lead to the process getting stuck as it waited
4698 for a non-existent subprocess. Forks do occasionally fail when resources
4699 run out. I reviewed all the other calls to fork(); they all seem to check
4702 PH/20 When checking for unexpected SMTP input at connect time (before writing
4703 the banner), Exim was not dealing correctly with a non-positive return
4704 from the read() function. If the client had disconnected by this time,
4705 the result was a log entry for a synchronization error with an empty
4706 string after "input=" when read() returned zero. If read() returned -1
4707 (an event I could not check), uninitialized data bytes were printed.
4708 There were reports of junk text (parts of files, etc) appearing after
4711 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4713 PH/22 Added support for macro redefinition, and (re)definition in between
4714 driver and ACL definitions.
4716 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4717 forgetting to use the resulting value; it was using the unexpanded value.
4719 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4720 hadn't been configured. The fix is from Juergen Kreileder, who
4721 understands it better than I do:
4723 "Here's what I see happening with three configured cyrus_sasl
4724 authenticators configured (plain, login, cram-md5):
4726 On startup auth_cyrus_sasl_init() gets called for each of these.
4727 This means three calls to sasl_listmech() without a specified mech_list.
4728 => SASL tests which mechs of all available mechs actually work
4729 => three warnings about OTP not working
4730 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4732 With the patch, sasl_listmech() also gets called three times. But now
4733 SASL's mech_list option is set to the server_mech specified in the the
4734 authenticator. Or in other words, the answer from sasl_listmech()
4735 gets limited to just the mech you're testing for (which is different
4737 => the return list contains just 'plain' or 'login', 'cram-md5' or
4738 nothing depending on the value of ob->server_mech.
4740 I've just tested the patch: Authentication still works fine,
4741 unavailable mechs specified in the exim configuration are still
4742 caught, and the auth.log warnings about OTP are gone."
4744 PH/25 When debugging is enabled, the contents of the command line are added
4745 to the debugging output, even when log_selector=+arguments is not
4748 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4749 answer is "GNU", and only if the return is "GNU/something" is the answer
4752 PH/27 $acl_verify_message is now set immediately after the failure of a
4753 verification in an ACL, and so is available in subsequent modifiers. In
4754 particular, the message can be preserved by coding like this:
4756 warn !verify = sender
4757 set acl_m0 = $acl_verify_message
4759 Previously, $acl_verify_message was set only while expanding "message"
4760 and "log_message" when a very denied access.
4762 PH/28 Modified OS/os.c-Linux with
4764 -#ifndef OS_LOAD_AVERAGE
4765 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4767 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4768 nomenclature these days.)
4770 PH/29 Installed patch from the Sieve maintainer that adds the options
4771 sieve_useraddress and sieve_subaddress to the redirect router.
4773 PH/30 In these circumstances:
4774 . Two addresses routed to the same list of hosts;
4775 . First host does not offer TLS;
4776 . First host accepts first address;
4777 . First host gives temporary error to second address;
4778 . Second host offers TLS and a TLS session is established;
4779 . Second host accepts second address.
4780 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4781 and peerdn, if requested) that were in fact used only for the second
4784 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4785 attention to any local part prefix or suffix that was matched by the
4786 router that accepted the address. It now behaves in the same way as it
4787 does for delivery: the affixes are removed from the local part unless
4788 rcpt_include_affixes is set on the transport.
4790 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4791 timeout during the DATA phase of an incoming message.
4793 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4794 applied a patch sanctioned by the Sieve maintainer.
4796 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4797 the uid or gid is negative. A case of a negative gid caused this to be
4798 noticed. The fix allows for either to be negative.
4800 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4801 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4804 PH/36 The change PH/12 above was broken. Fixed it.
4806 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4807 the grounds that routing the same address twice would always produce the
4808 same answer. This might have been true once, but it is certainly no
4809 longer true now. Routing a child address may depend on the previous
4810 routing that produced that child. Some complicated redirection strategies
4811 went wrong when messages had multiple recipients, and made Exim's
4812 behaviour dependent on the order in which the addresses were given.
4814 I have moved the duplicate checking until after the routing is complete.
4815 Exim scans the addresses that are assigned to local and remote
4816 transports, and removes any duplicates. This means that more work will be
4817 done, as duplicates will always all be routed, but duplicates are
4818 presumably rare, so I don't expect this is of any significance.
4820 For deliveries to pipes, files, and autoreplies, the duplicate checking
4821 still happens during the routing process, since they are not going to be
4824 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4825 It corrects a timeout issue with spamd. This is Ian's comment: "The
4826 background is that sometimes spamd either never reads data from a
4827 connection it has accepted, or it never writes response data. The exiscan
4828 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4829 blindly assumes that writes won't block so it may never time out."
4831 PH/39 Allow G after quota size as well as K and M.
4833 PH/40 The value set for $authenticated_id in an authenticator may not contain
4834 binary zeroes or newlines because the value is written to log lines and
4835 to spool files. There was no check on this. Now the value is run through
4836 the string_printing() function so that such characters are converted to
4837 printable escape sequences.
4839 PH/41 $message_linecount is a new variable that contains the total number of
4840 lines in the message. Compare $body_linecount, which is the count for the
4843 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4844 bounce and delay warning messages, except in certain special cases, which
4847 (a) An SMTP error message from a remote host;
4848 (b) A message specified in a :fail: redirection;
4849 (c) A message specified in a "fail" command in a system filter;
4850 (d) A message specified in a FAIL return from the queryprogram router;
4851 (e) A message specified by the cannot_route_message router option.
4853 In these cases only, Exim does include the error details in bounce and
4854 warning messages. There are also a few cases where bland messages such
4855 as "unrouteable address" or "local delivery error" are given.
4857 PH/43 $value is now also set for the "else" part of a ${run expansion.
4859 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4860 being worked on, but at least Exim now implements the latest version to
4863 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4864 process to complete was treated as a delivery failure, a timeout while
4865 writing the message to the pipe was logged, but erroneously treated as a
4866 successful delivery. Such timeouts include transport filter timeouts. For
4867 consistency with the overall process timeout, these timeouts are now
4868 treated as errors, giving rise to delivery failures by default. However,
4869 there is now a new Boolean option for the pipe transport called
4870 timeout_defer, which, if set TRUE, converts the failures into defers for
4871 both kinds of timeout. A transport filter timeout is now identified in
4874 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4875 systems where "make" and "gmake" are different, calling "gmake" at top
4876 level broke things. I've arranged for the value of $(MAKE) to be passed
4877 from the Makefile to this script so that it can call the same version of
4881 A note about Exim versions 4.44 and 4.50
4882 ----------------------------------------
4884 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4885 changes of various kinds. As a consequence, a big documentation update was
4886 needed. This delayed the release for rather longer than seemed good, especially
4887 in the light of a couple of (minor) security issues. Therefore, the changes
4888 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4889 release. So 4.44 and 4.50 are in effect two different branches that both start
4892 I have left the 4.50 change log unchanged; it contains all the changes since
4893 4.43. The change log for 4.44 is below; many of its items are identical to
4894 those for 4.50. This seems to be the most sensible way to preserve the
4895 historical information.
4901 1. Minor wording change to the doc/README.SIEVE file.
4903 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4904 computation of the current number of files was incorrect.
4906 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4907 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4910 4. Give more explanation in the error message when the command for a transport
4911 filter fails to execute.
4913 5. There are several places where Exim runs a non-Exim command in a
4914 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4915 was being done only for the command run by the queryprogram router. It is
4916 now done for all such subprocesses. The other cases are: ${run, transport
4917 filters, and the commands run by the lmtp and pipe transports.
4919 6. Added CONFIGURE_GROUP build-time option.
4921 7. Some older OS have a limit of 256 on the maximum number of file
4922 descriptors. Exim was using setrlimit() to set 1000 as a large value
4923 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4924 systems. I've change it so that if it can't get 1000, it tries for 256.
4926 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4927 was an oversight, and furthermore, ever since the addition of extra
4928 controls (e.g. 4.43/32), the checks on when to allow different forms of
4929 "control" were broken. There should now be diagnostics for all cases when a
4930 control that does not make sense is encountered.
4932 9. Added the /retain_sender option to "control=submission".
4934 10. $recipients is now available in the predata ACL (oversight).
4936 11. Tidy the search cache before the fork to do a delivery from a message
4937 received from the command line. Otherwise the child will trigger a lookup
4938 failure and thereby defer the delivery if it tries to use (for example) a
4939 cached ldap connection that the parent has called unbind on.
4941 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4942 of $address_data from the recipient verification was clobbered by the
4943 sender verification.
4945 13. The value of address_data from a sender verification is now available in
4946 $sender_address_data in subsequent conditions in the ACL statement.
4948 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4950 15. Added a new option "connect=<time>" to callout options, to set a different
4953 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4954 was its contents. (It was OK if the option was not defined at all.)
4956 17. A "Completed" log line is now written for messages that are removed from
4957 the spool by the -Mrm option.
4959 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4960 information about exactly what failed.
4962 19. Added -dd to debug only the daemon process.
4964 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4965 handles timeouts, both on the server side and network timeouts. Renamed the
4966 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4968 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4969 It is now set to "smtps".
4971 22. $host_address is now set to the target address during the checking of
4972 ignore_target_hosts.
4974 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4975 being passed; this would have caused $sender_host_name to have been used if
4976 matching the list had actually called for a host name (not very likely,
4977 since this list is usually IP addresses). A host name is now passed as
4980 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4981 code that specifies a non-restarting handler (typically sigaction() in
4982 modern systems) in an attempt to fix a rare and obscure crash bug.
4984 25. Narrowed the window for a race in the daemon that could cause it to ignore
4985 SIGCHLD signals. This is not a major problem, because they are used only to
4986 wake it up if nothing else does.
4988 26. A malformed maildirsize file could cause Exim to calculate negative values
4989 for the mailbox size or file count. Odd effects could occur as a result.
4990 The maildirsize information is now recalculated if the size or filecount
4993 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4994 support for a long time. Removed HAVE_SYS_VFS_H.
4996 28. Installed the latest version of exipick from John Jetmore.
4998 29. In an address list, if the pattern was not a regular expression, an empty
4999 subject address (from a bounce message) matched only if the pattern was an
5000 empty string. Non-empty patterns were not even tested. This was the wrong
5001 because it is perfectly reasonable to use an empty address as part of a
5002 database query. An empty address is now tested by patterns that are
5003 lookups. However, all the other forms of pattern expect the subject to
5004 contain a local part and a domain, and therefore, for them, an empty
5005 address still always fails if the pattern is not itself empty.
5007 30. Exim went into a mad DNS loop when attempting to do a callout where the
5008 host was specified on an smtp transport, and looking it up yielded more
5009 than one IP address.
5011 31. Re-factored the code for checking spool and log partition space into a
5012 function that finds that data and another that does the check. The former
5013 is then used to implement four new variables: $spool_space, $log_space,
5014 $spool_inodes, and $log_inodes.
5016 32. The RFC2047 encoding function was originally intended for short strings
5017 such as real names; it was not keeping to the 75-character limit for
5018 encoded words that the RFC imposes. It now respects the limit, and
5019 generates multiple encoded words if necessary. To be on the safe side, I
5020 have increased the buffer size for the ${rfc2047: expansion operator from
5023 33. It is now permitted to omit both strings after an "if" condition; if the
5024 condition is true, the result is "true". As before, when the second string
5025 is omitted, a false condition yields an empty string. This makes it less
5026 cumbersome to write custom ACL and router conditions.
5028 34. Failure to deliver a bounce message always caused it to be frozen, even if
5029 there was an errors_to setting on the router. The errors_to setting is now
5032 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5033 canonical form (fully expanded) before being placed in
5034 $sender_host_address.
5036 36. The table in the code that translates DNS record types into text (T_A to
5037 "A" for instance) was missing entries for NS and CNAME. It is just possible
5038 that this could have caused confusion if both these types were looked up
5039 for the same domain, because the text type is used as part of Exim's
5040 per-process caching. But the chance of anyone hitting this buglet seems
5043 37. The dnsdb lookup has been extended in a number of ways.
5045 (1) There is a new type, "zns", which walks up the domain tree until it
5046 finds some nameserver records. It should be used with care.
5048 (2) There is a new type, "mxh", which is like "mx" except that it returns
5049 just the host names, not the priorities.
5051 (3) It is now possible to give a list of domains (or IP addresses) to be
5052 looked up. The behaviour when one of the lookups defers can be
5053 controlled by a keyword.
5055 (4) It is now possible to specify the separator character for use when
5056 multiple records are returned.
5058 38. The dnslists ACL condition has been extended: it is now possible to supply
5059 a list of IP addresses and/or domains to be looked up in a particular DNS
5062 39. Added log_selector=+queue_time_overall.
5064 40. When running the queue in the test harness, wait just a tad after forking a
5065 delivery process, to get repeatability of debugging output.
5067 41. Include certificate and key file names in error message when GnuTLS fails
5068 to set them up, because the GnuTLS error message doesn't include the name
5069 of the failing file when there is a problem reading it.
5071 42. Allow both -bf and -bF in the same test run.
5073 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
5075 44. The "Exiscan patch" is now merged into the mainline Exim source.
5077 45. Sometimes the final signoff response after QUIT could fail to get
5078 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5079 < 0 before doing a fflush(). This bug looks as though it goes back to the
5080 introduction of TLS in release 3.20, but "sometimes" must have been rare
5081 because the tests only now provoked it.
5083 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
5084 (this can affect the format of dates).
5086 47. exim_tidydb, when checking for the continued existence of a message for
5087 which it has found a message-specific retry record, was not finding
5088 messages that were in split spool directories. Consequently, it was
5089 deleting retry records that should have stayed in existence.
5091 48. Steve fixed some bugs in eximstats.
5093 49. The SPA authentication driver was not abandoning authentication and moving
5094 on to the next authenticator when an expansion was forced to fail,
5095 contradicting the general specification for all authenticators. Instead it
5096 was generating a temporary error. It now behaves as specified.
5098 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
5099 (the order specifies the preference for clients). The order is now AES256,
5100 AES128, 3DES, ARCFOUR128.
5102 51. Small patch to Sieve code - explicitly set From: when generating an
5105 52. Exim crashed if a remote delivery caused a very long error message to be
5106 recorded - for instance if somebody sent an entire SpamAssassin report back
5107 as a large number of 550 error lines. This bug was coincidentally fixed by
5108 increasing the size of one of Exim's internal buffers (big_buffer) that
5109 happened as part of the Exiscan merge. However, to be on the safe side, I
5110 have made the code more robust (and fixed the comments that describe what
5113 53. Now that there can be additional text after "Completed" in log lines (if
5114 the queue_time_overall log selector is set), a one-byte patch to exigrep
5115 was needed to allow it to recognize "Completed" as not the last thing in
5118 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
5119 patch that reportedly fixes this has been added. I am not expert enough to
5120 create a test for it. This is what the patch creator wrote:
5122 "I found a little strange behaviour of ldap code when working with
5123 Windows 2003 AD Domain, where users was placed in more than one
5124 Organization Units. When I tried to give exim partial DN, the exit code
5125 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
5126 But simultaneously result of request was absolutely normal ldap result,
5127 so I produce this patch..."
5129 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
5130 REFERENCE, so I have modified the code to exclude the patch when that macro
5133 55. Some experimental protocols are using DNS PTR records for new purposes. The
5134 keys for these records are domain names, not reversed IP addresses. The
5135 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5136 leaves it alone. Component reversal etc. now happens only for IP addresses.
5139 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
5141 57. Double the size of the debug message buffer (to 2048) so that more of very
5142 long debug lines gets shown.
5144 58. The exicyclog utility now does better if the number of log files to keep
5145 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
5147 59. Two changes related to the smtp_active_hostname option:
5149 (1) $smtp_active_hostname is now available as a variable.
5150 (2) The default for smtp_banner uses $smtp_active_hostname instead
5151 of $primary_hostname.
5153 60. The host_aton() function is supposed to be passed a string that is known
5154 to be a valid IP address. However, in the case of IPv6 addresses, it was
5155 not checking this. This is a hostage to fortune. Exim now panics and dies
5156 if the condition is not met. A case was found where this could be provoked
5157 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5158 components; fortuitously, this particular loophole had already been fixed
5159 by change 4.50/55 above.
5161 If there are any other similar loopholes, the new check in host_aton()
5162 itself should stop them being exploited. The report I received stated that
5163 data on the command line could provoke the exploit when Exim was running as
5164 exim, but did not say which command line option was involved. All I could
5165 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5166 running as the user.
5169 61. There was a buffer overflow vulnerability in the SPA authentication code
5170 (which came originally from the Samba project). I have added a test to the
5171 spa_base64_to_bits() function which I hope fixes it.
5174 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
5175 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
5177 63. The daemon start-up calls getloadavg() while still root for those OS that
5178 need the first call to be done as root, but it missed one case: when
5179 deliver_queue_load_max is set with deliver_drop_privilege. This is
5180 necessary for the benefit of the queue runner, because there is no re-exec
5181 when deliver_drop_privilege is set.
5183 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
5184 This has been fixed.
5186 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5187 was in use, was not putting the data itself into the right store pool;
5188 consequently, it could be overwritten for a subsequent message in the same
5189 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5192 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
5194 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
5195 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
5196 the function were treating the return as a boolean value, which happened to
5197 work because 0=false and not-0=true, but is not correct code.
5199 68. The host_aton() function was not handling scoped IPv6 addresses (those
5200 with, for example, "%eth0" on the end) correctly.
5202 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
5203 negated items (that is, ~something) in unsigned ints. Some compilers
5204 apparently mutter when there is no cast.
5206 70. If an address verification called from an ACL failed, and did not produce a
5207 user-specific message (i.e. there was only a "system" message), nothing was
5208 put in $acl_verify_message. In this situation, it now puts the system
5211 71. Change 4.23/11 added synchronization checking at the start of an SMTP
5212 session; change 4.31/43 added the unwanted input to the log line - except
5213 that it did not do this in the start of session case. It now does.
5215 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
5216 This is wrong and can cause the other end to generate a synchronization
5217 error if it is another Exim or anything else that does the synchronization
5218 check. A QUIT command is no longer sent after a timeout.
5220 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
5221 during host lookups.
5223 74. The defer_ok option of callout verification was not working if it was used
5224 when verifying addresses in header lines, that is, for this case:
5226 verify = header_sender/callout=defer_ok
5228 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
5229 those file descriptors could be used for SMTP connections. If anything
5230 wrote to stderr (the example that came up was "warn" in embedded Perl), it
5231 could be sent to the SMTP client, causing chaos. The daemon now opens
5232 stdin, stdout, and stderr to /dev/null when it puts itself into the
5235 76. Arrange for output from Perl's "warn" command to be written to Exim's main
5236 log by default. The user can override this with suitable Perl magic.
5238 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
5239 the log message when discard triggers, was not working for the DATA ACL or
5240 for the non-SMTP ACL.
5242 78. Error message wording change in sieve.c.
5244 79. If smtp_accept_max_per_host was set, the number of connections could be
5245 restricted to fewer than expected, because the daemon was trying to set up
5246 a new connection before checking whether the processes handling previous
5247 connections had finished. The check for completed processes is now done
5248 earlier. On busy systems, this bug wouldn't be noticed because something
5249 else would have woken the daemon, and it would have reaped the completed
5252 80. If a message was submitted locally by a user whose login name contained one
5253 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
5254 It caused a spool format error. I have fixed the spool reading code. A
5255 related problem was that the "from" clause in the Received: line became
5256 illegal because of the space(s). It is now covered by ${quote_local_part.
5258 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
5261 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
5263 83. Patch from Sieve maintainer to fix unterminated string problem in
5264 "vacation" handling.
5266 84. Some minor changes to the Linux configuration files to help with other
5267 OS variants using glibc.
5269 85. One more patch for Sieve to update vacation handling to latest spec.
5272 ----------------------------------------------------
5273 See the note above about the 4.44 and 4.50 releases.
5274 ----------------------------------------------------
5280 1. Change 4.43/35 introduced a bug that caused file counts to be
5281 incorrectly computed when quota_filecount was set in an appendfile
5284 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5285 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5288 3. Give more explanation in the error message when the command for a transport
5289 filter fails to execute.
5291 4. There are several places where Exim runs a non-Exim command in a
5292 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5293 was being done only for the command run by the queryprogram router. It is
5294 now done for all such subprocesses. The other cases are: ${run, transport
5295 filters, and the commands run by the lmtp and pipe transports.
5297 5. Some older OS have a limit of 256 on the maximum number of file
5298 descriptors. Exim was using setrlimit() to set 1000 as a large value
5299 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5300 systems. I've change it so that if it can't get 1000, it tries for 256.
5302 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5303 was an oversight, and furthermore, ever since the addition of extra
5304 controls (e.g. 4.43/32), the checks on when to allow different forms of
5305 "control" were broken. There should now be diagnostics for all cases when a
5306 control that does not make sense is encountered.
5308 7. $recipients is now available in the predata ACL (oversight).
5310 8. Tidy the search cache before the fork to do a delivery from a message
5311 received from the command line. Otherwise the child will trigger a lookup
5312 failure and thereby defer the delivery if it tries to use (for example) a
5313 cached ldap connection that the parent has called unbind on.
5315 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5316 of $address_data from the recipient verification was clobbered by the
5317 sender verification.
5319 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5320 was its contents. (It was OK if the option was not defined at all.)
5322 11. A "Completed" log line is now written for messages that are removed from
5323 the spool by the -Mrm option.
5325 12. $host_address is now set to the target address during the checking of
5326 ignore_target_hosts.
5328 13. When checking ignore_target_hosts for an ipliteral router, no host name was
5329 being passed; this would have caused $sender_host_name to have been used if
5330 matching the list had actually called for a host name (not very likely,
5331 since this list is usually IP addresses). A host name is now passed as
5334 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5335 code that specifies a non-restarting handler (typically sigaction() in
5336 modern systems) in an attempt to fix a rare and obscure crash bug.
5338 15. Narrowed the window for a race in the daemon that could cause it to ignore
5339 SIGCHLD signals. This is not a major problem, because they are used only to
5340 wake it up if nothing else does.
5342 16. A malformed maildirsize file could cause Exim to calculate negative values
5343 for the mailbox size or file count. Odd effects could occur as a result.
5344 The maildirsize information is now recalculated if the size or filecount
5347 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5348 support for a long time. Removed HAVE_SYS_VFS_H.
5350 18. Updated exipick to current release from John Jetmore.
5352 19. Allow an empty sender to be matched against a lookup in an address list.
5353 Previously the only cases considered were a regular expression, or an
5356 20. Exim went into a mad DNS lookup loop when doing a callout where the
5357 host was specified on the transport, if the DNS lookup yielded more than
5360 21. The RFC2047 encoding function was originally intended for short strings
5361 such as real names; it was not keeping to the 75-character limit for
5362 encoded words that the RFC imposes. It now respects the limit, and
5363 generates multiple encoded words if necessary. To be on the safe side, I
5364 have increased the buffer size for the ${rfc2047: expansion operator from
5367 22. Failure to deliver a bounce message always caused it to be frozen, even if
5368 there was an errors_to setting on the router. The errors_to setting is now
5371 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5372 canonical form (fully expanded) before being placed in
5373 $sender_host_address.
5375 24. Updated eximstats to version 1.33
5377 25. Include certificate and key file names in error message when GnuTLS fails
5378 to set them up, because the GnuTLS error message doesn't include the name
5379 of the failing file when there is a problem reading it.
5381 26. Expand error message when OpenSSL has problems setting up cert/key files.
5384 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5385 (this can affect the format of dates).
5387 28. exim_tidydb, when checking for the continued existence of a message for
5388 which it has found a message-specific retry record, was not finding
5389 messages that were in split spool directories. Consequently, it was
5390 deleting retry records that should have stayed in existence.
5392 29. eximstats updated to version 1.35
5393 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5394 1.35 - bugfix such that pie charts by volume are generated correctly
5396 30. The SPA authentication driver was not abandoning authentication and moving
5397 on to the next authenticator when an expansion was forced to fail,
5398 contradicting the general specification for all authenticators. Instead it
5399 was generating a temporary error. It now behaves as specified.
5401 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5402 (the order specifies the preference for clients). The order is now AES256,
5403 AES128, 3DES, ARCFOUR128.
5405 31. Small patch to Sieve code - explicitly set From: when generating an
5408 32. Exim crashed if a remote delivery caused a very long error message to be
5409 recorded - for instance if somebody sent an entire SpamAssassin report back
5410 as a large number of 550 error lines. This bug was coincidentally fixed by
5411 increasing the size of one of Exim's internal buffers (big_buffer) that
5412 happened as part of the Exiscan merge. However, to be on the safe side, I
5413 have made the code more robust (and fixed the comments that describe what
5416 33. Some experimental protocols are using DNS PTR records for new purposes. The
5417 keys for these records are domain names, not reversed IP addresses. The
5418 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5419 leaves it alone. Component reversal etc. now happens only for IP addresses.
5422 34. The host_aton() function is supposed to be passed a string that is known
5423 to be a valid IP address. However, in the case of IPv6 addresses, it was
5424 not checking this. This is a hostage to fortune. Exim now panics and dies
5425 if the condition is not met. A case was found where this could be provoked
5426 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5427 components; fortuitously, this particular loophole had already been fixed
5428 by change 4.50/55 or 4.44/33 above.
5430 If there are any other similar loopholes, the new check in host_aton()
5431 itself should stop them being exploited. The report I received stated that
5432 data on the command line could provoke the exploit when Exim was running as
5433 exim, but did not say which command line option was involved. All I could
5434 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5435 running as the user.
5438 35. There was a buffer overflow vulnerability in the SPA authentication code
5439 (which came originally from the Samba project). I have added a test to the
5440 spa_base64_to_bits() function which I hope fixes it.
5443 36. The daemon start-up calls getloadavg() while still root for those OS that
5444 need the first call to be done as root, but it missed one case: when
5445 deliver_queue_load_max is set with deliver_drop_privilege. This is
5446 necessary for the benefit of the queue runner, because there is no re-exec
5447 when deliver_drop_privilege is set.
5449 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5450 was in use, was not putting the data itself into the right store pool;
5451 consequently, it could be overwritten for a subsequent message in the same
5452 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5455 38. Sometimes the final signoff response after QUIT could fail to get
5456 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5457 < 0 before doing a fflush(). This bug looks as though it goes back to the
5458 introduction of TLS in release 3.20, but "sometimes" must have been rare
5459 because the tests only now provoked it.
5465 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5466 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5467 Now it returns an integer. A number of calls were still expecting a T/F
5468 return. Fortuitously, in all cases, the tests worked in OK situations,
5469 which is the norm. However, things would have gone wrong on any write
5470 failures on the smtp file descriptor. This function is used when sending
5471 messages over SMTP and also when doing verify callouts.
5473 2. When Exim is called to do synchronous delivery of a locally submitted
5474 message (the -odf or -odi options), it no longer closes stderr before doing
5477 3. Implemented the mua_wrapper option.
5479 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5481 5. Implemented the functions header_remove(), header_testname(),
5482 header_add_at_position(), and receive_remove_recipient(), and exported them
5485 6. If an ACL "warn" statement specified the addition of headers, Exim already
5486 inserted X-ACL-Warn: at the start if there was no header name. However, it
5487 was not making this test for the second and subsequent header lines if
5488 there were newlines in the string. This meant that an invalid header could
5489 be inserted if Exim was badly configured.
5491 7. Allow an ACL "warn" statement to add header lines at the start or after all
5492 the Received: headers, as well as at the end.
5494 8. Added the rcpt_4xx retry error code.
5496 9. Added postmaster_mailfrom=xxx to callout verification option.
5498 10. Added mailfrom=xxxx to the callout verification option, for verify=
5501 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5502 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5504 12. Inserted some casts to stop certain compilers warning when using pointer
5505 differences as field lengths or precisions in printf-type calls (mostly
5506 affecting debugging statements).
5508 13. Added optional readline() support for -be (dynamically loaded).
5510 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5511 same clock tick as a message's arrival, so that its received time was the
5512 same as the "first fail" time on the retry record, and that message
5513 remained on the queue past the ultimate address timeout, every queue runner
5514 would try a delivery (because it was past the ultimate address timeout) but
5515 after another failure, the ultimate address timeout, which should have then
5516 bounced the address, did not kick in. This was a "< instead of <=" error;
5517 in most cases the first failure would have been in the next clock tick
5518 after the received time, and all would be well.
5520 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5521 being recognized when the domain list was tested by the match_domain
5522 condition in an expansion string.
5524 16. Added the ${str2b64: operator.
5526 17. Exim was always calling setrlimit() to set a large limit for the number of
5527 processes, without checking whether the existing limit was already
5528 adequate. (It did check for the limit on file descriptors.) Furthermore,
5529 errors from getrlimit() and setrlimit() were being ignored. Now they are
5530 logged to the main and panic logs, but Exim does carry on, to try to do its
5531 job under whatever limits there are.
5533 18. Imported PCRE 5.0.
5535 19. Trivial typo in log message " temporarily refused connection" (the leading
5538 20. If the log selector return_path_on_delivery was set and an address was
5539 redirected to /dev/null, the delivery process crashed because it assumed
5540 that a return path would always be set for a "successful" delivery. In this
5541 case, the whole delivery is bypassed as an optimization, and therefore no
5544 21. Internal re-arrangement: the function for sending a challenge and reading
5545 a response while authentication was assuming a zero-terminated challenge
5546 string. It's now changed to take a pointer and a length, to allow for
5547 binary data in such strings.
5549 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5551 23. Exim was not respecting finduser_retries when seeking the login of the
5552 uid under which it was called; it was always trying 10 times. (The default
5553 setting of finduser_retries is zero.) Also, it was sleeping after the final
5554 failure, which is pointless.
5556 24. Implemented tls_on_connect_ports.
5558 25. Implemented acl_smtp_predata.
5560 26. If the domain in control=submission is set empty, Exim assumes that the
5561 authenticated id is a complete email address when it generates From: or
5562 Sender: header lines.
5564 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5565 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5566 chown and chgrp in /bin and hostname in /usr/bin.
5568 28. Exim was keeping the "process log" file open after each use, just as it
5569 does for the main log. This opens the possibility of it remaining open for
5570 long periods when the USR1 signal hits a daemon. Occasional processlog
5571 errors were reported, that could have been caused by this. Anyway, it seems
5572 much more sensible not to leave this file open at all, so that is what now
5575 29. The long-running daemon process does not normally write to the log once it
5576 has entered its main loop, and it closes the log before doing so. This is
5577 so that log files can straightforwardly be renamed and moved. However,
5578 there are a couple of unusual error situations where the daemon does write
5579 log entries, and I had neglected to close the log afterwards.
5581 30. The text of an SMTP error response that was received during a remote
5582 delivery was being truncated at 512 bytes. This is too short for some of
5583 the long messages that one sometimes sees. I've increased the limit to
5586 31. It is now possible to make retry rules that apply only when a message has a
5587 specific sender, in particular, an empty sender.
5589 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5590 it possible to be selective about when SMTP synchronization is enforced.
5592 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5594 32. Implemented hosts_connection_nolog.
5596 33. Added an ACL for QUIT.
5598 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5601 35. Added mailbox_size and mailbox_filecount to appendfile.
5603 36. Added control = no_multiline_responses to ACLs.
5605 37. There was a bug in the logic of the code that waits for the clock to tick
5606 in the case where the clock went backwards by a substantial amount such
5607 that the microsecond fraction of "now" was more than the microsecond
5608 fraction of "then" (but the whole seconds number was less).
5610 38. Added support for the libradius Radius client library this is found on
5611 FreeBSD (previously only the radiusclient library was supported).
5617 1. When certain lookups returned multiple values in the form name=value, the
5618 quoting of the values was not always being done properly. Specifically:
5619 (a) If the value started with a double quote, but contained no whitespace,
5621 (b) If the value contained whitespace other than a space character (i.e.
5622 tabs or newlines or carriage returns) it was not quoted.
5623 This fix has been applied to the mysql and pgsql lookups by writing a
5624 separate quoting function and calling it from the lookup code. The fix
5625 should probably also be applied to nisplus, ibase and oracle lookups, but
5626 since I cannot test any of those, I have not disturbed their existing code.
5628 2. A hit in the callout cache for a specific address caused a log line with no
5629 reason for rejecting RCPT. Now it says "Previous (cached) callout
5630 verification failure".
5632 3. There was an off-by-one bug in the queryprogram router. An over-long
5633 return line was truncated at 256 instead of 255 characters, thereby
5634 overflowing its buffer with the terminating zero. As well as fixing this, I
5635 have increased the buffer size to 1024 (and made a note to document this).
5637 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5638 when Exim is waiting for an SMTP response from a remote server, Exim
5639 restarts its select() call on the socket, thereby resetting its timeout.
5640 This is not a problem when such interrupts are rare. Somebody set up a cron
5641 job to run exiwhat every 2 minutes, which is less than the normal select()
5642 timeout (5 or 10 minutes). This meant that the select() timeout never
5643 kicked in because it was always reset. I have fixed this by comparing the
5644 time when an interrupt arrives with the time at the start of the first call
5645 to select(). If more time than the timeout has elapsed, the interrupt is
5646 treated as a timeout.
5648 5. Some internal re-factoring in preparation for the addition of Sieve
5649 extensions (by MH). In particular, the "personal" test is moved to a
5650 separate function, and given an option for scanning Cc: and Bcc: (which is
5651 not set for Exim filters).
5653 6. When Exim created an email address using the login of the caller as the
5654 local part (e.g. when creating a From: or Sender: header line), it was not
5655 quoting the local part when it contained special characters such as @.
5657 7. Installed new OpenBSD configuration files.
5659 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5660 try to make them clearer.
5662 9. Callout options, other than the timeout value, were being ignored when
5663 verifying sender addresses in header lines. For example, when using
5665 verify = header_sender/callout=no_cache
5667 the cache was (incorrectly) being used.
5669 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5670 only the Cygwin environment.
5672 11. When return_path_on_delivery was set as a log selector, if different remote
5673 addresses in the same message used different return paths and parallel
5674 remote delivery occurred, the wrong values would sometimes be logged.
5675 (Whenever a remote delivery process finished, the return path value from
5676 the most recently started remote delivery process was logged.)
5678 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5679 lines when AUTH and/or TLS are in use. This is the "received protocol"
5680 field. Exim used to use "asmtp" for authenticated SMTP, without any
5681 indication (in the protocol name) for TLS use. Now it follows the RFC and
5682 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5683 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5684 names appear in log lines as well as in Received: header lines.
5686 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5687 extensions, and comparison tests, and to fix some bugs.
5689 14. Changes to the "personal" filter test:
5691 (1) The test was buggy in that it was just doing the equivalent of
5692 "contains" tests on header lines. For example, if a user's address was
5693 anne@some.where, the "personal" test would incorrectly be true for
5695 To: susanne@some.where
5697 This test is now done by extracting each address from the header in turn,
5698 and checking the entire address. Other tests that are part of "personal"
5699 are now done using regular expressions (for example, to check local parts
5700 of addresses in From: header lines).
5702 (2) The list of non-personal local parts in From: addresses has been
5703 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5704 taken from the Sieve specification recommendations.
5706 (3) If the message contains any header line starting with "List-" it is
5707 treated as non-personal.
5709 (4) The test for "circular" in the Subject: header line has been removed
5710 because it now seems ill-conceived.
5712 15. Minor typos in src/EDITME comments corrected.
5714 16. Installed latest exipick from John Jetmore.
5716 17. If headers_add on a router specified a text string that was too long for
5717 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5718 of string_sprintf() is now avoided.
5720 18. $message_body_size was not set (it was always zero) when running the DATA
5721 ACL and the local_scan() function.
5723 19. For the "mail" command in an Exim filter, no default was being set for
5724 the once_repeat time, causing a random time value to be used if "once" was
5725 specified. (If the value happened to be <= 0, no repeat happened.) The
5726 default is now 0s, meaning "never repeat". The "vacation" command was OK
5727 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5728 (I found it when inspecting the code).
5730 20. There is now an overall timeout for performing a callout verification. It
5731 defaults to 4 times the callout timeout, which applies to individual SMTP
5732 commands during the callout. The overall timeout applies when there is more
5733 than one host that can be tried. The timeout is checked before trying the
5734 next host. This prevents very long delays if there are a large number of
5735 hosts and all are timing out (e.g. when the network connections are timing
5736 out). The value of the overall timeout can be changed by specifying an
5737 additional sub-option for "callout", called "maxwait". For example:
5739 verify = sender/callout=5s,maxwait=20s
5741 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5742 to the end before writing, but this should make it even safer).
5744 22. Exim was forgetting that it had advertised PIPELINING for the second and
5745 subsequent messages on an SMTP connection. It was also not resetting its
5746 memory on STARTTLS and an internal HELO.
5748 23. When Exim logs an SMTP synchronization error within a session, it now
5749 records whether PIPELINING has been advertised or not.
5751 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5752 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5753 rather than long int.
5755 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5757 26. Added the never_mail option to autoreply.
5763 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5764 crash if the getsockname() call failed; this can happen if a connection is
5765 closed very soon after it is established. The problem was simply in the
5766 order in which certain operations were done, causing Exim to try to write
5767 to the SMTP stream before it had set up the file descriptor. The bug has
5768 been fixed by making things happen in the correct order.
5774 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5775 before the connection was closed, thus losing the rejection response.
5777 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5778 some early Solaris releases, but causes trouble in current releases where
5779 socklen_t is defined.
5781 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5784 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5787 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5788 that releasing the top of it at the end releases what was used for sub-
5789 expansions (unless the block got too big). However, discard this block if
5790 the first thing is a variable or header, so that we can use its block when
5791 it is dynamic (useful for very large $message_headers, for example).
5793 6. Lookups now cache *every* query, not just the most recent. A new, separate
5794 store pool is used for this. It can be recovered when all lookup caches are
5795 flushed. Lookups now release memory at the end of their result strings.
5796 This has involved some general refactoring of the lookup sources.
5798 7. Some code has been added to the store_xxx() functions to reduce the amount
5799 of flapping under certain conditions.
5801 8. log_incoming_interface used to affect only the <= reception log lines. Now
5802 it causes the local interface and port to be added to several more SMTP log
5803 lines, for example "SMTP connection from", and rejection lines.
5805 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5807 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5809 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5810 could be overwritten at the end of the current message (or the start of a
5811 new message if it was set in a HELO ACL). The value is now preserved for
5812 the duration of the SMTP connection.
5814 12. If a transport had a headers_rewrite setting, and a matching header line
5815 contained an unqualified address, that address was qualified, even if it
5816 did not match any rewriting rules. The underlying bug was that the values
5817 of the flags that permit the existence of unqualified sender and recipient
5818 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5819 non-local messages, and by -bnq for local messages) were not being
5820 preserved with the message after it was received.
5822 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5823 "next input=" as part of the text comprising the host identity instead of
5824 the correct text. The code was using the same buffer for two different
5825 strings. However, depending on which order the printing function evaluated
5826 its arguments, the bug did not always show up. Under Linux, for example, my
5827 test suite worked just fine.
5829 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5830 doesn't work with some older versions of Perl. It has been changed to "my",
5831 which in any case is probably the better facility to use.
5833 15. A really picky compiler found some instances of statements for creating
5834 error messages that either had too many or two few arguments for the format
5837 16. The size of the buffer for calls to the DNS resolver has been increased
5838 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5839 for addresses that have a lot of PTR records. This alleviates a problem; it
5840 does not fully solve it.
5842 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5843 buffer now truncates the list and logs the incident, which is the same
5844 action as happens when Exim is looking up a host name and its aliases.
5845 Previously in this situation something unpredictable would happen;
5846 sometimes it was "internal error: store_reset failed".
5848 18. If a server dropped the connection unexpectedly when an Exim client was
5849 using GnuTLS and trying to read a response, the client delivery process
5850 crashed while trying to generate an error log message.
5852 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5853 string, for example:
5855 warn message = H1: something\nH2: something
5857 the text was added as a single header line from Exim's point of view
5858 though it ended up OK in the delivered message. However, searching for the
5859 second and subsequent header lines using $h_h2: did not work. This has been
5860 fixed. Similarly, if a system filter added multiple headers in this way,
5861 the routers could not see them.
5863 20. Expanded the error message when iplsearch is called with an invalid key to
5864 suggest using net-iplsearch in a host list.
5866 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5867 ACLs are no longer actually imposed (and a message to that effect is
5870 22. If a "gecos" field in a passwd entry contained escaped characters, in
5871 particular, if it contained a \" sequence, Exim got it wrong when building
5872 a From: or a Sender: header from that name. A second bug also caused
5873 incorrect handling when an unquoted " was present following a character
5874 that needed quoting.
5876 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5877 was not being matched caselessly.
5879 24. Arranged for all hyphens in the exim.8 source to be escaped with
5882 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5883 properly. Recipient callout cache records were still being keyed to include
5884 the sender, even when use_sender was set false. This led to far more
5885 callouts that were necessary. The sender is no longer included in the key
5886 when use_sender is false.
5888 26. Added "control = submission" modifier to ACLs.
5890 27. Added the ${base62d: operator to decode base 62 numbers.
5892 28. dnsdb lookups can now access SRV records.
5894 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5895 the configuration file.
5897 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5898 (-v) mode. This makes the output for a verbose queue run more intelligible.
5900 31. Added a use_postmaster feature to recipient callouts.
5902 32. Added the $body_zerocount variable, containing the number of binary zero
5903 bytes in the message body.
5905 33. The time of last modification of the "new" subdirectory is now used as the
5906 "mailbox time last read" when there is a quota error for a maildir
5909 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5911 35. Added +ignore_unknown as a special item in host lists.
5913 36. Code for decoding IPv6 addresses in host lists is now included, even if
5914 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5915 address was recognized as an IP address, but was then not correctly decoded
5916 into binary, causing unexpected and incorrect effects when compared with
5923 1. Very minor rewording of debugging text in manualroute to say "list of
5924 hosts" instead of "hostlist".
5926 2. If verify=header_syntax was set, and a header line with an unqualified
5927 address (no domain) and a large number of spaces between the end of the
5928 name and the colon was received, the reception process suffered a buffer
5929 overflow, and (when I tested it) crashed. This was caused by some obsolete
5930 code that should have been removed. The fix is to remove it!
5932 3. When running in the test harness, delay a bit after writing a bounce
5933 message to get a bit more predictability in the log output.
5935 4. Added a call to search_tidyup() just before forking a reception process. In
5936 theory, someone could use a lookup in the expansion of smtp_accept_max_
5937 per_host which, without the tidyup, could leave open a database connection.
5939 5. Added the variables $recipient_data and $sender_data which get set from a
5940 lookup success in an ACL "recipients" or "senders" condition, or a router
5941 "senders" option, similar to $domain_data and $local_part_data.
5943 6. Moved the writing of debug_print from before to after the "senders" test
5946 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5947 problems for message scanning, either using a data ACL, or using
5948 local_scan() because the Received: header was not generated till after they
5949 were called (in order to set the time as the time of reception completion).
5950 I have revised the way this works. The header is now generated after the
5951 body is received, but before the ACL or local_scan() are called. After they
5952 are run, the timestamp in the header is updated.
5958 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5959 before starting a queue runner without re-exec. This happened only when
5960 deliver_drop_privilege was set or when the Exim user was set to root. The
5961 effect of the bug was that timeouts during subsequent deliveries caused
5962 crashes instead of being properly handled. The handler is now left at its
5963 default (and expected) setting.
5965 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5966 message, again when deliver_drop_privilege is set or Exim is run as root.
5967 The bug described in (1) was not present in this case, but the tidying up
5968 of the other signals was missing. I have made the two cases consistent.
5970 3. The ignore_target_hosts setting on a manualroute router was being ignored
5971 for hosts that were looked up using the /MX notation.
5973 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5976 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5977 operated on the sender address. After changing the $sender_address to <>
5978 for the sender address verify, Exim was re-instated it as the original
5979 (before rewriting) address, but remembering that it had rewritten it, so it
5980 wasn't rewriting it again. This bug also had the effect of breaking the
5981 sender address verification caching when the sender address was rewritten.
5983 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5984 This has been changed so that if the ip literal address matches
5985 ignore_target_hosts, the router declines.
5987 7. Added expansion conditions match_domain, match_address, and match_local_
5988 part (NOT match_host).
5990 8. The placeholder for the Received: header didn't have a length field set.
5992 9. Added code to Exim itself and to exim_lock to test for a specific race
5993 condition that could lead to file corruption when using MBX delivery. The
5994 issue is with the lockfile that is created in /tmp. If this file is removed
5995 after a process has opened it but before that process has acquired a lock,
5996 there is the potential for a second process to recreate the file and also
5997 acquire a lock. This could lead to two Exim processes writing to the file
5998 at the same time. The added code performs the same test as UW imapd; it
5999 checks after acquiring the lock that its file descriptor still refers to
6000 the same named file.
6002 10. The buffer for building added header lines was of fixed size, 8192 bytes.
6003 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
6006 11. Added the smtp_active_hostname option. If used, this will typically be made
6007 to depend on the incoming interface address. Because $interface_address is
6008 not set up until the daemon has forked a reception process, error responses
6009 that can happen earlier (such as "too many connections") no longer contain
6012 12. If an expansion in a condition on a "warn" statement fails because a lookup
6013 defers, the "warn" statement is abandoned, and the next ACL statement is
6014 processed. Previously this caused the whole ACL to be aborted.
6016 13. Added the iplsearch lookup type.
6018 14. Added ident_timeout as a log selector.
6020 15. Added tls_certificate_verified as a log selector.
6022 16. Added a global option tls_require_ciphers (compare the smtp transport
6023 option of the same name). This controls incoming TLS connections.
6025 17. I finally figured out how to make tls_require_ciphers do a similar thing
6026 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
6027 before starting the TLS session.
6029 18. Tabs are now shown as \t in -bP output.
6031 19. If the log selector return_path_on_delivery was set, Exim crashed when
6032 bouncing a message because it had too many Received: header lines.
6034 20. If two routers both had headers_remove settings, and the first one included
6035 a superfluous trailing colon, the final name in the first list and the
6036 first name in the second list were incorrectly joined into one item (with a
6037 colon in the middle).
6043 1. Added -C and -D options to the exinext utility, mainly to make it easier
6044 to include in the automated testing, but these could be helpful when
6045 multiple configurations are in use.
6047 2. The exinext utility was not formatting the output nicely when there was
6048 an alternate port involved in the retry record key, nor when there was a
6049 message id as well (for retries that were specific to a specific message
6050 and a specific host). It was also confused by IPv6 addresses, because of
6051 the additional colons they contain. I have fixed the IPv4 problem, and
6052 patched it up to do a reasonable job for IPv6.
6054 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
6055 delivery, the log line now contains "pipelined" if PIPELINING was used.
6057 4. An SMTP transport process used to panic and die if the bind() call to set
6058 an explicit outgoing interface failed. This has been changed; it is now
6059 treated in the same way as a connect() failure.
6061 5. A reference to $sender_host_name in the part of a conditional expansion
6062 that was being skipped was still causing a DNS lookup. This no longer
6065 6. The def: expansion condition was not recognizing references to header lines
6066 that used bh_ and bheader_.
6068 7. Added the _cache feature to named lists.
6070 8. The code for checking quota_filecount in the appendfile transport was
6071 allowing one more file than it should have been.
6073 9. For compatibility with Sendmail, the command line option
6081 and sets the incoming protocol and host name (for trusted callers). The
6082 host name and its colon can be omitted when only the protocol is to be set.
6083 Note the Exim already has two private options, -pd and -ps, that refer to
6084 embedded Perl. It is therefore impossible to set a protocol value of "d" or
6085 "s", but I don't think that's a major issue.
6087 10. A number of refactoring changes to the code, none of which should affect
6090 (a) The number of logging options was getting close to filling up the
6091 32-bit word that was used as a bit map. I have split them into two classes:
6092 those that are passed in the argument to log_write(), and those that are
6093 only ever tested independently outside of that function. These are now in
6094 separate 32-bit words, so there is plenty of room for expansion again.
6095 There is no change in the user interface or the logging behaviour.
6097 (b) When building, for example, log lines, the code previously used a
6098 macro that called string_cat() twice, in order to add two strings. This is
6099 not really sufficiently general. Furthermore, there was one instance where
6100 it was actually wrong because one of the argument was used twice, and in
6101 one call a function was used. (As it happened, calling the function twice
6102 did not affect the overall behaviour.) The macro has been replaced by a
6103 function that can join an arbitrary number of extra strings onto a growing
6106 (c) The code for expansion conditions now uses a table and a binary chop
6107 instead of a serial search (which was left over from when there were very
6108 few conditions). Also, it now recognizes conditions like "pam" even when
6109 the relevant support is not compiled in: a suitably worded error message is
6110 given if an attempt is made to use such a condition.
6112 11. Added ${time_interval:xxxxx}.
6114 12. A bug was causing one of the ddress fields not to be passed back correctly
6115 from remote delivery subprocesses. The field in question was not being
6116 subsequently used, so this caused to problems in practice.
6118 13. Added new log selectors queue_time and deliver_time.
6120 14. Might have fixed a bug in maildirsizefile handling that threw up
6121 "unexpected character" debug warnings, and recalculated the data
6122 unnecessarily. In any case, I expanded the warning message to give more
6125 15. Added the message "Restricted characters in address" to the statements in
6126 the default ACL that block characters like @ and % in local parts.
6128 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
6129 Three changes have been made:
6131 (a) There was a serious bug; a negative response to MAIL caused the whole
6132 recipient domain to be cached as invalid, thereby blocking all messages
6133 to all local parts at the same domain, from all senders. This bug has
6134 been fixed. The domain is no longer cached after a negative response to
6135 MAIL if the sender used is not empty.
6137 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
6140 (c) A new callout option, "use_sender" has been added for people who want
6141 the modified behaviour.
6147 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
6150 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
6151 indeed breaks things for older releases.
6153 3. Added additional logging to the case where there is a problem reading data
6154 from a filter that is running in a subprocess using a pipe, in order to
6155 try to track down a specific problem.
6157 4. Testing facility fudge: when running in the test harness and attempting
6158 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
6159 getting "No route to host". Convert this to a timeout.
6161 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
6164 6. Some OS don't have socklen_t but use size_t instead. This affects the
6165 fifth argument of getsockopt() amongst other things. This is now
6166 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
6167 can be set for individual OS. I have set it for SunOS5, OSF1, and
6168 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
6169 some earlier ones do not.
6171 7. Change 4.30/15 was not doing the test caselessly.
6173 8. The standard form for an IPv6 address literal was being rejected by address
6174 parsing in, for example, MAIL and RCPT commands. An example of this kind of
6175 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6176 this, as well as the form without the "IPv6" on the front (but only when
6177 address literals are enabled, of course).
6179 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
6181 10. Exim crashed if a message with an empty sender address specified by -f
6182 encountered a router with an errors_to setting. This could be provoked only
6183 by a command such as
6187 where an empty string was supplied; "<>" did not hit this bug.
6189 11. Installed PCRE release 4.5.
6191 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
6192 remained set. It is now erased.
6194 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
6195 times from message ids (which are base 36 rather than the normal 62).
6197 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
6198 were being counted as actual protocol errors, and logged if the log
6199 selector +smtp_protocol_error was set. One cannot be perfect in this test,
6200 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
6201 and DATA following a set of rejected RCPTs do not count as protocol errors.
6202 In other words, Exim assumes they were pipelined, though this may not
6203 actually be the case. Of course, in all cases the client gets an
6204 appropriate error code.
6206 15. If a lookup fails in an ACL condition, a message about the failure may
6207 be available; it is used if testing the ACL cannot continue, because most
6208 such messages specify what the cause of the deferral is. However, some
6209 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
6210 that caused an old message to be retained and used if a later statement
6211 caused a defer, replacing the real cause of the deferral.
6213 16. If an IP address had so many PTR records that the DNS lookup buffer
6214 was not large enough to hold them, Exim could crash while trying to process
6215 the truncated data. It now detects and logs this case.
6217 17. Further to 4.21/58, another change has been made: if (and only if) the
6218 first line of a message (the first header line) ends with CRLF, a bare LF
6219 in a subsequent header line has a space inserted after it, so as not to
6220 terminate the header.
6222 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
6223 unnecessarily, used atoi() instead of strtol(), and didn't check the
6224 termination when getting file sizes from file names by regex.
6226 19. Completely re-implemented the support for maildirsize files, in the light
6227 of a number of problems with the previous contributed implementation
6228 (4.30/29). In particular:
6230 . If the quota is zero, the maildirsize file is maintained, but no quota is
6233 . If the maildir directory does not exist, it is created before any attempt
6234 to write a maildirsize file.
6236 . The quota value in the file is just a cache; if the quota is changed in
6237 the transport, the new value overrides.
6239 . A regular expression is available for excluding directories from the
6242 20. The autoreply transport checks the characters in options that define the
6243 message's headers; it allows continued headers, but it was checking with
6244 isspace() after an embedded newline instead of explicitly looking for a
6247 21. If all the "regular" hosts to which an address was routed had passed their
6248 expiry times, and had not reached their retry times, the address was
6249 bounced, even if fallback hosts were defined. Now Exim should go on to try
6252 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
6253 equivalent code in the SMTP transport. Some hosts send humungous responses
6254 to HELO/EHLO, more than 1024 it seems.
6256 23. Refactoring: code in filter.c used (void *) for "any old type" but this
6257 gives compiler warnings in some environments. I've now done it "properly",
6260 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
6261 (because of problems with the built-in one) was declared to return uschar *
6262 instead of char *, causing compiler failure.
6264 25. Fixed a file descriptor leak when processing alias/forward files.
6266 26. Fixed a minor format string issue in dbfn.c.
6268 27. Typo in exim.c: ("dmbnz" for "dbmnz").
6270 28. If a filter file refered to $h_xxx or $message_headers, and the headers
6271 contained RFC 2047 "words", Exim's memory could, under certain conditions,
6274 29. When a sender address is verified, it is cached, to save repeating the test
6275 when there is more than one recipient in a message. However, when the
6276 verification involves a callout, it is possible for different callout
6277 options to be set for different recipients. It is too complicated to keep
6278 track of this in the cache, so now Exim always runs a verification when a
6279 callout is required, relying on the callout cache for the optimization.
6280 The overhead is duplication of the address routing, but this should not be
6283 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
6284 to be verified with callout=postmaster, and the main callout worked but the
6285 postmaster check failed, the verification correctly failed. However, if a
6286 subsequent RCPT command asked for sender verification *without* the
6287 postmaster check, incorrect caching caused this verification also to fail,
6290 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
6291 it was not caching the DNS options (qualify_single, search_parents) that
6292 were used when the lookup failed. A subsequent lookup with different
6293 options therefore always gave the same answer, though there were cases
6294 where it should not have. (Example: a "domains = !$mx_any" option on a
6295 dnslookup router: the "domains" option is always processed without any
6296 widening, but the router might have qualify_single set.) Now Exim uses the
6297 cached value only when the same options are set.
6299 32. Added John Jetmore's "exipick" utility to the distribution.
6301 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
6302 than a timeout (e.g. a certificate is required, and is not provided), an
6303 Exim server now closes the connection immediately. Previously it waited for
6304 the client to close - but if the client is SSL, it seems that they each
6305 wait for each other, leading to a delay before one of them times out.
6307 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
6308 maintained 0.8.x compatibility because I don't think many are using it, and
6309 it is clearly obsolete.
6311 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
6314 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
6315 client certificate was expired. A simple patch fixes this, though I don't
6316 understand the full logic of why the verify callback is called multiple
6319 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
6320 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
6321 which causes problems with some clients (such as the Certicom SSL Plus
6322 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
6323 disables the coutermeasure allowing Eudora to connect."
6325 38. Exim was not checking that a write() to a log file succeeded. This could
6326 lead to Bad Things if a log got too big, in particular if it hit a file
6327 size limit. Exim now panics and dies if it cannot write to a log file, just
6328 as it does if it cannot open a log file.
6330 39. Modified OS/Makefile-Linux so that it now contains
6332 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
6334 The two -D definitions ensure that Exim is compiled with large file
6335 support, which makes it possible to handle log files that are bigger than
6338 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
6339 instance) a domain was checked against a named list that involved a lookup,
6340 causing $domain_data to be set, then another domain was checked against the
6341 same list, then the first domain was re-checked, the value of $domain_data
6342 after the final check could be wrong. In particular, if the second check
6343 failed, it could be set empty. This bug probably also applied to
6346 41. The strip_trailing_dot option was not being applied to the address given
6347 with the -f command-line option.
6349 42. The code for reading a message's header from the spool was incrementing
6350 $received_count, but never initializing it. This meant that the value was
6351 incorrect (doubled) while delivering a message in the same process in which
6352 it was received. In the most common configuration of Exim, this never
6353 happens - a fresh exec is done - but it can happen when
6354 deliver_drop_privilege is set.
6356 43. When Exim logs an SMTP synchronization error - client data sent too soon -
6357 it now includes up to 150 characters of the unexpected data in the log
6360 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
6361 and building data strings. The size of both of these buffers was 10 000
6362 bytes - far larger than anybody would *ever* want, thought I. Needless to
6363 say, somebody hit the limit. I have increased the maximum line length to
6364 20 000 and the maximum data length of concatenated lines to 100 000. I have
6365 also fixed two bugs, because there was no checking on these buffers. Tsk,
6366 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
6367 buffer is too small.
6369 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6370 lsearch lookups. Now it does.
6372 46. When parsing a route_list item in a manualroute router, a fixed-length
6373 buffer was used for the list of hosts. I made this 1024 bytes long,
6374 thinking that nobody would ever have a list of hosts that long. Wrong.
6375 Somebody had a whole pile of complicated expansion conditions, and the
6376 string was silently truncated, leading to an expansion error. It turns out
6377 that it is easier to change to an unlimited length (owing to other changes
6378 that have happened since this code was originally written) than to build
6379 structure for giving a limitation error. The length of the item that
6380 expands into the list of hosts is now unlimited.
6382 47. The lsearch lookup could not handle data where the length of text line was
6383 more than 4095 characters. Such lines were truncated, leading to shortened
6384 data being returned. It should now handle lines of any length.
6386 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6387 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6390 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6391 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6392 no longer happens in comment lines. A list of replacements is now placed
6393 at the head of all of the source files, except those whose only change is
6394 to replace PERL_COMMAND in the very first #! line.
6396 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6397 messages on the queue, with a bottom-up merge sort, using code contributed
6398 by Michael Haardt. This should make operations like -bp somewhat faster on
6399 large queues. It won't affect queue runners, except when queue_run_in_order
6402 51. Installed eximstats 1.31 in the distribution.
6404 52. Added support for SRV lookups to the dnslookup router.
6406 53. If an ACL referred to $message_body or $message_body_end, the value was not
6407 reset for any messages that followed in the same SMTP session.
6409 54. The store-handling optimization for building very long strings was not
6410 differentiating between the different store pools. I don't think this
6411 actually made any difference in practice, but I've tidied it.
6413 55. While running the routers to verify a sender address, $sender_address
6414 was still set to the sender address. This is wrong, because when routing to
6415 send a bounce to the sender, it would be empty. Therefore, I have changed
6416 it so that, while verifying a sender address, $sender_address is set to <>.
6417 (There is no change to what happens when verifying a recipient address.)
6419 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6420 target A or AAAA records (if not already returned) without resetting the
6421 qualify_single or search_parents options of the DNS resolver. These are
6422 inappropriate in this case because the targets of MX and SRV records must
6423 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6424 target that, when qualified, matched something in the local domain. These
6425 two options are now turned off when doing these lookups.
6427 57. It seems that at least some releases of Reiserfs (which does not have the
6428 concept of a fixed number of inodes) returns zero and not -1 for the
6429 number of available inodes. This interacted badly with check_spool_inodes,
6430 which assumed that -1 was the "no such thing" setting. What I have done is
6431 to check that the total number of inodes is greater than zero before doing
6432 the test of how many are available.
6434 58. When a "warn" ACL statement has a log_message modifier, the message is
6435 remembered, and not repeated. This is to avoid a lot of repetition when a
6436 message has many recipients that cause the same warning to be written.
6437 However, Exim was preserving the list of already written lines for an
6438 entire SMTP session, which doesn't seem right. The memory is now reset if a
6439 new message is started.
6441 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6442 debugging output unless log_rewrite was also set.
6444 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6445 of (int)(handle) when we know that handle contains (void *)(-1).
6447 61. The Exim daemon panic-logs an error return when it closes the incoming
6448 connection. However "connection reset by peer" seems to be common, and
6449 isn't really an error worthy of noting specially, so that particular error
6452 62. When Exim is trying to find all the local interfaces, it used to panic and
6453 die if the ioctl to get the interface flags failed. However, it seems that
6454 on at least one OS (Solaris 9) it is possible to have an interface that is
6455 included in the list of interfaces, but for which you get a failure error
6456 for this call. This happens when the interface is not "plumbed" into a
6457 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6458 failure of the "get flags" call assumes that the interface is down.
6460 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6461 makes life easier for people who are doing arithmetic on fields extracted
6462 from dates, where you often get leading zeros that should not be
6463 interpreted as octal.
6465 64. Added qualify_domain to the redirect router, to override the global
6468 65. If a pathologically long header line contained very many addresses (the
6469 report of this problem mentioned 10 000) and each of them was rewritten,
6470 Exim could use up a very large amount of memory. (It kept on making new
6471 copies of the header line as it rewrote, and never released the old ones.)
6472 At the expense of a bit more processing, the header rewriting function has
6473 been changed so that it no longer eats memory in this way.
6475 66. The generation of the Received: header has been moved from the time that a
6476 message starts to be received, to the time that it finishes. The timestamp
6477 in the Received: header should now be very close to that of the <= log
6478 line. There are two side-effects of this change:
6480 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6481 logged header lines no longer include the local Received: line, because
6482 it has not yet been created. The same applies to a copy of the message
6483 that is returned to a non-SMTP sender when a message is rejected.
6485 (b) When a filter file is tested using -bf, no additional Received: header
6486 is added to the test message. After some thought, I decided that this
6489 This change does not affect the value of $received_for. It is still set
6490 after address rewriting, but before local_scan() is called.
6492 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6494 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6495 gave an unhelpful panic error message, and a defer error. I have managed to
6496 change this behaviour so that it now rejects any supplied certificate,
6497 which seems right, as the list of acceptable certificates is empty.
6499 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6500 gave an unhelpful defer error. I have not managed to make this reject any
6501 supplied certificates, but the error message it gives is "no certificate
6502 supplied", which is not helpful.
6504 70. exigrep's output now also includes lines that are not associated with any
6505 message, but which match the given pattern. Implemented by a patch from
6506 Martin Sluka, which also tidied up the Perl a bit.
6508 71. Recipient callout verification, like sender verification, was using <> in
6509 the MAIL FROM command. This isn't really the right thing, since the actual
6510 sender may affect whether the remote host accepts the recipient or not. I
6511 have changed it to use the actual sender in the callout; this means that
6512 the cache record is now keyed on a recipient/sender pair, not just the
6513 recipient address. There doesn't seem to be a real danger of callout loops,
6514 since a callout by the remote host to check the sender would use <>.
6515 [SEE ABOVE: changed after hitting problems.]
6517 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6518 temporary errors. However, in the case of such a code being given after
6519 the end of a data transmission (i.e. after ".") Exim was failing to write
6520 a retry record for the message. (Yes, there was some broken host that was
6521 actually sending 8xx at this point.)
6523 73. An unknown lookup type in a host list could cause Exim to panic-die when
6524 the list was checked. (An example that provoked this was putting <; in the
6525 middle of a list instead of at the start.) If this happened during a DATA
6526 ACL check, a -D file could be left lying around. This kind of configuration
6527 error no longer causes Exim to die; instead it causes a defer error. The
6528 incident is still logged to the main and panic logs.
6530 74. Buglet left over from Exim 3 conversion. The message "too many messages
6531 in one connection" was written to the rejectlog but not the mainlog, except
6532 when address rewriting (yes!) was being logged.
6534 75. Added write_rejectlog option.
6536 76. When a system filter was run not as root (that is, when system_filter_user
6537 was set), the values of the $n variables were not being returned to the
6538 main process; thus, they were not subsequently available in the $sn
6541 77. Added +return_path_on_delivery log selector.
6543 78. A connection timeout was being treated differently from recipients deferred
6544 when testing hosts_max_try with a message that was older than the host's
6545 retry timeout. (The host should not be counted, thus allowing all hosts to
6546 be tried at least once before bouncing.) This may have been the cause of an
6547 occasionally reported bug whereby a message would remain on the queue
6548 longer than the retry timeout, but would be bounced if a delivery was
6549 forced. I say "may" because I never totally pinned down the problem;
6550 setting up timeout/retry tests is difficult. See also the next item.
6552 79. The ultimate address timeout was not being applied to errors that involved
6553 a combination of host plus message (for example, a timeout on a MAIL
6554 command). When an address resolved to a number of possible hosts, and they
6555 were not all tried for each delivery (e.g. because of hosts_max_try), a
6556 message could remain on the queue longer than the retry timeout.
6558 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6561 81. Fixed an obscure SMTP outgoing bug which required at least the following
6562 conditions: (a) there was another message waiting for the same server;
6563 (b) the server returned 5xx to all RCPT commands in the first message so
6564 that the message was not completed; (c) the server dropped the connection
6565 or gave a negative response to the RSET that Exim sends to abort the
6566 transaction. The observed case was a dropped connection after DATA that had
6567 been sent in pipelining mode. That is, the server had advertised PIPELINING
6568 but was not implementing it correctly. The effect of the bug was incorrect
6569 behaviour, such as trying another host, and this could lead to a crash.
6575 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6576 and daemon.c were passed as pointers to ints; they should have been
6577 pointers to socklen_t variables (which are typically unsigned ints).
6579 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6582 3. Fixed a really odd bug that affected only the testing scheme; patching a
6583 certain fixed string in the binary changed the value of another string that
6584 happened to be identical to the end of the original first string.
6586 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6587 name", it returns that address as the IP address. On some operating
6588 systems (e.g. Solaris), it also passes back the IP address string as the
6589 "host name". However, on others (e.g. Linux), it passes back an empty
6590 string. Exim wasn't checking for this, and was changing the host name to an
6591 empty string, assuming it had been canonicalized.
6593 5. Although rare, it is permitted to have more than one PTR record for a given
6594 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6595 all the names associated with an address, because they do in Solaris.
6596 However, it seems that they do not in Linux for data that comes from the
6597 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6598 I found this out when I moved to a new Linux workstation and tried to run
6599 the Exim test suite.
6601 To get round this problem I have changed the code so that it now does its
6602 own call to the DNS to look up PTR records when searching for a host name.
6603 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6604 addresses that are only in /etc/hosts are still found.
6606 This behaviour is, however, controlled by an option called host_lookup_
6607 order, which defaults to "bydns:byaddr". If people want to use the other
6608 order, or indeed, just use one or the other means of lookup, they can
6609 specify it in this variable.
6611 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6612 some operating systems, this comes back from gethostbyaddr() as an empty
6613 string, and this is what Exim used to test for. However, it seems that in
6614 other systems, "." is yielded. Exim now tests for this case too.
6616 7. The values of check_spool_space and check_log_space are now held internally
6617 as a number of kilobytes instead of an absolute number of bytes. If a
6618 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6619 kilobyte. This means that much larger values can be stored.
6621 8. Exim monitor: an attempt to get the action menu when not actually pointing
6622 at a message produces an empty menu entitled "No message selected". This
6623 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6624 no entries in it ("Shell widget menu has zero width and/or height"). So I
6625 have added a single, blank menu entry in this case.
6627 9. Added ${quote_local_part.
6629 10. MIME decoding is now applied to the contents of Subject: header lines when
6632 11. Now that a reference to $sender_host_address automatically causes a reverse
6633 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6634 host lookup before query-style lookups in lists that might use this
6635 variable. This has therefore been abolished, and the "net-" prefix is no
6636 longer necessary for query-style lookups.
6638 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6639 have been a typo for LFLAGS, so it has been changed.
6641 13. The install script calls Exim with "-C /dev/null" in order to find the
6642 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6643 to be output. However, since Exim outputs its version number before the
6644 error, it didn't break the script. It just looked ugly. I fixed this by
6645 always allowing "-C /dev/null" if the caller is root.
6647 14. Ignore overlarge ACL variable number when reading spool file - insurance
6648 against a later release with more variables having written the file.
6650 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6651 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6652 this, as well as the form without the "IPv6" on the front.
6654 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6655 OS/Makefile-Darwin file.
6657 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6658 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6660 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6661 variable. However, some OS use ints for inodes. I've added cast to long int
6662 to get rid of the compiler warning.
6664 19. I had forgotten to lock out "/../" in configuration file names when
6665 ALT_CONFIG_PREFIX was set.
6667 20. Routers used for verification do not need to specify transports. However,
6668 if such a router generated a host list, and callout was configured, Exim
6669 crashed, because it could not find a port number from the (non-existent)
6670 transport. It now assumes port 25 in this circumstance.
6672 21. Added the -t option to exigrep.
6674 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6675 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6676 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6677 them is set, LOOKUP_LSEARCH is forced.
6679 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6681 24. Added sender and host information to the "rejected by local_scan()" log
6682 line; previously there was no indication of these.
6684 25. Added .include_if_exists.
6686 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6687 Linux. It turns out that not all file systems support this. Apparently some
6688 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6689 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6690 not supported on the file descriptor, is now ignored when Exim is trying to
6691 sync a directory. This applies only to Linux.
6693 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6695 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6696 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6699 29. Added support for maildirsize files from supplied patch (modified a bit).
6701 30. The use of :fail: followed by an empty string could lead Exim to respond to
6702 sender verification failures with (e.g.):
6704 550 Verification failed for <xxx>
6705 550 Sender verify failed
6707 where the first response line was missing the '-' that indicates it is not
6708 the final line of the response.
6710 31. The loop for finding the name of the user that called Exim had a hardwired
6711 limit of 10; it now uses the value of finduser_retries, which is used for
6712 all other user lookups.
6714 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6717 33. Exim was neglecting to zero errno before one call of strtol() when
6718 expanding a string and expecting an integer value. On some systems this
6719 resulted in spurious "integer overflow" errors. Also, it was casting the
6720 result into an int without checking.
6722 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6723 did not work. The code looks as if it has *never* worked, though it appears
6724 to have been documented since at least release 1.62. I have made it work.
6726 35. The "timeout_DNS" error in retry rules, also documented since at least
6727 1.62, also never worked. As it isn't clear exactly what this means, and
6728 clearly it isn't a major issue, I have abolished the feature by treating it
6729 as "timeout", and writing a warning to the main and panic logs.
6731 36. The display of retry rules for -brt wasn't always showing the error code
6734 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6735 timeout_connect_A, timeout_connect_MX.
6737 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6738 to the empty sender.
6740 39. The daemon was not analysing the content of -oX till after it had closed
6741 stderr and disconnected from the controlling terminal. This meant that any
6742 syntax errors were only noted on the panic log, and the return code from
6743 the command was 0. By re-arranging the code a little, I've made the
6744 decoding happen first, so such errors now appear on stderr, and the return
6745 code is 1. However, the actual setting up of the sockets still happens in
6746 the disconnected process, so errors there are still only recorded on the
6749 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6750 connections (as happens on some IP stacks) was logged at start up time as
6751 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6752 it from "IPv6 and IPv4", which means that two separate sockets are being
6755 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6756 says whether AF_INET or AF_INET6 was passed as an argument.
6758 42. Exiwhat output was messed up when time zones were included in log
6761 43. Exiwhat now gives more information about the daemon's listening ports,
6762 and whether -tls-on-connect was used.
6764 44. The "port" option of the smtp transport is now expanded.
6766 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6767 silently ignored. Now an error message is written to the main and panic
6770 46. There's a new ACL modifier called "logwrite" which writes to a log file
6771 as soon as it is encountered.
6773 47. Added $local_user_uid and $local_user_gid at routing time.
6775 48. Exim crashed when trying to verify a sender address that was being
6778 49. Exim was recognizing only a space character after ".include". It now also
6779 recognizes a tab character.
6781 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6782 extracting the relevant information from the specification. The man page no
6783 longer contains scrambled data for the -d option, and I've added a section
6784 at the front about calling Exim under different names.
6786 51. Added "extra_headers" argument to the "mail" command in filter files.
6788 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6791 53. Installed eximstats 1.29.
6793 54. Added transport_filter_timeout as a generic transport option.
6795 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6796 Cc: header lines. This was required by RFC 822, but it not required by RFC
6799 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6800 incoming messages that did not have them. Now it does so only if the
6801 message originates locally, that is, if there is no associated remote host
6802 address. When Resent- header lines are present, this applies to the Resent-
6803 lines rather than the non-Resent- lines.
6805 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6806 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6808 58. Messages for configuration errors now include the name of the main
6809 configuration file - useful now that there may be more than one file in a
6810 list (.included file names were always shown).
6812 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6813 for those rare installations that do not start the daemon as root or run it
6814 setuid root. I've cut out the call to initgroups() if the daemon is not
6817 60. The Exim user and group can now be bound into the binary as text strings
6818 that are looked up at the start of Exim's processing.
6820 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6822 62. Added $mailstore_basename variable.
6824 63. Installed patch to sieve.c from Michael Haardt.
6826 64. When Exim failed to open the panic log after failing to open the main log,
6827 the original message it was trying to log was written to stderr and debug
6828 output, but if they were not available (the usual case in production), it
6829 was lost. Now it is written to syslog before the two lines that record the
6830 failures to open the logs.
6832 65. Users' Exim filters run in subprocesses under the user's uid. It is
6833 possible for a "deliver" command or an alias in a "personal" command to
6834 provoke an address rewrite. If logging of address rewriting is configured,
6835 this fails because the process is not running as root or exim. There may be
6836 a better way of dealing with this, but for the moment (because 4.30 needs
6837 to be released), I have disabled address rewrite logging when running a
6838 filter in a non-root, non-exim process.
6844 1. The buildconfig auxiliary program wasn't quoting the value set for
6845 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6846 not defined. This bug was masked in 4.22 by the effect that was fixed in
6849 2. Some messages that were rejected after a message id was allocated were
6850 shown as "incomplete" by exigrep. It no longer does this for messages that
6851 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6853 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6854 have allow_domain_literals set, the ID did not get logged in the <= line.
6855 Domain literals are now always recognized in Message-ID: header lines.
6857 4. The first argument for a ${extract expansion item is the key name or field
6858 number. Leading and trailing spaces in this item were not being ignored,
6859 causing some misleading effects.
6861 5. When deliver_drop_privilege was set, single queue runner processes started
6862 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6863 same command in the process it spins off) were not dropping privilege.
6865 6. When the daemon running as "exim" started a queue runner, it always
6866 re-executed Exim in the spun-off process. This is a waste of effort when
6867 deliver_drop_privilege is set. The new process now just calls the
6868 queue-runner function directly.
6874 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6877 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6878 ignored. Though the use of -oP was forcing the writing of a pid file, it
6879 was always written to the default place.
6881 3. If the message "no IP address found for host xxxx" is generated during
6882 incoming verification, it is now followed by identification of the incoming
6883 connection (so you can more easily find what provoked it).
6885 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6887 5. Added some features to "harden" Exim a bit more against certain attacks:
6889 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6890 be put in Local/Makefile. This is like the never_users runtime option,
6891 but it cannot be overridden. The default setting is "root".
6893 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6894 prefix string with which any file named in a -C command line option
6897 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6898 is retained for -C and -D only if the caller of Exim is root. Without
6899 it, the exim user may also use -C and -D and retain privilege.
6901 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6902 command line option is disabled.
6904 6. Macro names set by the -D option must start with an upper case letter, just
6905 like macro names defined in the configuration file.
6907 7. Added "dereference=" facility to LDAP.
6909 8. Two instances of the typo "uknown" in the source files are fixed.
6911 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6912 the Configure-Makefile script screwed up while processing it.
6914 10. Incorporated PCRE 4.4.
6916 11. The SMTP synchronization check was not operating right at the start of an
6917 SMTP session. For example, it could not catch a HELO sent before the client
6918 waited for the greeting. There is now a check for outstanding input at the
6919 point when the greeting is written. Because of the duplex, asynchronous
6920 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6921 way, but not yet received, when the check is performed.
6923 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6924 on TCP/IP sockets, because this apparently causes some broken clients to
6927 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6928 unchanged) from the Cygwin maintainer.
6930 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6931 was supported instead of testing for mbx. Effectively a typo.
6933 15. The spa authenticator server code was not checking that the input it
6934 received was valid base64.
6936 16. The debug output line for the "set" modifier in ACLs was not showing the
6937 name of the variable that was being set.
6939 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6941 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6942 lookup on its own. Something else was needed to trigger the lookup. For
6943 example, a match in host_lookup or the need for a host name in a host list.
6944 Now, if $sender_host_name is referenced and the host name has not yet been
6945 looked up, a lookup is performed. If the lookup fails, the variable remains
6946 empty, and $host_lookup_failed is set to "1".
6948 19. Added "eqi" as a case-independent comparison operator.
6950 20. The saslauthd authentication condition could segfault if neither service
6951 nor realm was specified.
6953 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6954 error message that was logged was misleading, and incoming SMTP
6955 connections were dropped. The message is now more accurate, and temporary
6956 errors are given to SMTP connections.
6958 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6959 (except RSET) with a 421 error, until QUIT is received. However, it was
6960 failing to send a response to QUIT.
6962 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6963 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6964 ACL is now run after the helo_try_verify_hosts code.
6966 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6967 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6968 case-independent, so other case variants are also recognized). Apparently
6969 some systems use these upper case variants.
6971 25. If more than two messages were waiting for the same host, and a transport
6972 filter was specified for the transport, Exim sent two messages over the
6973 same TCP/IP connection, and then failed with "socket operation on non-
6974 socket" when it tried to send the third.
6976 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6978 27. The extern definition of crypt16() in expand.c was not being excluded when
6979 the OS had its own crypt16() function.
6981 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6982 as a preferred synonym for return_size_limit, both as an option and as an
6985 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6987 30. Changed the default configuration ACL to relax the local part checking rule
6988 for addresses that are not in any local domains. For these addresses,
6989 slashes and pipe symbols are allowed within local parts, but the sequence
6990 /../ is explicitly forbidden.
6992 31. SPA server authentication was not clearing the challenge buffer before
6995 32. log_message in a "warn" ACL statement was writing to the reject log as
6996 well as to the main log, which contradicts the documentation and doesn't
6997 seem right (because no rejection is happening). So I have stopped it.
6999 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
7000 However, I am unable to do any testing of this.
7002 34. Fixed an infelicity in the appendfile transport. When checking directories
7003 for a mailbox, to see if any needed to be created, it was accidentally
7004 using path names with one or more superfluous leading slashes; tracing
7005 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
7007 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
7008 contents are added to the log line that is written for every discarded
7009 recipient. (Previously a log_message setting was ignored.)
7011 36. The ${quote: operator now quotes the string if it is empty.
7013 37. The install script runs exim in order to find its version number. If for
7014 some reason other than non-existence or emptiness, which it checks, it
7015 could not run './exim', it was installing it with an empty version number,
7016 i.e. as "exim-". This error state is now caught, and the installation is
7019 38. An argument was missing from the function that creates an error message
7020 when Exim fails to connect to the socket for saslauthd authentication.
7021 This could cause Exim to crash, or give a corrupted message.
7023 39. Added isip, isip4, and isip6 to ${if conditions.
7025 40. The ACL variables $acl_xx are now saved with the message, and can be
7026 accessed later in routers, transports, and filters.
7028 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
7029 strings in the file are not string-expanded.
7031 42. If a MAIL command specified a SIZE value that was too large to fit into an
7032 int variable, the check against message_size_limit failed. Such values are
7033 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
7034 day this will have to be increased, but I don't think I want to be around
7035 when emails are that large.
7042 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
7043 iconv() is not standard in FreeBSD.
7045 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
7046 IPv6 enabled. The observed symptom was a segmentation fault on return from
7047 the function os_common_find_running_interfaces() in src/os.c.
7049 3. In the check_special_case() function in daemon.c I had used "errno" as an
7050 argument name, which causes warnings on some systems. This was basically a
7051 typo, since it was named "eno" in the comments!
7053 4. The code that waits for the clock to tick (at a resolution of some fraction
7054 of a second) so as to ensure message-id uniqueness was always waiting for
7055 at least one whole tick, when it could have waited for less. [This is
7056 almost certainly not relevant at current processor speeds, where it is
7057 unlikely to ever wait at all. But we try to future-proof.]
7059 5. The function that sleeps for a time interval that includes fractions of a
7060 second contained a race. It did not block SIGALRM between setting the
7061 timer, and suspending (a couple of lines later). If the interval was short
7062 and the sigsuspend() was delayed until after it had expired, the suspension
7063 never ended. On busy systems this could lead to processes getting stuck for
7066 6. Some uncommon configurations may cause a lookup to happen in a queue runner
7067 process, before it forks any delivery processes. The open lookup caching
7068 mechanism meant that the open file or database connection was passed into
7069 the delivery process. The problem was that delivery processes always tidy
7070 up cached lookup data. This could cause a problem for the next delivery
7071 process started by the queue runner, because the external queue runner
7072 process does not know about the closure. So the next delivery process
7073 still has data in the lookup cache. In the case of a file lookup, there was
7074 no problem because closing a file descriptor in a subprocess doesn't affect
7075 the parent. However, if the lookup was caching a connection to a database,
7076 the connection was closed, and the second delivery process was likely to
7077 see errors such as "PGSQL: query failed: server closed the connection
7078 unexpectedly". The problem has been fixed by closing all cached lookups
7079 in a queue runner before running a delivery process.
7081 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
7082 seem to have the "const" qualifier which it has on other OS. I've
7085 8. Change 4.21/2 was too strict. It is only if there are two authenticators
7086 *of the same type* (client or server) with the same public name that an
7087 error should be diagnosed.
7089 9. When Exim looked up a host name for an IP address, but failed to find the
7090 original IP address when looking up the host name (a safety check), it
7091 output the message "<ip address> does not match any IP for NULL", which was
7092 confusing, to say the least. The bug was that the host name should have
7093 appeared instead of "NULL".
7095 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
7096 user that is built into the binary, and the -C or -D options is used, root
7097 privilege is dropped before the configuration file is read. In addition,
7098 logging is switched to stderr instead of the normal log files. If the
7099 configuration then re-defines the Exim user, the unprivileged environment
7100 is probably not what is expected, so Exim logs a panic warning message (but
7103 However, if deliver_drop_privilege is set, the unprivileged state may well
7104 be exactly what is intended, so the warning has been cut out in that case,
7105 and Exim is allowed to try to write to its normal log files.
7111 1. smtp_return_error_details was not giving details for temporary sender
7112 or receiver verification errors.
7114 2. Diagnose a configuration error if two authenticators have the same public
7117 3. Exim used not to create the message log file for a message until the first
7118 delivery attempt. This could be confusing when incoming messages were held
7119 for policy or load reasons. The message log file is now created at the time
7120 the message is received, and an initial "Received" line is written to it.
7122 4. The automatically generated man page for command line options had a minor
7123 bug that caused no ill effects; however, a more serious problem was that
7124 the procedure for building the man page automatically didn't always
7125 operate. Consequently, release 4.20 contains an out-of-date version. This
7126 shouldn't happen again.
7128 5. When building Exim with embedded Perl support, the script that builds the
7129 Makefile was calling 'perl' to find its compile-time parameters, ignoring
7130 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
7132 6. The freeze_tell option was not being used for messages that were frozen on
7133 arrival, either by an ACL or by local_scan().
7135 7. Added the smtp_incomplete_transaction log selector.
7137 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
7138 was accepting AUTH without a new EHLO.
7140 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
7141 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
7144 10. Logging of TCP/IP connections (when configured) now happens in the main
7145 daemon process instead of the child process, so that the TCP/IP connection
7146 count is more accurate (but it can never be perfect).
7148 11. The use of "drop" in a nested ACL was not being handled correctly in the
7149 outer ACL. Now, if condition failure induced by the nested "drop" causes
7150 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
7151 or "require"), the connection is dropped.
7153 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
7154 that yield "discard" can now be used with an "accept" or a "discard" verb,
7155 but an error is generated for any others (because I can't see a useful way
7156 to define what should happen).
7158 13. When an ACL is read dynamically from a file (or anywhere else), the lines
7159 are now processed in the same way as lines in the Exim configuration file.
7160 In particular, continuation lines are supported.
7162 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
7164 15. Added -ti meaning -t -i.
7166 16. Check for letters, digits, hyphens, and dots in the names of dnslist
7167 domains, and warn by logging if others are found.
7169 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
7170 returned from GIFCONF when Exim is trying to find the list of interfaces on
7171 a host. The code in os.c has been modified to copy each ifreq to an aligned
7172 structure in all cases.
7174 Also, in some cases, the returned ifreq's were being copied to a 'struct
7175 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
7176 means the last couple of bytes of an IPv6 address could be chopped if the
7177 ifreq contained only a normal sockaddr (14 bytes storage).
7179 18. Named domain lists were not supported in the hosts_treat_as_local option.
7180 An entry such as +xxxx was not recognized, and was treated as a literal
7183 19. Ensure that header lines added by a DATA ACL are included in the reject log
7184 if the ACL subsequently rejects the message.
7186 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
7187 MD5 (which is deprecated).
7189 21. When testing a filter file using -bf, Exim was writing a message when it
7190 took the sender from a "From " line in the message, but it was not doing so
7191 when it took $return_path from a Return-Path: header line. It now does.
7193 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
7194 with a valid header line field name (a series of printing characters
7195 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
7197 23. Changed "disc" in the source to "disk" to conform to the documentation and
7198 the book and for uniformity.
7200 24. Ignore Sendmail's -Ooption=value command line item.
7202 25. When execve() failed while trying to run a command in a pipe transport,
7203 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
7204 could be confused with a return value of 69 from the command itself. This
7205 has been changed to 127, the value the shell returns if it is asked to run
7206 a non-existent command. The wording for the related log line suggests a
7207 non-existent command as the problem.
7209 26. If received_header_text expands to an empty string, do not add a Received:
7210 header line to the message. (Well, it adds a token one on the spool, but
7211 marks it "old" so that it doesn't get used or transmitted.)
7213 27. Installed eximstats 1.28 (addition of -nt option).
7215 28. There was no check for failure on the call to getsockname() in the daemon
7216 code. This can fail if there is a shortage of resources on the system, with
7217 ENOMEM, for example. A temporary error is now given on failure.
7219 29. Contrary to the C standard, it seems that in some environments, the
7220 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
7221 program. Exim now does this explicitly; it affects the formatting of
7222 timestamps using strftime().
7224 30. If exiqsumm was given junk data, it threw up some uninitialized variable
7225 complaints. I've now initialized all the variables, to avoid this.
7227 32. Header lines added by a system filter were not being "seen" during
7228 transport-time rewrites.
7230 33. The info_callback() function passed to OpenSSL is set up with type void
7231 (*)(SSL *, int, int), as described somewhere. However, when calling the
7232 function (actually a macro) that sets it up, the type void(*)() is
7233 expected. I've put in a cast to prevent warnings from picky compilers.
7235 34. If a DNS black list lookup found a CNAME record, but there were no A
7236 records associated with the domain it pointed at, Exim crashed.
7238 35. If a DNS black list lookup returned more than one A record, Exim ignored
7239 all but the first. It now scans all returned addresses if a particular IP
7240 value is being sought. In this situation, the contents of the
7241 $dnslist_value variable are a list of all the addresses, separated by a
7244 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
7245 to accept a host name and all its aliases if the forward lookup for any of
7246 them yielded the IP address of the incoming connection. Now it accepts only
7247 those names whose forward lookup yields the correct IP address. Any other
7248 names are discarded. This closes a loophole whereby a rogue DNS
7249 administrator could create reverse DNS records to break through a
7250 wildcarded host restriction in an ACL.
7252 37. If a user filter or a system filter that ran in a subprocess used any of
7253 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
7254 the wrong values were passed to the pipe command ($thisaddress had the
7255 value of $0, $0 had the value of $1, etc). This bug was introduced by
7256 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
7258 38. Improved the line breaking for long SMTP error messages from ACLs.
7259 Previously, if there was no break point between 40 and 75 characters, Exim
7260 left the rest of the message alone. Two changes have been made: (a) I've
7261 reduced the minimum length to 35 characters; (b) if it can't find a break
7262 point between 35 and 75 characters, it looks ahead and uses the first one
7263 that it finds. This may give the occasional overlong line, but at least the
7264 remaining text gets split now.
7266 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
7267 file descriptors might be low, and that setting 1000 would always raise it.
7268 It turns out that in some environments, the limit is already over 1000 and
7269 that lowering it causes trouble. So now Exim takes care not to decrease it.
7271 40. When delivering a message, the value of $return_path is set to $sender_
7272 address at the start of routing (routers may change the value). By an
7273 oversight, this default was not being set up when an address was tested by
7274 -bt or -bv, which affected the outcome if any router or filter referred to
7277 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
7278 log only when "message" or "log_message" are set. However, if one of the
7279 conditions was an address verification, or a call to a nested ACL, the
7280 messages generated by the underlying test were being passed through. This
7281 no longer happens. The underlying message is available in $acl_verify_
7282 message for both "message" and "log_message" expansions, so it can be
7283 passed through if needed.
7285 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
7286 new expansion $bh_ to give the encoded byte string without charset
7287 translation. Translation happens only if iconv() is available; HAVE_ICONV
7288 indicates this at build time. HEADERS_CHARSET gives the charset to
7289 translate to; headers_charset can change it in the configuration, and
7290 "headers charset" can change it in an individual filter file.
7292 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
7293 that creates RFC 2047 encoded "words" labels them as that charset instead
7294 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
7295 expansion operator; (ii) when Exim creates a From: line for a local
7296 message; (iii) when a header line is rewritten to include a "phrase" part.
7298 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
7299 buggy, causing it to skip the first lines of messages whose message ID
7300 ended in 'D'. This would not have bitten before Exim release 4.14, because
7301 message IDs were unlikely to end in 'D' before then. The effect was to have
7302 incorrect size information for certain domains.
7304 45. #include "config.h" was missing at the start of the crypt16.c module. This
7305 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
7308 46. If there was a timeout during a "random" callout check, Exim treated it as
7309 a failure of the random address, and carried on sending RSET and the real
7310 address. If the delay was just some slowness somewhere, the response to the
7311 original RCPT would be taken as a response to RSET and so on, causing
7312 mayhem of various kinds.
7314 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
7315 when I implemented it. It didn't allow for the fact that some option values
7316 may legitimately be negative (e.g. size_addition), and it didn't even do
7317 the right test for positive values.
7319 48. Domain names in DNS records are case-independent. Exim always looks them up
7320 in lower case. Some resolvers return domain names in exactly the case they
7321 appear in the zone file, that is, they may contain uppercase letters. Not
7322 all resolvers do this - some return always lower case. Exim was treating a
7323 change of case by a resolver as a change of domain, similar to a widening
7324 of a domain abbreviation. This triggered its re-routing code and so it was
7325 trying to route what was effectively the same domain again. This normally
7326 caused routing to fail (because the router wouldn't handle the domain
7327 twice). Now Exim checks for this case specially, and just changes the
7328 casing of the domain that it ultimately uses when it transmits the message
7331 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
7334 50. If a filter generated a file delivery with a non-absolute name (possible if
7335 no home directory exists for the router), the forbid_file option was not
7338 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
7339 the existing equality matching.
7341 52. Exim was using ints instead of ino_t variables in some places where it was
7342 dealing with inode numbers.
7344 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
7345 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
7346 called TMPDIR, and if it finds it is different, it changes its value.
7348 54. The smtp_printf() function is now made available to local_scan() so
7349 additional output lines can be written before returning. There is also an
7350 smtp_fflush() function to enable the detection of a dropped connection.
7351 The variables smtp_input and smtp_batched_input are exported to
7354 55. Changed the default runtime configuration: the message "Unknown user"
7355 has been removed from the ACL, and instead placed on the localuser router,
7356 using the cannot_route_message feature. This means that any verification
7357 failures that generate their own messages won't get overridden. Similarly,
7358 the "Unrouteable address" message that was in the ACL for unverifiable
7359 relay addresses has also been removed.
7361 56. Added hosts_avoid_esmtp to the smtp transport.
7363 57. The exicyclog script was not checking for the esoteric option
7364 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
7365 will work only if exicyclog is run under the appropriate euid.
7367 58. Following a discussion on the list, the rules by which Exim recognises line
7368 endings on incoming messages have been changed. The -dropcr and drop_cr
7369 options are now no-ops, retained only for backwards compatibility. The
7370 following line terminators are recognized: LF CRLF CR. However, special
7371 processing applies to CR:
7373 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7374 nor a local message in the state where . is a terminator.
7376 (ii) If a bare CR is encountered in a header line, an extra space is added
7377 after the line terminator so as not to end the header. The reasoning
7378 behind this is that bare CRs in header lines are most likely either
7379 to be mistakes, or people trying to play silly games.
7381 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7382 was being incorrectly given as 18 bytes larger than it should have been.
7383 This is a VOB (very old bug).
7385 60. This may never have affected anything current, but just in case it has:
7386 When the local host is found other than at the start of a list of hosts,
7387 the local host, those with the same MX, and any that follow, are discarded.
7388 When the list in question was part of a longer list of hosts, the following
7389 hosts (not currently being processed) were also being discarded. This no
7390 longer happens. I'm not sure if this situation could ever has previously
7393 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7396 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7397 header. This is something that is recommended in a new Internet Draft, and
7398 is something that is documented as being done by Sendmail. There are two
7399 possible values. For messages generated by the autoreply transport, Exim
7402 Auto-Submitted: auto-replied
7404 whereas for all other generated messages (e.g. bounces) it adds
7406 Auto-Submitted: auto-generated
7408 63. The "personal" condition in filters now includes a test for the
7409 Auto-Submitted: header. If it contains the string "auto-" the message it
7410 not considered personal.
7412 64. Added rcpt_include_affixes as a generic transport option.
7414 65. Added queue_only_override (default true).
7416 66. Added the syslog_duplication option.
7418 67. If what should have been the first header line of a message consisted of
7419 a space followed by a colon, Exim was mis-interpreting it as a header line.
7420 It isn't of course - it is syntactically invalid and should therefore be
7421 treated as the start of the message body. The misbehaviour could have
7422 caused a number of strange effects, including loss of data in subsequent
7423 header lines, and spool format errors.
7425 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7426 client host had authenticated. This control can now be exercised by an ACL
7427 for more flexibility.
7429 69. By default, callouts do not happen when testing with -bh. There is now a
7430 variant, -bhc, which does actually run the callout code, including
7431 consulting and updating the callout cache.
7433 70. Added support for saslauthd authentication, courtesy of Alexander
7436 71. If statvfs() failed on the spool or log directories while checking their
7437 size for availability, Exim confusingly gave the error "space shortage".
7438 Furthermore, in debugging mode it crashed with a floating point exception.
7439 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7440 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7441 serious problem, Exim now writes to the main and panic logs when this
7442 happens, with details of the failure. It then refuses to accept the
7443 incoming message, giving the message "spool directory problem" or "log
7444 directory problem" with a 421 code for SMTP messages.
7446 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7447 0, 1, and 2 exist, because some OS complain for execs without them (see
7448 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7449 descriptors. However, the code omitted to check that the open succeeded,
7450 causing mysterious errors if for some reason the permissions on /dev/null
7451 got screwed. Now Exim writes a message to the main and panic logs, and
7452 bombs out if it can't open /dev/null.
7454 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7455 interact so that it is all more flexible. It is supposed to remain
7456 backwards compatible. Also added extra_local_interfaces.
7458 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7459 to bomb out with an assertion failure - to the client this appears as a
7460 connection drop. This problem occurs in the part of the code that was taken
7461 from the Samba project. Fortunately, the assertion is in a very simple
7462 function, so I have fixed this by reproducing the function inline in the
7463 one place where it is called, and arranging for authentication to fail
7464 instead of killing the process with assert().
7466 75. The SPA client code was not working when the server requested OEM rather
7467 than Unicode encoding.
7469 76. Added code to make require_files with a specific uid setting more usable in
7470 the case where statting the file as root fails - usually a non-root-mounted
7471 NFS file system. When this happens and the failure is EACCES, Exim now
7472 forks a subprocess and does the per-uid checking as the relevant uid.
7474 77. Added process_log_path.
7476 78. If log_file_path was not explicitly set, a setting of check_log_space or
7477 check_log_inodes was ignored.
7479 79. If a space check for the spool or log partitions fails, the incident is now
7480 logged. Of course, in the latter case the data may get lost...
7482 80. Added the %p formatting code to string_format() so that it can be used to
7483 print addresses in debug_print(). Adjusted all the address printing in the
7484 debugging in store.c to use %p rather than %d.
7486 81. There was a concern that a line of code in smtp_in.c could overflow a
7487 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7488 initially expressed, the concern was not well-founded, because trailing
7489 spaces are removed early. However, if the trailing spaces were followed by
7490 a NULL, they did not get removed, so the overflow was possible. Two fixes
7493 (a) I re-wrote the offending code in a cleaner fashion.
7494 (b) If an incoming SMTP command contains a NULL character, it is rejected
7497 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7498 runs initgroups(), so that if the Exim user is in any additional groups,
7499 they will be used during message reception.
7505 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob