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 Replace the store_release() internal interface with store_newblock(),
12 which internalises the check required to safely use the old one, plus
13 the allocate and data copy operations duplicated in both (!) of the
16 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
17 modifier. This matches the restriction on the commandline.
19 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
20 Previously only the last row was returned.
22 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
23 we assumed that tags in the header were well-formed, and parsed the
24 element content after inspecting only the first char of the tag.
25 Assumptions at that stage could crash the receive process on malformed
28 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
29 While running the DKIM ACL we operate on the Permanent memory pool so that
30 variables created with "set" persist to the DATA ACL. Also (at any time)
31 DNS lookups that fail create cache records using the Permanent pool. But
32 expansions release any allocations made on the current pool - so a dnsdb
33 lookup expansion done in the DKIM ACL releases the memory used for the
34 DNS negative-cache, and bad things result. Solution is to switch to the
35 Main pool for expansions.
36 While we're in that code, add checks on the DNS cache during store_reset,
37 active in the testsuite.
38 Problem spotted, and debugging aided, by Wolfgang Breyha.
40 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
41 When none of the hosts presented to a transport match an already-open
42 connection, close it and proceed with the list. Previously we would
43 queue the message. Spotted by Lena with Yahoo, probably involving
46 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
47 Previously a spurious "250 OK id=" response was appended to the proper
50 JH/08 The "support for" informational output now, which built with Content
51 Scanning support, has a line for the malware scanner interfaces compiled
52 in. Interface can be individually included or not at build time.
54 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
55 by the template makefile "src/EDITME". The "STREAM" support for an older
56 ClamAV interface method is removed.
58 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
59 rows affected is given instead).
61 JH/11 The runtime Berkeley DB library version is now additionally output by
62 "exim -d -bV". Previously only the compile-time version was shown.
64 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
65 SMTP connection. Previously, when one had more receipients than the
66 first, an abortive onward connection was made. Move to full support for
67 multiple onward connections in sequence, handling cutthrough connection
68 for all multi-message initiating connections.
70 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
71 routers. Previously, a multi-recipient message would fail to match the
72 onward-connection opened for the first recipient, and cause its closure.
74 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
75 a timeout on read on a GnuTLS initiating connection, resulting in the
76 initiating connection being dropped. This mattered most when the callout
77 was marked defer_ok. Fix to keep the two timeout-detection methods
80 JH/15 Relax results from ACL control request to enable cutthrough, in
81 unsupported situations, from error to silently (except under debug)
82 ignoring. This covers use with PRDR, frozen messages, queue-only and
85 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
86 metadata, resulting in a crash in free().
88 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
89 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
90 Broken also in d185889f4, with init system revamp.
96 JH/01 Rework error string handling in TLS interface so that the caller in
97 more cases is responsible for logging. This permits library-sourced
98 string to be attached to addresses during delivery, and collapses
99 pairs of long lines into single ones.
101 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
102 during configuration. Wildcards are allowed and expanded.
104 JH/02 Rework error string handling in DKIM to pass more info back to callers.
105 This permits better logging.
107 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
108 do not close it down and have the child transport start it up again on
109 the passed-on TCP connection. Instead, proxy the child (and any
110 subsequent ones) for TLS via a unix-domain socket channel. Logging is
111 affected: the continued delivery log lines do not have any DNSSEC, TLS
112 Certificate or OCSP information. TLS cipher information is still logged.
114 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
115 identical IP addresses on different listening ports. Will also affect
118 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
119 add noisy ifdef guards to special-case this sillyness.
120 Patch from Bernd Kuhls.
122 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
123 than 255 are no longer allowed.
125 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
126 Disable the session-cache too, which might reduce our load. Since we
127 currrectly use a new context for every connection, both as server and
128 client, there is no benefit for these.
129 GnuTLS appears to not support tickets server-side by default (we don't
130 call gnutls_session_ticket_enable_server()) but client side is enabled
131 by default on recent versions (3.1.3 +) unless the PFS priority string
134 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
135 <https://reproducible-builds.org/specs/source-date-epoch/>.
137 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
138 the check for any unsuccessful recipients did not notice the limit, and
139 erroneously found still-pending ones.
141 JH/08 Pipeline CHUNKING command and data together, on kernels that support
142 MSG_MORE. Only in-clear (not on TLS connections).
144 JH/09 Avoid using a temporary file during transport using dkim. Unless a
145 transport-filter is involved we can buffer the headers in memory for
146 creating the signature, and read the spool data file once for the
147 signature and again for transmission.
149 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
150 4.77 as the kernel support then wasn't solid, having issues in 64bit
151 mode. Now, it's been long enough. Add support for FreeBSD also.
153 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
154 case where the routing stage had gathered several addresses to send to
155 a host before calling the transport for the first, we previously failed
156 to close down TLS in the old transport process before passing the TCP
157 connection to the new process. The new one sent a STARTTLS command
158 which naturally failed, giving a failed delivery and bloating the retry
159 database. Investigation and fix prototype from Wolfgang Breyha.
161 JH/12 Fix check on SMTP command input synchronisation. Previously there were
162 false-negatives in the check that the sender had not preempted a response
163 or prompt from Exim (running as a server), due to that code's lack of
164 awareness of the SMTP input buffering.
166 PP/04 Add commandline_checks_require_admin option.
167 Exim drops privileges sanely, various checks such as -be aren't a
168 security problem, as long as you trust local users with access to their
169 own account. When invoked by services which pass untrusted data to
170 Exim, this might be an issue. Set this option in main configuration
171 AND make fixes to the calling application, such as using `--` to stop
174 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
175 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
176 responses to those, into a single TLS record each way (this usually means
177 a single packet). As a side issue, smtp_enforce_sync now works on TLS
180 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
181 affects you only if you're dancing at the edge of the param size limits.
182 If you are, and this message makes sense to you, then: raise the
183 configured limit or use OpenSSL 1.1. Nothing we can do for older
186 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
187 cmdline element to get the documented default one. Previously it was
190 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
191 are unsuable for non-message-specific reasons - eg. connection timeout,
192 banner-time rejection.
194 JH/16 Fix logging of delivery remote port, when specified by router, under
197 PP/06 Repair manualroute's ability to take options in any order, even if one
198 is the name of a transport.
201 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
203 JH/17 Change the list-building routines interface to use the expanding-string
204 triplet model, for better allocation and copying behaviour.
206 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
207 Previously it was constructed the first time a possibly-matching string
208 was met in the configuration file input during startup; now it is done
211 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
212 compatible one, to avoid the (poorly documented) possibility of a config
213 file in the working directory redirecting the DB files, possibly correpting
214 some existing file. CVE-2017-10140 assigned for BDB.
216 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
217 cache-hot. Previously, although the result was properly cached, the
218 initial verify call returned a defer.
220 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
221 the main verify for receipient in uncached-mode.
223 JH/22 Retire historical build files to an "unsupported" subdir. These are
224 defined as "ones for which we have no current evidence of testing".
226 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
227 if present. Previously it was ignored.
229 JH/24 Start using specified-initialisers in C structure init coding. This is
230 a C99 feature (it's 2017, so now considered safe).
232 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
233 if was a fixed-sized field and bitmask ops via macros; it is now more
236 PP/07 GitHub PR 56: Apply MariaDB build fix.
237 Patch provided by Jaroslav Škarvada.
239 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
240 during Coverity cleanups [4.87 JH/47]
241 Diagnosis and fix provided by Michael Fischer v. Mollard.
243 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
244 the right size to place the terminating semicolon on its own folded
245 line, the header hash was calculated to an incorrect value thanks to
246 the (relaxed) space the fold became.
248 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
249 and confused the parent.
251 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
252 which could crash as a result. This could lead to undeliverable messages.
254 JH/28 Logging: "next input sent too soon" now shows where input was truncated
257 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
258 matters on fast-turnover and PID-randomising systems, which were getting
259 out-of-order delivery.
261 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
262 a possibly-overlapping copy. The symptom was that "Remote host closed
263 connection in response to HELO" was logged instead of the actual 4xx
266 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
267 Previously only that bufferd was discarded, resulting in SYMTP command
270 JH/32 DKIM: when a message has multiple signatures matching an identity given
271 in dkim_verify_signers, run the dkim acl once for each. Previously only
272 one run was done. Bug 2189.
274 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
275 "panic the current process" to "deliberately defer". The panic log is
276 still written with the problem list name; the mail and reject logs now
277 get a temp-reject line for the message that was being handled, saying
278 something like "domains check lookup or other defer". The SMTP 451
279 message is still "Temporary local problem".
281 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
282 A crafted sequence of BDAT commands could result in in-use memory beeing
283 freed. CVE-2017-16943.
285 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
286 from SMTP input. Previously it was always done; now only done for DATA
287 and not BDAT commands. CVE-2017-16944.
289 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
290 to the message (such as an overlong header line). Previously this was
291 not done and we did not exit BDAT mode. Followon from the previous item
292 though a different problem.
298 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
299 than -2003 did; needs libidn2 in addition to libidn.
301 JH/02 The path option on a pipe transport is now expanded before use.
303 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
304 Patch provided by "Björn", documentation fix added too.
306 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
307 missing a wire-to-host endian conversion.
309 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
310 close after a BDAT command line could be taken as a following command,
311 giving a synch failure. Fix by only checking for synch immediately
312 before acknowledging the chunk.
314 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
315 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
316 macro. Patches provided by Josh Soref.
318 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
319 Previously we did not; the RFC seems ambiguous and VRFY is not listed
320 by IANA as a service extension. However, John Klensin suggests that we
323 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
324 the dkim code may be unix-mode line endings rather than smtp wire-format
325 CRLF, so prepend a CR to any bare LF.
327 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
328 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
330 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
331 the first were themselves being wrongly included in the feed into dkim
332 processing; with most chunk sizes in use this resulted in an incorrect
333 body hash calculated value.
335 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
336 DKIM signature block, for verification. Although advised against by
337 standards it is specifically not ruled illegal.
339 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
341 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
342 missing a body hash (the bh= tag).
344 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
345 It seems that HAProxy sends the Proxy Protocol information in clear and
346 only then does a TLS startup, so do the same.
348 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
349 TCP connections (such as for Spamd) unless the daemon successfully set
350 Fast Open mode on its listening sockets. This fixes breakage seen on
351 too-old kernels or those not configured for Fast Open, at the cost of
352 requiring both directions being enabled for TFO, and TFO never being used
353 by non-daemon-related Exim processes.
355 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
356 endings, at least on the first header line. Try to canonify any that get
357 past that check, despite the cost.
359 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
360 now limited to an arbitrary five deep, while parsing addresses with the
361 strip_excess_angle_brackets option enabled.
363 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
364 instead leave the unprompted TLS handshake in socket buffer for the
365 TLS library to consume.
367 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
369 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
371 JH/16 Drop variables when they go out of scope. Memory management drops a whole
372 region in one operation, for speed, and this leaves assigned pointers
373 dangling. Add checks run only under the testsuite which checks all
374 variables at a store-reset and panics on a dangling pointer; add code
375 explicitly nulling out all the variables discovered. Fixes one known
376 bug: a transport crash, where a dangling pointer for $sending_ip_address
377 originally assigned in a verify callout, is re-used.
379 PP/06 Drop '.' from @INC in various Perl scripts.
381 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
383 PP/08 Reduce a number of compilation warnings under clang; building with
384 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
385 should be warning-free.
387 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
389 HS/01 Fix portability problems introduced by PP/08 for platforms where
390 realloc(NULL) is not equivalent to malloc() [SunOS et al].
392 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
393 chunk. This allows us to accept broken chunked messages. We need a more
394 general solution here.
396 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
397 already-broken messages in the queue.
399 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
401 JH/19 Fix reference counting bug in routing-generated-address tracking.
407 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
408 supports it and a size is available (ie. the sending peer gave us one).
410 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
411 years of being deprecated). The replacements are the ACLs
412 acl_smtp_mime and acl_not_smtp_mime.
414 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
415 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
416 or even an in-clear connection were permitted. Now, if the host lookup
417 was dnssec and dane was requested then the host is only used if the
418 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
419 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
420 if one fails this test.
421 This means that a poorly-configured remote DNS will make it incommunicado;
422 but it protects against a DNS-interception attack on it.
424 JH/04 Bug 1810: make continued-use of an open smtp transport connection
425 non-noisy when a race steals the message being considered.
427 JH/05 If main configuration option tls_certificate is unset, generate a
428 self-signed certificate for inbound TLS connections.
430 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
431 in rewrites and routers.
433 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
434 and logged a warning sing 4.83; now they are a configuration file error.
436 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
437 (lacking @domain). Apply the same qualification processing as RCPT.
439 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
441 JH/10 Support ${sha256:} applied to a string (as well as the previous
444 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
445 a cutthrough deliver is pending, as we always want to make a connection.
446 This also avoids re-routing the message when later placing the cutthrough
447 connection after a verify cache hit.
448 Do not update it with the verify result either.
450 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
451 when routing results in more than one destination address.
453 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
454 signing (which inhibits the cutthrough capability). Previously only
455 the presence of an option was tested; now an expansion evaluating as
456 empty is permissible (obviously it should depend only on data available
457 when the cutthrough connection is made).
459 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
460 the relevant preceding SMTP command did not note the pipelining mode.
462 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
463 Previously they were not counted.
465 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
466 as one having no matching records. Previously we deferred the message
467 that needed the lookup.
469 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
470 distinguished as "(=".
472 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
473 for missing MX records. Previously it only worked for missing A records.
475 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
477 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
478 after the data-go-ahead and data-ack. Patch from Jason Betts.
480 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
481 even for a "none" policy. Patch from Tony Meyer.
483 JH/22 Fix continued use of a connection for further deliveries. If a port was
484 specified by a router, it must also match for the delivery to be
487 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
488 When one of the recipients of a message was unsuitable for the connection
489 (has no matching addresses), we lost track of needing to mark it
490 deferred. As a result mail would be lost.
492 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
494 JH/25 Decoding ACL controls is now done using a binary search; the source code
495 takes up less space and should be simpler to maintain. Merge the ACL
496 condition decode tables also, with similar effect.
498 JH/26 Fix problem with one_time used on a redirect router which returned the
499 parent address unchanged. A retry would see the parent address marked as
500 delivered, so not attempt the (identical) child. As a result mail would
503 JH/27 Fix a possible security hole, wherein a process operating with the Exim
504 UID can gain a root shell. Credit to http://www.halfdog.net/ for
505 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
508 JH/28 Enable {spool,log} filesystem space and inode checks as default.
509 Main config options check_{log,spool}_{inodes,space} are now
510 100 inodes, 10MB unless set otherwise in the configuration.
512 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
513 Previously it only applied to the main-section connection policy
516 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
518 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
519 by me. Added RFC7919 DH primes as an alternative.
521 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
522 are not in the system include path.
524 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
525 GnuTLS, when a session startup failed (eg because the client disconnected)
526 Exim did stdio operations after fclose. This was exposed by a recent
527 change which nulled out the file handle after the fclose.
529 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
530 signed directly by the cert-signing cert, rather than an intermediate
531 OCSP-signing cert. This is the model used by LetsEncrypt.
533 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
535 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
536 an incoming connection.
538 HS/02 Bug 1802: Do not half-close the connection after sending a request
541 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
542 fallback to "prime256v1".
544 JH/34 SECURITY: Use proper copy of DATA command in error message.
545 Could leak key material. Remotely exploitable. CVE-2016-9963.
551 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
552 and 3.4.4 - once the server is enabled to respond to an OCSP request
553 it does even when not requested, resulting in a stapling non-aware
554 client dropping the TLS connection.
556 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
557 support variable-length bit vectors. No functional change.
559 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
560 The I= interface field on outgoing lines is now after the H= remote
561 host field, same as incoming lines. There is a separate
562 outgoing_interface log selector which allows you to disable the
565 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
566 If not running log_selector +smtp_connection the mainlog would be held
567 open indefinitely after a "too many connections" event, including to a
568 deleted file after a log rotate. Leave the per net connection logging
569 leaving it open for efficiency as that will be quickly detected by the
570 check on the next write.
572 HS/01 Bug 1671: Fix post transport crash.
573 Processing the wait-<transport> messages could crash the delivery
574 process if the message IDs didn't exist for some reason. When
575 using 'split_spool_directory=yes' the construction of the spool
576 file name failed already, exposing the same netto behaviour.
578 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
579 mime_regex ACL conditions.
581 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
582 to DSN fail messages (bounces): remote IP, remote greeting, remote response
583 to HELO, local diagnostic string.
585 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
586 log line to debug. Even when configured with a tls authenticator many
587 client connections are expected to not authenticate in this way, so
588 an authenticate fail is not an error.
590 HS/02 Add the Exim version string to the process info. This way exiwhat
591 gives some more detail about the running daemon.
593 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
594 matter for fast-change records such as DNSBLs.
596 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
597 retry record, even if constant. There may be multiple transports with
598 different interface settings and the retry behaviour needs to be kept
601 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
603 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
605 JH/10 Bug 840: fix log_defer_output option of pipe transport
607 JH/11 Bug 830: use same host for all RCPTS of a message, even under
608 hosts_randomize. This matters a lot when combined with mua_wrapper.
610 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
611 ${quote_pgsql:<string>} operator.
613 JH/13 Bug 1708: avoid misaligned access in cached lookup.
615 JH/14 Change header file name for freeradius-client. Relevant if compiling
616 with Radius support; from the Gentoo tree and checked under Fedora.
618 JH/15 Bug 1712: Introduce $prdr_requested flag variable
620 JH/16 Bug 1714: Permit an empty string as expansion result for transport
621 option transport_filter, meaning no filtering.
623 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
625 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
626 defaults to "*" (all hosts). The variable is now available when not built
627 with TLS, default unset, mainly to enable keeping the testsuite sane.
628 If a server certificate is not supplied (via tls_certificate) an error is
629 logged, and clients will find TLS connections fail on startup. Presumably
630 they will retry in-clear.
631 Packagers of Exim are strongly encouraged to create a server certificate
632 at installation time.
634 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
635 with the $config_file variable.
637 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
638 in transport context, after the attempt, and per-recipient. The latter type
639 is per host attempted. The event data is the error message, and the errno
640 information encodes the lookup type (A vs. MX) used for the (first) host,
641 and the trailing two digits of the smtp 4xx response.
643 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
644 to write to mainlog (or rejectlog, paniclog) in the window between file
645 creation and permissions/ownership being changed. Particularly affects
646 installations where exicyclog is run as root, rather than exim user;
647 result is that the running daemon panics and dies.
649 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
651 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
652 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
653 "pri" and "weight". Note that the previous implicit priority given by the
654 list order is no longer honoured.
656 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
659 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
660 by defining SUPPORT_SOCKS.
662 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
663 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
664 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
665 variables are renamed to proxy_{local,external}_{address,port}.
667 JH/27 Move Internationalisation support from Experimental to mainline, enabled
668 for a build by defining SUPPORT_I18N
670 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
671 of the query string, and make ${quote_redis:} do that quoting.
673 JH/29 Move Events support from Experimental to mainline, enabled by default
674 and removable for a build by defining DISABLE_EVENT.
676 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
678 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
679 cached by the daemon.
681 JH/32 Move Redis support from Experimental to mainline, enabled for a build
682 by defining LOOKUP_REDIS. The libhiredis library is required.
684 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
685 keys are given for lookup.
687 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
688 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
689 only supported when built with TLS support. The PolarSSL SHA routines
690 are still used when the TLS library is too old for convenient support.
692 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
693 openssl_options), for security. OpenSSL forces this from version 1.1.0
694 server-side so match that on older versions.
696 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
697 allocation for $value could be released as the expansion processing
698 concluded, but leaving the global pointer active for it.
700 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
701 and to use the domains and local_parts ACL conditions.
703 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
704 incorrectly not doubled on cutthrough transmission, hence seen as a
705 body-termination at the receiving system - resulting in truncated mails.
706 Commonly the sender saw a TCP-level error, and retransmitted the message
707 via the normal store-and-forward channel. This could result in duplicates
708 received - but deduplicating mailstores were liable to retain only the
709 initial truncated version.
711 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
713 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
715 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
716 we're in there, support oversigning also; bug 1309.
718 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
720 HS/04 Add support for keep_environment and add_environment options.
722 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
723 either intentional arithmetic overflow during PRNG, or testing config-
726 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
727 delivery resulted in actual delivery. Cancel cutthrough before DATA
730 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
731 rejects a recipient: pass the reject to the originator.
733 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
734 Many were false-positives and ignorable, but it's worth fixing the
737 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
738 for the new environment-manipulation done at startup. Move the routines
739 from being local to tls.c to being global via the os.c file.
741 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
742 an extract embedded as result-arg for a map, the first arg for extract
743 is unavailable so we cannot tell if this is a numbered or keyed
744 extraction. Accept either.
750 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
753 JH/02 The smtp transport option "multi_domain" is now expanded.
755 JH/03 The smtp transport now requests PRDR by default, if the server offers
758 JH/04 Certificate name checking on server certificates, when exim is a client,
759 is now done by default. The transport option tls_verify_cert_hostnames
760 can be used to disable this per-host. The build option
761 EXPERIMENTAL_CERTNAMES is withdrawn.
763 JH/05 The value of the tls_verify_certificates smtp transport and main options
764 default to the word "system" to access the system default CA bundle.
765 For GnuTLS, only version 3.0.20 or later.
767 JH/06 Verification of the server certificate for a TLS connection is now tried
768 (but not required) by default. The verification status is now logged by
769 default, for both outbound TLS and client-certificate supplying inbound
772 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
775 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
776 Status Notification (bounce) messages are now MIME format per RFC 3464.
777 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
778 under the control of the dsn_advertise_hosts option, and routers may
779 have a dsn_lasthop option.
781 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
782 default, modifiable by a malware= option. The list separator for
783 the options can now be changed in the usual way. Bug 68.
785 JH/10 The smtp_receive_timeout main option is now expanded before use.
787 JH/11 The incoming_interface log option now also enables logging of the
788 local interface on delivery outgoing connections.
790 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
791 if the interface and destination host and port all match.
793 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
796 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
797 Patch from Andrew Lewis.
799 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
800 now supports optional time-restrictions, weighting, and priority
801 modifiers per server. Patch originally by <rommer@active.by>.
803 JH/16 The spamd_address main option now supports a mixed list of local
804 and remote servers. Remote servers can be IPv6 addresses, and
805 specify a port-range.
807 JH/17 Bug 68: The spamd_address main option now supports an optional
808 timeout value per server.
810 JH/18 Bug 1581: Router and transport options headers_add/remove can
811 now have the list separator specified.
813 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
816 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
819 JH/21 Support for the A6 type of dns record is withdrawn.
821 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
822 rather than the verbs used.
824 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
825 from 255 to 1024 chars.
827 JH/24 Verification callouts now attempt to use TLS by default.
829 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
830 are generic router options now. The defaults didn't change.
832 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
833 Original patch from Alexander Shikoff, worked over by JH.
835 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
836 files if ZCAT_COMMAND is not executable.
838 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
840 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
842 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
843 Normally benign, it bites when the pair was led to by a CNAME;
844 modern usage is to not canonicalize the domain to a CNAME target
845 (and we were inconsistent anyway for A-only vs AAAA+A).
847 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
849 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
850 when evaluating $sender_host_dnssec.
852 JH/31 Check the HELO verification lookup for DNSSEC, adding new
853 $sender_helo_dnssec variable.
855 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
857 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
859 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
861 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
862 documented as working, but never had. Support all but $spam_report.
864 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
865 added for tls authenticator.
867 HS/03 Add perl_taintmode main config option
873 TL/01 When running the test suite, the README says that variables such as
874 no_msglog_check are global and can be placed anywhere in a specific
875 test's script, however it was observed that placement needed to be near
876 the beginning for it to behave that way. Changed the runtest perl
877 script to read through the entire script once to detect and set these
878 variables, reset to the beginning of the script, and then run through
879 the script parsing/test process like normal.
881 TL/02 The BSD's have an arc4random API. One of the functions to induce
882 adding randomness was arc4random_stir(), but it has been removed in
883 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
884 function when detected.
886 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
887 cause callback expansion.
889 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
890 syntax errors in an expansion can be treated as a string instead of
891 logging or causing an error, due to the internal use of bool_lax
892 instead of bool when processing it.
894 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
895 server certificates when making smtp deliveries.
897 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
899 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
901 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
903 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
904 Merged patch from Sebastian Wiedenroth.
906 JH/05 Fix results-pipe from transport process. Several recipients, combined
907 with certificate use, exposed issues where response data items split
908 over buffer boundaries were not parsed properly. This eventually
909 resulted in duplicates being sent. This issue only became common enough
910 to notice due to the introduction of connection certificate information,
911 the item size being so much larger. Found and fixed by Wolfgang Breyha.
913 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
914 size buffer was used, resulting in syntax errors when an expansion
917 JH/07 Add support for directories of certificates when compiled with a GnuTLS
918 version 3.3.6 or later.
920 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
921 is EXPERIMENTAL_EVENT, the main-configuration and transport options
922 both become "event_action", the variables become $event_name, $event_data
923 and $event_defer_errno. There is a new variable $verify_mode, usable in
924 routers, transports and related events. The tls:cert event is now also
925 raised for inbound connections, if the main configuration event_action
928 TL/06 In test suite, disable OCSP for old versions of openssl which contained
929 early OCSP support, but no stapling (appears to be less than 1.0.0).
931 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
932 server certificate names available under the smtp transport option
933 "tls_verify_cert_hostname" now do not permit multi-component wildcard
936 JH/10 Time-related extraction expansions from certificates now use the main
937 option "timezone" setting for output formatting, and are consistent
938 between OpenSSL and GnuTLS compilations. Bug 1541.
940 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
941 encoded parameter in the incoming message. Bug 1558.
943 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
944 include certificate info, eximon was claiming there were spoolfile
947 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
949 JH/14 Log delivery-related information more consistently, using the sequence
950 "H=<name> [<ip>]" wherever possible.
952 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
953 are problematic for Debian distribution, omit them from the release
956 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
958 JH/16 Fix string representation of time values on 64bit time_t architectures.
961 JH/17 Fix a null-indirection in certextract expansions when a nondefault
962 output list separator was used.
967 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
968 checkers that were complaining about end of non-void function with no
971 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
972 This was a regression introduced in 4.83 by another bugfix.
974 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
976 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
977 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
983 TF/01 Correctly close the server side of TLS when forking for delivery.
985 When a message was received over SMTP with TLS, Exim failed to clear up
986 the incoming connection properly after forking off the child process to
987 deliver the message. In some situations the subsequent outgoing
988 delivery connection happened to have the same fd number as the incoming
989 connection previously had. Exim would try to use TLS and fail, logging
990 a "Bad file descriptor" error.
992 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
993 utilities have not been installed.
995 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
996 temporary space as the ACL may create new global variables.
998 TL/01 LDAP support uses per connection or global context settings, depending
999 upon the detected version of the libraries at build time.
1001 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1002 to extract and use the src ip:port in logging and expansions as if it
1003 were a direct connection from the outside internet. PPv2 support was
1004 updated based on HAProxy spec change in May 2014.
1006 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1008 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1009 Properly escape header and check for NULL return.
1011 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1014 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1016 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1017 characters in header names, implemented as a verify condition.
1018 Contributed by Michael Fischer v. Mollard.
1020 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1021 results permerror and temperror. Previous values are deprecated but
1022 still accepted. In a future release, err_perm and err_temp will be
1023 completely removed, which will be a backward incompatibility if the
1024 ACL tests for either of these two old results. Patch contributed by
1025 user bes-internal on the mailing list.
1027 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1029 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1030 selectors, in both main and reject logs.
1032 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1035 JH/07 Add malware type "sock" for talking to simple daemon.
1037 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1039 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1040 routers/transports under cutthrough routing.
1042 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1043 numbers. Touch up "bool" conditional to keep the same definition.
1045 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1047 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1049 JH/12 Expand items in router/transport headers_add or headers_remove lists
1050 individually rather than the list as a whole. Bug 1452.
1052 Required for reasonable handling of multiple headers_ options when
1053 they may be empty; requires that headers_remove items with embedded
1054 colons must have them doubled (or the list-separator changed).
1056 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1057 view the policy declared in the DMARC record. Currently, $dmarc_status
1058 is a combined value of both the record presence and the result of the
1061 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1063 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1064 dnslookup router and the smtp transport (applying to the forward
1067 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1068 of ldap servers used for a specific lookup. Patch provided by Heiko
1071 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1072 New variable $lookup_dnssec_authenticated for observability.
1074 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1075 Patch submitted by Lars Timman.
1077 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1079 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1080 Requires trusted mode and valid format message id, aborts otherwise.
1081 Patch contributed by Heiko Schlichting.
1083 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1084 certextract with support for various fields. Bug 1358.
1086 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1087 is requested by default, modifiable by smtp transport option
1090 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1091 operate on certificate variables to give certificate fingerprints
1092 Also new ${sha256:cert_variable}.
1094 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1096 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1097 Christian Aistleitner.
1099 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1101 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1102 file. Patch from Wolfgang Breyha.
1104 JH/25 Expand the coverage of the delivery $host and $host_address to
1105 client authenticators run in verify callout. Bug 1476.
1107 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1108 align with daemon_smtp_ports. Bug 72.
1110 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1111 support and error reporting did not work properly.
1113 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1114 and is readable. Patch from Andrew Colin Kissa.
1116 TL/14 Enhance documentation of ${run expansion and how it parses the
1117 commandline after expansion, particularly in the case when an
1118 unquoted variable expansion results in an empty value.
1120 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1122 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1123 by renaming away from it. Addresses GH issue 15, reported by
1126 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1128 TL/15 SECURITY: prevent double expansion in math comparison functions
1129 (can expand unsanitized data). Not remotely exploitable.
1136 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1138 PP/02 Make -n do something, by making it not do something.
1139 When combined with -bP, the name of an option is not output.
1141 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1144 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1145 $sender_host_name and config options to manage this, and basic check
1148 PP/05 DSCP support for outbound connections and control modifier for inbound.
1150 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1151 (Only plugin which currently uses this is kerberos4, which nobody should
1152 be using, but we should make it available and other future plugins might
1153 conceivably use it, even though it would break NAT; stuff *should* be
1154 using channel bindings instead).
1156 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1157 name; added for Sendmail compatibility; requires admin caller.
1158 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1159 just ignore it); requires trusted caller.
1160 Also parse but ignore: -Ac -Am -X<logfile>
1163 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1165 TL/02 Add +smtp_confirmation as a default logging option.
1167 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1168 Patch by Magnus Holmgren from 2007-02-20.
1170 TL/04 Bugzilla 1281 - Spec typo.
1171 Bugzilla 1283 - Spec typo.
1172 Bugzilla 1290 - Spec grammar fixes.
1174 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1176 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1178 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1179 mailing list by by Dmitry Isaikin.
1181 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1183 JH/02 Support "G" suffix to numbers in ${if comparisons.
1185 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1187 NM/01 Bugzilla 1197 - Spec typo
1188 Bugzilla 1196 - Spec examples corrections
1190 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1192 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1193 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1196 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1197 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1199 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1200 "acl {{name}{arg}...}", and optional args on acl condition
1203 JH/05 Permit multiple router/transport headers_add/remove lines.
1205 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1207 JH/07 Avoid using a waiting database for a single-message-only transport.
1208 Performance patch from Paul Fisher. Bugzilla 1262.
1210 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1213 JH/09 Add $headers_added variable, with content from use of ACL modifier
1214 add_header (but not yet added to the message). Bugzilla 199.
1216 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1217 Pulled from Bugzilla 817 by Wolfgang Breyha.
1219 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1221 (nb: this is the same fix as in Exim 4.80.1)
1223 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1226 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1229 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1231 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1232 advertise SMTP AUTH mechanism to us, instead of a generic
1233 protocol violation error. Also, make Exim more robust to bad
1234 data from the Dovecot auth socket.
1236 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1238 When a queue runner is handling a message, Exim first routes the
1239 recipient addresses, during which it prunes them based on the retry
1240 hints database. After that it attempts to deliver the message to
1241 any remaining recipients. It then updates the hints database using
1244 So if a recipient address works intermittently, it can get repeatedly
1245 deferred at routing time. The retry hints record remains fresh so the
1246 address never reaches the final cutoff time.
1248 This is a fairly common occurrence when a user is bumping up against
1249 their storage quota. Exim had some logic in its local delivery code
1250 to deal with this. However it did not apply to per-recipient defers
1251 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1253 This change adds a proper retry rule check during routing so that the
1254 final cutoff time is checked against the message's age. We only do
1255 this check if there is an address retry record and there is not a
1256 domain retry record; this implies that previous attempts to handle
1257 the address had the retry_use_local_parts option turned on. We use
1258 this as an approximation for the destination being like a local
1259 delivery, as in LMTP.
1261 I suspect this new check makes the old local delivery cutoff check
1262 redundant, but I have not verified this so I left the code in place.
1264 TF/02 Correct gecos expansion when From: is a prefix of the username.
1266 Test 0254 submits a message to Exim with the header
1270 When I ran the test suite under the user fanf2, Exim expanded
1271 the header to contain my full name, whereas it should have added
1272 a Resent-Sender: header. It erroneously treats any prefix of the
1273 username as equal to the username.
1275 This change corrects that bug.
1277 GF/01 DCC debug and logging tidyup
1278 Error conditions log to paniclog rather than rejectlog.
1279 Debug lines prefixed by "DCC: " to remove any ambiguity.
1281 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1283 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1284 Bug spotted by Jeremy Harris; was flawed since initial commit.
1285 Would have resulted in OCSP responses post-SNI triggering an Exim
1286 NULL dereference and crash.
1288 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1290 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1291 Bug detection, analysis and fix by Samuel Thibault.
1292 Bugzilla 1331, Debian bug #698092.
1294 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1296 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1297 Server implementation by Todd Lyons, client by JH.
1298 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1299 config variable "prdr_enable" controls whether the server
1300 advertises the facility. If the client requests PRDR a new
1301 acl_data_smtp_prdr ACL is called once for each recipient, after
1302 the body content is received and before the acl_smtp_data ACL.
1303 The client is controlled by both of: a hosts_try_prdr option
1304 on the smtp transport, and the server advertisement.
1305 Default client logging of deliveries and rejections involving
1306 PRDR are flagged with the string "PRDR".
1308 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1309 fclose(). Diagnosis by Todd Lyons.
1311 PP/17 Update configure.default to handle IPv6 localhost better.
1312 Patch by Alain Williams (plus minor tweaks).
1315 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1316 This is now consistent with GnuTLS, and is now documented: the
1317 previous undocumented portable approach to treating the option as
1318 unset was to force an expansion failure. That still works, and
1319 an empty string is now equivalent.
1321 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1322 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1323 not performing validation itself.
1325 PP/20 Added force_command boolean option to pipe transport.
1326 Patch from Nick Koston, of cPanel Inc.
1328 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1331 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1333 PP/21 Fix eximon continuous updating with timestamped log-files.
1334 Broken in a format-string cleanup in 4.80, missed when I repaired the
1335 other false fix of the same issue.
1336 Report and fix from Heiko Schlichting.
1339 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1340 Report from Prashanth Katuri.
1342 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1343 It's SecureTransport, so affects any MacOS clients which use the
1344 system-integrated TLS libraries, including email clients.
1346 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1347 using a MIME ACL for non-SMTP local injection.
1348 Report and assistance in diagnosis by Warren Baker.
1350 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1352 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1354 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1355 last id that failed so it may be referenced in subsequent ACL's.
1357 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1360 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1361 ldap library initialization, allowing self-signed CA's to be
1362 used. Also properly sets require_cert option later in code by
1363 using NULL (global ldap config) instead of ldap handle (per
1364 session). Bug diagnosis and testing by alxgomz.
1366 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1367 the src/util/ subdirectory.
1369 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1370 renamed to Transport Post Delivery Action by Jeremy Harris, as
1373 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1374 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1375 redis_servers = needs to be configured which will be used by the redis
1376 lookup. Patch from Warren Baker, of The Packet Hub.
1378 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1380 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1381 hostname or reverse DNS when processing a host list. Used suggestions
1382 from multiple comments on this bug.
1384 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1386 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1387 Missed a few lines, added it to make the runtest require no keyboard
1390 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1391 contains upper case chars. Make router use caseful_local_part.
1393 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1394 support when GnuTLS has been built with p11-kit.
1400 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1402 This, or similar/improved, will also be change PP/11 of 4.82.
1408 PP/01 Handle short writes when writing local log-files.
1409 In practice, only affects FreeBSD (8 onwards).
1410 Bugzilla 1053, with thanks to Dmitry Isaikin.
1412 NM/01 Bugzilla 949 - Documentation tweak
1414 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1417 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1419 PP/02 Implemented gsasl authenticator.
1421 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1423 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1424 `pkg-config foo` for cflags/libs.
1426 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1427 with rest of GSASL and with heimdal_gssapi.
1429 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1430 `pkg-config foo` for cflags/libs for the TLS implementation.
1432 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1433 properties get this fed in as external SSF. A number of robustness
1434 and debugging improvements to the cyrus_sasl authenticator.
1436 PP/08 cyrus_sasl server now expands the server_realm option.
1438 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1439 Patch by Jeremy Harris.
1441 PP/10 Added dbmjz lookup type.
1443 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1445 PP/12 MAIL args handles TAB as well as SP, for better interop with
1446 non-compliant senders.
1447 Analysis and variant patch by Todd Lyons.
1449 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1450 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1451 Patch from Dirk Mueller <dmueller@suse.com>
1453 PP/13 tls_peerdn now print-escaped for spool files.
1454 Observed some $tls_peerdn in wild which contained \n, which resulted
1455 in spool file corruption.
1457 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1458 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1459 or write after TLS renegotiation, which otherwise led to messages
1462 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1463 as a tracking header (ie: a signed header comes before the signature).
1464 Patch from Wolfgang Breyha.
1466 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1467 comma-sep list; embedded commas doubled.
1469 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1471 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1473 Report and patch from Dmitry Banschikov.
1475 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1476 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1477 failures. We appear to now support TLS1.1+ with Exim.
1479 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1480 lets Exim select keys and certificates based upon TLS SNI from client.
1481 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1482 before an outbound SMTP session. New log_selector, +tls_sni.
1484 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1485 NULL dereference. Report and patch from Alun Jones.
1487 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1488 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1489 Not seeing resolver debug output on NetBSD, but suspect this is a
1490 resolver implementation change.
1492 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1493 Left warnings. Added "eximon gdb" invocation mode.
1495 PP/21 Defaulting "accept_8bitmime" to true, not false.
1497 PP/22 Added -bw for inetd wait mode support.
1499 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1500 locate the relevant includes and libraries. Made this the default.
1502 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1503 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1505 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1506 This may cause build issues on older platforms.
1508 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1509 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1510 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1511 Added SNI support via GnuTLS too.
1512 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1514 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1516 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1517 Applied second patch from Janne, implementing suggestion to default
1518 multiple-strings-in-record handling to match SPF spec.
1520 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1522 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1523 read-only, out of scope).
1524 Patch from Wolfgang Breyha, report from Stuart Northfield.
1526 PP/29 Fix three issues highlighted by clang analyser static analysis.
1527 Only crash-plausible issue would require the Cambridge-specific
1528 iplookup router and a misconfiguration.
1529 Report from Marcin Mirosław.
1531 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1533 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1534 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1535 As part of this, removing so much warning spew let me fix some minor
1536 real issues in debug logging.
1538 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1539 assignment on my part. Fixed.
1541 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1542 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1543 Janne Snabb (who went above and beyond: thank you).
1545 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1546 string otherwise requires a connection and a bunch more work and it's
1547 relatively easy to get wrong. Should also expose TLS library linkage
1550 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1551 64-bit ${eval} (JH/03).
1553 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1554 GNU libc to support some of the 64-bit stuff, should not lead to
1555 conflicts. Defined before os.h is pulled in, so if a given platform
1556 needs to override this, it can.
1558 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1559 protection layer was required, which is not implemented.
1560 Bugzilla 1254, patch from Wolfgang Breyha.
1562 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1563 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1564 tls_dhparam take prime identifiers. Also unbreak combination of
1565 OpenSSL+DH_params+TLSSNI.
1567 PP/39 Disable SSLv2 by default in OpenSSL support.
1573 PP/01 Solaris build fix for Oracle's LDAP libraries.
1574 Bugzilla 1109, patch from Stephen Usher.
1576 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1578 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1581 TF/02 Fix a couple more cases where we did not log the error message
1582 when unlink() failed. See also change 4.74-TF/03.
1584 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1585 lock up or crash if it happened to be inside a call to libc when it
1586 got a SIGUSR1 from exiwhat.
1588 The SIGUSR1 handler appends the current process status to the process
1589 log which is later printed by exiwhat. It used to use the general
1590 purpose logging code to do this, but several functions it calls are
1591 not safe for signals.
1593 The new output code in the SIGUSR1 handler is specific to the process
1594 log, and simple enough that it's easy to inspect for signal safety.
1595 Removing some special cases also simplifies the general logging code.
1596 Removing the spurious timestamps from the process log simplifies
1599 TF/04 Improved ratelimit ACL condition.
1601 The /noupdate option has been deprecated in favour of /readonly which
1602 has clearer semantics. The /leaky, /strict, and /readonly update modes
1603 are mutually exclusive. The update mode is no longer included in the
1604 database key; it just determines when the database is updated. (This
1605 means that when you upgrade Exim will forget old rate measurements.)
1607 Exim now checks that the per_* options are used with an update mode that
1608 makes sense for the current ACL. For example, when Exim is processing a
1609 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1610 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1611 must specify per_mail/readonly. If you omit the update mode it defaults to
1612 /leaky where that makes sense (as before) or /readonly where required.
1614 The /noupdate option is now undocumented but still supported for
1615 backwards compatibility. It is equivalent to /readonly except that in
1616 ACLs where /readonly is required you may specify /leaky/noupdate or
1617 /strict/noupdate which are treated the same as /readonly.
1619 A useful new feature is the /count= option. This is a generalization
1620 of the per_byte option, so that you can measure the throughput of other
1621 aggregate values. For example, the per_byte option is now equivalent
1622 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1624 The per_rcpt option has been generalized using the /count= mechanism
1625 (though it's more complicated than the per_byte equivalence). When it is
1626 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1627 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1628 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1629 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1630 Note that using per_rcpt with a non-readonly update mode in more than
1631 one ACL will cause the recipients to be double-counted. (The per_mail
1632 and per_byte options don't have this problem.)
1634 The handling of very low rates has changed slightly. If the computed rate
1635 is less than the event's count (usually one) then this event is the first
1636 after a long gap. In this case the rate is set to the same as this event's
1637 count, so that the first message of a spam run is counted properly.
1639 The major new feature is a mechanism for counting the rate of unique
1640 events. The new per_addr option counts the number of different
1641 recipients that someone has sent messages to in the last time period. It
1642 behaves like per_rcpt if all the recipient addresses are different, but
1643 duplicate recipient addresses do not increase the measured rate. Like
1644 the /count= option this is a general mechanism, so the per_addr option
1645 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1646 example, measure the rate that a client uses different sender addresses
1647 with the options per_mail/unique=$sender_address. There are further
1648 details in the main documentation.
1650 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1652 TF/06 Removed a few PCRE remnants.
1654 TF/07 Automatically extract Exim's version number from tags in the git
1655 repository when doing development or release builds.
1657 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1658 Bugzilla 879. Patch from Paul Fisher.
1660 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1661 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1664 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1666 PP/05 Variable $av_failed, true if the AV scanner deferred.
1667 Bugzilla 1078. Patch from John Horne.
1669 PP/06 Stop make process more reliably on build failure.
1670 Bugzilla 1087. Patch from Heiko Schlittermann.
1672 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1673 Bugzilla 1089. Patch from Heiko Schlittermann.
1675 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1676 Bugzilla 1131. Patch from Holger Weiß.
1678 PP/09 Handle IPv6 addresses with SPF.
1679 Bugzilla 860. Patch from Wolfgang Breyha.
1681 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1683 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1686 PP/11 match_* no longer expand right-hand-side by default.
1687 New compile-time build option, EXPAND_LISTMATCH_RHS.
1688 New expansion conditions, "inlist", "inlisti".
1690 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1692 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1694 PP/14 fix log_write() format string regression from TF/03.
1695 Bugzilla 1152. Patch from Dmitry Isaikin.
1701 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1703 PP/02 Harmonised TLS library version reporting; only show if debugging.
1704 Layout now matches that introduced for other libraries in 4.74 PP/03.
1706 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1708 PP/04 New "dns_use_edns0" global option.
1710 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1713 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1714 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1716 TK/01 Updated PolarSSL code to 0.14.2.
1717 Bugzilla 1097. Patch from Andreas Metzler.
1719 PP/07 Catch divide-by-zero in ${eval:...}.
1720 Fixes bugzilla 1102.
1722 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1725 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1726 format-string attack -- SECURITY: remote arbitrary code execution.
1728 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1729 time unintentionally subject to list matching rules, letting the header
1730 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1731 arbitrary string expansion). This allowed for information disclosure.
1733 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1734 INT_MIN/-1 -- value coerced to INT_MAX.
1740 NM/01 Workaround for PCRE version dependency in version reporting
1743 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1744 This fixes portability to compilers other than gcc, notably
1745 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1747 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1748 makefiles for portability to HP-UX and POSIX correctness.
1750 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1751 Also via indented variable definition in the Makefile.
1752 (Debugging by Oliver Heesakkers).
1754 PP/02 Restore caching of spamd results with expanded spamd_address.
1755 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1757 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1758 Improves build reliability. Fix from: Frank Elsner
1760 NM/02 Fix wide character breakage in the rfc2047 coding
1761 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1763 NM/03 Allow underscore in dnslist lookups
1764 Fixes bug 1026. Patch from Graeme Fowler
1766 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1767 Code patches from Adam Ciarcinski of NetBSD.
1769 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1772 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1773 is logged, to avoid truncation. Patch from John Horne.
1775 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1776 Patch from Jakob Hirsch.
1778 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1779 SQL string expansion failure details.
1780 Patch from Andrey Oktyabrski.
1782 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1783 Patch from Simon Arlott.
1785 PP/09 New lookups functionality failed to compile on old gcc which rejects
1786 extern declarations in function scope.
1787 Patch from Oliver Fleischmann
1789 PP/10 Use sig_atomic_t for flags set from signal handlers.
1790 Check getgroups() return and improve debugging.
1791 Fixed developed for diagnosis in bug 927 (which turned out to be
1794 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1795 Patch from Mark Zealey.
1797 PP/12 Bugzilla 1056: Improved spamd server selection.
1798 Patch from Mark Zealey.
1800 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1801 Based on patch from Heiko Schlittermann.
1803 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1804 Patch from Uwe Doering, sign-off by Michael Haardt.
1806 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1807 variable declaration deep within a block. Bug and patch from
1810 PP/15 lookups-Makefile IRIX compatibility coercion.
1812 PP/16 Make DISABLE_DKIM build knob functional.
1814 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1815 Patch by Simon Arlott
1817 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1818 variable argument macros. Our copy now differs from upstream.
1824 TF/01 Failure to get a lock on a hints database can have serious
1825 consequences so log it to the panic log.
1827 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1828 controlled using the smtp_confirmation log selector.
1830 TF/03 Include the error message when we fail to unlink a spool file.
1832 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1833 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1834 for maintaining out-of-tree patches for some time.
1836 PP/01 Bugzilla 139: Documentation and portability issues.
1837 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1838 Handle per-OS dynamic-module compilation flags.
1840 PP/02 Let /dev/null have normal permissions.
1841 The 4.73 fixes were a little too stringent and complained about the
1842 permissions on /dev/null. Exempt it from some checks.
1843 Reported by Andreas M. Kirchwitz.
1845 PP/03 Report version information for many libraries, including
1846 Exim version information for dynamically loaded libraries. Created
1847 version.h, now support a version extension string for distributors
1848 who patch heavily. Dynamic module ABI change.
1850 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1851 privilege escalation vulnerability whereby the Exim run-time user
1852 can cause root to append content of the attacker's choosing to
1855 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1858 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1859 If dropping privileges for untrusted macros, we disabled normal logging
1860 on the basis that it would fail; for the Exim run-time user, this is not
1861 the case, and it resulted in successful deliveries going unlogged.
1862 Fixed. Reported by Andreas Metzler.
1868 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1869 only prepend for the Resent-* case. Fixes regression introduced in
1870 Exim 4.70 by NM/22 for Bugzilla 607.
1872 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1873 increasing numbers of administrators be bitten by this.
1875 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1877 PP/03 Bugzilla 994: added openssl_options main configuration option.
1879 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1881 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1883 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1885 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1886 without a peer certificate, leading to a segfault because of an
1887 assumption that peers always have certificates. Be a little more
1888 paranoid. Problem reported by Martin Tscholak.
1890 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1891 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1892 NB: ClamAV planning to remove STREAM in "middle of 2010".
1893 CL also introduces -bmalware, various -d+acl logging additions and
1894 more caution in buffer sizes.
1896 PP/09 Implemented reverse_ip expansion operator.
1898 PP/10 Bugzilla 937: provide a "debug" ACL control.
1900 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1902 PP/12 Bugzilla 973: Implement --version.
1904 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1906 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1908 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1910 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1911 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1912 ignore trailing whitespace.
1914 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1916 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1919 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1920 Notification from Dr Andrew Aitchison.
1922 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1923 ExtendedDetectionInfo response format.
1924 Notification from John Horne.
1926 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1929 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1930 XSL and documented dependency on system catalogs, with examples of how
1933 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1936 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1937 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1938 configuration file which is writeable by the Exim user or group.
1940 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1941 of configuration files to cover files specified with the -C option if
1942 they are going to be used with root privileges, not just the default
1945 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1946 option (effectively making it always true).
1948 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1949 files to be used while preserving root privileges.
1951 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1952 that rogue child processes cannot use them.
1954 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1955 run-time user, instead of root.
1957 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1958 Exim run-time user without dropping privileges.
1960 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1961 result string, instead of calling string_vformat() twice with the same
1964 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1965 for other users. Others should always drop root privileges if they use
1966 -C on the command line, even for a whitelisted configure file.
1968 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1970 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1976 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1977 $data_path, and $header_path variables; fixed documentation bugs and
1980 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1981 exipick to access non-standard spools, including the "frozen" queue
1984 NM/01 Bugzilla 965: Support mysql stored procedures.
1985 Patch from Alain Williams
1987 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1989 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1990 Patch from Andreas Metzler
1992 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1993 Patch from Kirill Miazine
1995 NM/05 Bugzilla 671: Added umask to procmail example.
1997 JJ/03 installed exipick 20100323.0, fixing doc bug
1999 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2000 directory. Notification and patch from Dan Rosenberg.
2002 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2004 TK/02 Improve log output when DKIM signing operation fails.
2006 MH/01 Treat the transport option dkim_domain as a colon separated
2007 list, not as a single string, and sign the message with each element,
2008 omitting multiple occurences of the same signer.
2010 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2011 Bugzilla 985, 986. Patch by Simon Arlott
2013 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2014 Patch by Simon Arlott
2016 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2017 MBX locking. Notification from Dan Rosenberg.
2023 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2025 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2027 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2029 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2031 NM/04 Bugzilla 915: Flag broken perl installation during build.
2037 TK/01 Added patch by Johannes Berg that expands the main option
2038 "spamd_address" if it starts with a dollar sign.
2040 TK/02 Write list of recipients to X-Envelope-Sender header when building
2041 the mbox-format spool file for content scanning (suggested by Jakob
2044 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2045 (http://www.dcc-servers.net/) support via dccifd. Activated by
2046 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2048 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2049 by Mark Daniel Reidel <mr@df.eu>.
2051 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2052 When building exim an external PCRE library is now needed -
2053 PCRE is a system library on the majority of modern systems.
2054 See entry on PCRE_LIBS in EDITME file.
2056 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2057 conversation. Added nologin parameter to request.
2058 Patch contributed by Kirill Miazine.
2060 TF/01 Do not log submission mode rewrites if they do not change the address.
2062 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2064 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2065 log files in place. Contributed by Roberto Lima.
2067 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2069 TF/03 Bugzilla 615: When checking the local_parts router precondition
2070 after a local_part_suffix or local_part_prefix option, Exim now
2071 does not use the address's named list lookup cache, since this
2072 contains cached lookups for the whole local part.
2074 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2075 Robert Millan. Documentation is in experimental-spec.txt.
2077 TF/04 Bugzilla 668: Fix parallel build (make -j).
2079 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2081 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2082 Patch provided by Jan Srzednicki.
2084 TF/05 Leading white space used to be stripped from $spam_report which
2085 wrecked the formatting. Now it is preserved.
2087 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2088 that they are available at delivery time.
2090 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2092 TF/08 TLS error reporting now respects the incoming_interface and
2093 incoming_port log selectors.
2095 TF/09 Produce a more useful error message if an SMTP transport's hosts
2096 setting expands to an empty string.
2098 NM/06 Bugzilla 744: EXPN did not work under TLS.
2099 Patch provided by Phil Pennock.
2101 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2102 Patch provided by Richard Godbee.
2104 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2105 acl_smtp_notquit, added index entry.
2107 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2108 Patch provided by Eugene Bujak.
2110 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2111 accept(). Patch provided by Maxim Dounin.
2113 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2114 Patch provided by Phil Pennock.
2116 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2118 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2119 Patch provided by Brad "anomie" Jorsch.
2121 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2122 Patch provided by Dean Brooks.
2124 TK/05 Add native DKIM support (does not depend on external libraries).
2126 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2127 Patch provided by Graeme Fowler.
2129 NM/16 Bugzilla 851: Documentation example syntax fix.
2131 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2133 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2136 NM/19 Bugzilla 745: TLS version reporting.
2137 Patch provided by Phil Pennock.
2139 NM/20 Bugzilla 167: bool: condition support.
2140 Patch provided by Phil Pennock.
2142 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2143 clients. Patch provided by Phil Pennock.
2145 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2146 Patch provided by Brad "anomie" Jorsch.
2148 NM/23 Bugzilla 687: Fix misparses in eximstats.
2149 Patch provided by Heiko Schlittermann.
2151 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2152 Patch provided by Heiko Schlittermann.
2154 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2155 plus update to original patch.
2157 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2159 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2160 Patch provided by David Brownlee.
2162 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2164 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2166 NM/30 Bugzilla 888: TLS documentation bugfixes.
2168 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2170 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2171 Unlike the original bugzilla I have changed all shell scripts in src tree.
2173 NM/33 Bugzilla 898: Transport filter timeout fix.
2174 Patch by Todd Rinaldo.
2176 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2177 Patch by Serge Demonchaux.
2179 NM/35 Bugzilla 39: Base64 decode bug fixes.
2180 Patch by Jakob Hirsch.
2182 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2184 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2186 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2188 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2194 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2195 ALT-N's libdkim that I have put here:
2196 http://duncanthrax.net/exim-experimental/
2198 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2199 were called 'true' and it seems that C99 defines that as a reserved
2200 keyword to be used with 'bool' variable types. That means you could
2201 not include C99-style headers which use bools without triggering
2202 build errors in sieve.c.
2204 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2205 as mailq or other aliases. Changed the --help handling significantly
2206 to do whats expected. exim_usage() emits usage/help information.
2208 SC/01 Added the -bylocaldomain option to eximstats.
2210 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2212 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2214 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2220 PH/01 Another patch from the Sieve maintainer.
2222 PH/02 When an IPv6 address is converted to a string for single-key lookup
2223 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2224 dots are used instead of colons so that keys in lsearch files need not
2225 contain colons. This was done some time before quoting was made available
2226 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2227 (notated using the quote facility) so as to distinguish them from IPv4
2228 keys. This meant that lookups for IP addresses in host lists did not work
2229 for iplsearch lookups.
2231 This has been fixed by arranging for IPv6 addresses to be expressed with
2232 colons if the lookup type is iplsearch. This is not incompatible, because
2233 previously such lookups could never work.
2235 The situation is now rather anomalous, since one *can* have colons in
2236 ordinary lsearch keys. However, making the change in all cases is
2237 incompatible and would probably break a number of configurations.
2239 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2242 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2243 conversion specification without a maximum field width, thereby enabling
2244 a rogue spamd server to cause a buffer overflow. While nobody in their
2245 right mind would setup Exim to query an untrusted spamd server, an
2246 attacker that gains access to a server running spamd could potentially
2247 exploit this vulnerability to run arbitrary code as the Exim user.
2249 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2250 $primary_hostname instead of what libspf2 thinks the hosts name is.
2252 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2253 a directory entry by the name of the lookup key. Previously, if a
2254 symlink pointed to a non-existing file or a file in a directory that
2255 Exim lacked permissions to read, a lookup for a key matching that
2256 symlink would fail. Now it is enough that a matching directory entry
2257 exists, symlink or not. (Bugzilla 503.)
2259 PH/03 The body_linecount and body_zerocount variables are now exported in the
2262 PH/04 Added the $dnslist_matched variable.
2264 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2265 This means they are set thereafter only if the connection becomes
2268 PH/06 Added the client_condition to authenticators so that some can be skipped
2269 by clients under certain conditions.
2271 PH/07 The error message for a badly-placed control=no_multiline_responses left
2272 "_responses" off the end of the name.
2274 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2276 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2277 (without spaces) instead of just copying the configuration text.
2279 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2281 PH/11 Added $max_received_linelength.
2283 PH/12 Added +ignore_defer and +include_defer to host lists.
2285 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2286 way in which PCRE > 7.0 is built.
2288 PH/14 Implemented queue_only_load_latch.
2290 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2291 MAIL command. The effect was to mangle the value on 64-bit systems.
2293 PH/16 Another patch from the Sieve maintainer.
2295 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2297 PH/18 If a system quota error occurred while trying to create the file for
2298 a maildir delivery, the message "Mailbox is full" was not appended to the
2299 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2300 only to a quota excession during the actual writing of the file.
2302 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2303 characters?) which causes problems in log lines. The DN values are now
2304 passed through string_printing() before being added to log lines.
2306 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2307 and InterBase are left for another time.)
2309 PH/21 Added message_body_newlines option.
2311 PH/22 Guard against possible overflow in moan_check_errorcopy().
2313 PH/23 POSIX allows open() to be a macro; guard against that.
2315 PH/24 If the recipient of an error message contained an @ in the local part
2316 (suitably quoted, of course), incorrect values were put in $domain and
2317 $local_part during the evaluation of errors_copy.
2323 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2324 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2327 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2328 issue a MAIL command.
2330 PH/02 In an ACL statement such as
2332 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2334 if a client was not listed at all, or was listed with a value other than
2335 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2336 the condition was not true (as it should be), so access was not denied.
2337 The bug was that the ! inversion was incorrectly passed on to the second
2338 item. This has been fixed.
2340 PH/03 Added additional dnslists conditions == and =& which are different from
2341 = and & when the dns lookup returns more than one IP address.
2343 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2344 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2346 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2347 FSYNC, which compiles an option called disable_fsync that allows for
2348 bypassing fsync(). The documentation is heavily laced with warnings.
2350 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2352 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2353 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2354 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2355 including adding "make clean"; (3) Added -fPIC when compiling the test
2356 dynamically loaded module, to get rid of a warning.
2358 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2359 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2360 = 0s. The bug is otherwise harmless.
2362 PH/07 There was a bug in the dovecot authenticator such that the value of
2363 $auth1 could be overwritten, and so not correctly preserved, after a
2364 successful authentication. This usually meant that the value preserved by
2365 the server_setid option was incorrect.
2367 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2369 PH/09 Installed PCRE release 7.0.
2371 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2372 run for batched SMTP input. It is now run at the start of every message
2373 in the batch. While fixing this I discovered that the process information
2374 (output by running exiwhat) was not always getting set for -bs and -bS
2375 input. This is fixed, and it now also says "batched" for BSMTP.
2377 PH/11 Added control=no_pipelining.
2379 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2380 patch, slightly modified), and move the expansion of helo_data till after
2381 the connection is made in the smtp transport (so it can use these
2384 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2386 PH/14 Added log_selector = +pid.
2388 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2390 PH/16 Add ${if forany and ${if forall.
2392 PH/17 Added dsn_from option to vary the From: line in DSNs.
2394 PH/18 Flush SMTP output before performing a callout, unless control =
2395 no_callout_flush is set.
2397 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2398 was true (the default) a successful delivery failed to delete the retry
2399 item, thus causing premature timeout of the address. The bug is now
2402 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2404 PH/21 Long custom messages for fakedefer and fakereject are now split up
2405 into multiline responses in the same way that messages for "deny" and
2406 other ACL rejections are.
2408 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2409 with slight modification.
2411 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2412 draft, changing the syntax and factoring some duplicate code".
2414 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2415 for deliveries of the second and subsequent messages over the same SMTP
2418 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2419 ${reduce, with only minor "tidies".
2421 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2423 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2424 expansion side effects.
2426 PH/27 When a message times out after an over-quota error from an Exim-imposed
2427 quota, the bounce message says "mailbox is full". This message was not
2428 being given when it was a system quota that was exceeded. It now should
2431 MH/03 Made $recipients available in local_scan(). local_scan() already has
2432 better access to the recipient list through recipients_list[], but
2433 $recipients can be useful in postmaster-provided expansion strings.
2435 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2436 in the case of a MAIL command with additional options following the
2437 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2438 were accidentally chopped off.
2440 PH/29 SMTP synchronization checks are implemented when a command is read -
2441 there is a check that no more input is waiting when there shouldn't be
2442 any. However, for some commands, a delay in an ACL can mean that it is
2443 some time before the response is written. In this time, more input might
2444 arrive, invalidly. So now there are extra checks after an ACL has run for
2445 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2446 pipelining has not been advertised.
2448 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2450 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2451 returned in the SMTP response when smtp_return_error_details was false.
2452 This has been fixed.
2454 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2455 instead of the C I/O that was originally supplied, because problems were
2456 reported on Solaris.
2458 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2459 Exim which did not show up earlier: it was assuming that a call to
2460 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2461 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2462 was a macro that became an assignment, so it seemed to work. This has
2463 changed to a proper function call with a void return, hence the compile
2464 error. Exim's code has been fixed.
2466 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2469 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2471 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2473 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2474 because it turns out to be too big or there is a timeout) there is no
2475 "Completed" line in the log. When some messages of this type were
2476 selected by exigrep, they were listed as "not completed". Others were
2477 picked up by some special patterns. I have improved the selection
2478 criteria to be more general.
2480 PH/38 The host_find_failed option in the manualroute router can now be set
2481 to "ignore", to completely ignore a host whose IP address cannot be
2482 found. If all hosts are ignored, the behaviour is controlled by the new
2483 host_all_ignored option.
2485 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2486 homing or because of multiple MX records with /mx) generated more than
2487 one IP address, and the following item turned out to be the local host,
2488 all the secondary addresses of the first item were incorrectly removed
2489 from the list, along with the local host and any following hosts (which
2490 is what is supposed to happen).
2492 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2493 whoever called Exim into the -H file. In the case of the daemon it was
2494 behaving confusingly. When first started, it used values for whoever
2495 started the daemon, but after a SIGHUP it used the Exim user (because it
2496 calls itself on a restart). I have changed the code so that it now always
2499 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2500 message are rejected with the same error (e.g. no authentication or bad
2501 sender address), and a DATA command is nevertheless sent (as can happen
2502 with PIPELINING or a stupid MUA), the error message that was given to the
2503 RCPT commands is included in the rejection of the DATA command. This is
2504 intended to be helpful for MUAs that show only the final error to their
2507 PH/42 Another patch from the Sieve maintainer.
2509 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2510 Eximstats - Fixed some broken HTML links and added missing column headers
2512 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2513 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2515 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2517 PH/43 Yet another patch from the Sieve maintainer.
2519 PH/44 I found a way to check for a TCP/IP connection going away before sending
2520 the response to the final '.' that terminates a message, but only in the
2521 case where the client has not sent further data following the '.'
2522 (unfortunately, this is allowed). However, in many cases there won't be
2523 any further data because there won't be any more messages to send. A call
2524 to select() can be used: if it shows that the input is "ready", there is
2525 either input waiting, or the socket has been closed. An attempt to read
2526 the next input character can distinguish the two cases. Previously, Exim
2527 would have sent an OK response which the client would never have see.
2528 This could lead to message repetition. This fix should cure that, at
2529 least in a lot of common cases.
2531 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2532 advertised in response to EHLO.
2538 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2539 fixed by 4.65/MH/01 (is this a record?) are fixed:
2541 (i) An empty string was always treated as zero by the numeric comparison
2542 operators. This behaviour has been restored.
2544 (ii) It is documented that the numeric comparison operators always treat
2545 their arguments as decimal numbers. This was broken in that numbers
2546 starting with 0 were being interpreted as octal.
2548 While fixing these problems I realized that there was another issue that
2549 hadn't been noticed. Values of message_size_limit (both the global option
2550 and the transport option) were treated as octal if they started with 0.
2551 The documentation was vague. These values are now always treated as
2552 decimal, and I will make that clear in the documentation.
2558 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2559 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2562 MH/01 Don't check that the operands of numeric comparison operators are
2563 integers when their expansion is in "skipping" mode (fixes bug
2564 introduced by 4.64-PH/07).
2566 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2567 child addresses, Exim now panics and dies. Previously, because the count
2568 is held in a short int, deliveries were likely to be lost. As such a
2569 large number of recipients for a single message is ridiculous
2570 (performance will be very, very poor), I have chosen to impose a limit
2571 rather than extend the field.
2577 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2578 leftover -K file (the existence of which was triggered by #402).
2579 While we were at it, introduced process PID as part of the -K
2580 filename. This should rule out race conditions when creating
2583 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2584 processing considerably. Previous code took too long for large mails,
2585 triggering a timeout which in turn triggers #401.
2587 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2588 in the DK code in transports.c. sendfile() is not really portable,
2589 hence the _LINUX specificness.
2591 TF/01 In the add_headers option to the mail command in an Exim filter,
2592 there was a bug that Exim would claim a syntax error in any
2593 header after the first one which had an odd number of characters
2596 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2597 callout verification, Exim cached a "reject" for the entire domain. This
2598 is correct for most verifications, but it is not correct for a recipient
2599 verification with use_sender or use_postmaster set, because in that case
2600 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2601 case of MAIL FROM:<> rejection from other early rejections (e.g.
2602 rejection of HELO). When verifying a recipient using a non-null MAIL
2603 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2604 Whatever the result of the callout, the value of the domain cache is
2605 left unchanged (for any other kind of callout, getting as far as trying
2606 RCPT means that the domain itself is ok).
2608 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2611 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2612 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2613 session, the daemon ignores ECONNECT errors and logs others; it now
2614 ignores EPIPE as well.
2616 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2617 (quoted-printable decoding).
2619 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2620 later the small subsequent patch to fix an introduced bug.
2622 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2624 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2626 PH/08 An error is now given if message_size_limit is specified negative.
2628 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2629 to be given (somewhat) arbitrary names.
2631 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2634 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2635 miscellaneous code fixes
2637 PH/10 Added the log_reject_target ACL modifier to specify where to log
2640 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2641 hostname. This is wrong, because it relates to the incoming message (and
2642 probably the interface on which it is arriving) and not to the outgoing
2643 callout (which could be using a different interface). This has been
2644 changed to use the value of the helo_data option from the smtp transport
2645 instead - this is what is used when a message is actually being sent. If
2646 there is no remote transport (possible with a router that sets up host
2647 addresses), $smtp_active_hostname is used.
2649 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2650 tweaks were necessary in order to get it to work (see also 21 below):
2651 (a) The code assumed that strncpy() returns a negative number on buffer
2652 overflow, which isn't the case. Replaced with Exim's string_format()
2654 (b) There were several signed/unsigned issues. I just did the minimum
2655 hacking in of casts. There is scope for a larger refactoring.
2656 (c) The code used strcasecmp() which is not a standard C function.
2657 Replaced with Exim's strcmpic() function.
2658 (d) The code set only $1; it now sets $auth1 as well.
2659 (e) A simple test gave the error "authentication client didn't specify
2660 service in request". It would seem that Dovecot has changed its
2661 interface. Fortunately there's a specification; I followed it and
2662 changed what the client sends and it appears to be working now.
2664 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2667 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2668 address A is aliased to B and C, where B exists and C does not. Without
2669 -v the output is "A verified" because verification stops after a
2670 successful redirection if more than one address is generated. However,
2671 with -v the child addresses are also verified. Exim was outputting "A
2672 failed to verify" and then showing the successful verification for C,
2673 with its parentage. It now outputs "B failed to verify", showing B's
2674 parentage before showing the successful verification of C.
2676 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2677 look up a TXT record in a specific list after matching in a combined
2680 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2681 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2682 they consult the DNS. I had assumed they would set it the way they
2683 wanted; and indeed my experiments on Linux seem to show that in some
2684 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2685 To be on the safe side, however, I have now made the interface to
2686 host_find_byname() similar to host_find_bydns(), with an argument
2687 containing the DNS resolver options. The host_find_byname() function now
2688 sets these options at its start, just as host_find_bydns() does. The smtp
2689 transport options dns_qualify_single and dns_search_parents are passed to
2690 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2691 of host_find_byname() use the default settings of RES_DEFNAMES
2692 (qualify_single) but not RES_DNSRCH (search_parents).
2694 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2695 spool_read_header() do less string testing, by means of a preliminary
2696 switch on the second character of optional "-foo" lines. (This is
2697 overdue, caused by the large number of possibilities that now exist.
2698 Originally there were few.) While I was there, I also converted the
2699 str(n)cmp tests so they don't re-test the leading "-" and the first
2700 character, in the hope this might squeeze out yet more improvement.
2702 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2703 flag allowing group syntax was set by the header_syntax check but not
2704 turned off, possible causing trouble later; (2) The flag was not being
2705 set at all for the header_verify test, causing "group"-style headers to
2706 be rejected. I have now set it in this case, and also caused header_
2707 verify to ignore an empty address taken from a group. While doing this, I
2708 came across some other cases where the code for allowing group syntax
2709 while scanning a header line wasn't quite right (mostly, not resetting
2710 the flag correctly in the right place). These bugs could have caused
2711 trouble for malformed header lines. I hope it is now all correct.
2713 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2714 with the "reply" argument non-NULL. The code, however (which originally
2715 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2716 but it didn't always do it. This confused somebody who was copying the
2717 code for some other use. I have removed all the tests.
2719 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2720 feature that was used to support insecure browsers during the U.S. crypto
2721 embargo. It requires special client support, and Exim is probably the
2722 only MTA that supported it -- and would never use it because real RSA is
2723 always available. This code has been removed, because it had the bad
2724 effect of slowing Exim down by computing (never used) parameters for the
2725 RSA_EXPORT functionality.
2727 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2728 authenticator to fail if there's a tab character in the incoming data
2729 (there should never be unless someone is messing about, as it's supposed
2730 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2731 if the connection is using TLS or if the remote IP is the same as the
2732 local IP, and the "valid-client-cert option" if a client certificate has
2735 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2736 authenticators. This can be used for authorization after authentication
2737 succeeds. (In the case of plaintext, it servers for both authentication
2740 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2741 if any retry times were supplied.
2743 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2744 connection, where there is no client IP address to check. In this
2745 situation, the verify now always succeeds.
2747 PH/25 Applied John Jetmore's -Mset patch.
2749 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2751 PH/27 In a string expansion for a processed (not raw) header when multiple
2752 headers of the same name were present, leading whitespace was being
2753 removed from all of them, but trailing whitespace was being removed only
2754 from the last one. Now trailing whitespace is removed from each header
2755 before concatenation. Completely empty headers in a concatenation (as
2756 before) are ignored.
2758 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2759 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2761 PH/29 [Removed. This was a change that I later backed out, and forgot to
2762 correct the ChangeLog entry (that I had efficiently created) before
2763 committing the later change.]
2765 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2766 address errors (4xx response to RCPT) over the same connection as other
2767 messages routed to the same hosts. Such deliveries are always "forced",
2768 so retry times are not inspected. This resulted in far too many retries
2769 for the affected addresses. The effect occurred only when there were more
2770 hosts than the hosts_max_try setting in the smtp transport when it had
2771 the 4xx errors. Those hosts that it had tried were not added to the list
2772 of hosts for which the message was waiting, so if all were tried, there
2773 was no problem. Two fixes have been applied:
2775 (i) If there are any address or message errors in an SMTP delivery, none
2776 of the hosts (tried or untried) are now added to the list of hosts
2777 for which the message is waiting, so the message should not be a
2778 candidate for sending over the same connection that was used for a
2779 successful delivery of some other message. This seems entirely
2780 reasonable: after all the message is NOT "waiting for some host".
2781 This is so "obvious" that I'm not sure why it wasn't done
2782 previously. Hope I haven't missed anything, but it can't do any
2783 harm, as the worst effect is to miss an optimization.
2785 (ii) If, despite (i), such a delivery is accidentally attempted, the
2786 routing retry time is respected, so at least it doesn't keep
2787 hammering the server.
2789 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2790 in ${readsocket because some servers need this prod.
2792 PH/32 Added some extra debug output when updating a wait-xxx database.
2794 PH/33 The hint "could be header name not terminated by colon", which has been
2795 given for certain expansion errors for a long time, was not being given
2796 for the ${if def:h_colon_omitted{... case.
2798 PH/34 The spec says: "With one important exception, whenever a domain list is
2799 being scanned, $domain contains the subject domain." There was at least
2800 one case where this was not true.
2802 PH/35 The error "getsockname() failed: connection reset by peer" was being
2803 written to the panic log as well as the main log, but it isn't really
2804 panic-worthy as it just means the connection died rather early on. I have
2805 removed the panic log writing for the ECONNRESET error when getsockname()
2808 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2809 runs only) independently of the message's sender address. This meant
2810 that, if the 4xx error was in fact related to the sender, a different
2811 message to the same recipient with a different sender could confuse
2812 things. In particular, this can happen when sending to a greylisting
2813 server, but other circumstances could also provoke similar problems.
2814 I have changed the default so that the retry time for these errors is now
2815 based a combination of the sender and recipient addresses. This change
2816 can be overridden by setting address_retry_include_sender=false in the
2819 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2820 remote server are returned as part of bounce messages. This was not
2821 happening for LMTP over a pipe (the lmtp transport), but now it is the
2822 same for both kinds of LMTP.
2824 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2825 in header lines that were in fact CNAMEs. This is no longer the case.
2827 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2828 and queue runs started by the daemon processed all messages. This has
2829 been fixed so that -R and -S can now usefully be given with -q<time>.
2831 PH/40 Import PCRE release 6.7 (fixes some bugs).
2833 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2835 PH/42 Give an error if -q is specified more than once.
2837 PH/43 Renamed the variables $interface_address and $interface_port as
2838 $received_ip_address and $received_port, to make it clear that these
2839 values apply to message reception, and not to the outgoing interface when
2840 a message is delivered. (The old names remain recognized, of course.)
2842 PH/44 There was no timeout on the connect() call when using a Unix domain
2843 socket in the ${readsocket expansion. There now is.
2845 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2846 be meaningful with "accept".
2848 SC/01 Eximstats V1.43
2849 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2851 SC/02 Eximstats V1.44
2852 Use a glob alias rather than an array ref in the generated
2853 parser. This improves both readability and performance.
2855 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2856 Collect SpamAssassin and rejection statistics.
2857 Don't display local sender or destination tables unless
2858 there is data to show.
2859 Added average volumes into the top table text output.
2861 SC/04 Eximstats V1.46
2862 Collect data on the number of addresses (recipients)
2863 as well as the number of messages.
2865 SC/05 Eximstats V1.47
2866 Added 'Message too big' to the list of mail rejection
2867 reasons (thanks to Marco Gaiarin).
2869 SC/06 Eximstats V1.48
2870 Mainlog lines which have GMT offsets and are too short to
2871 have a flag are now skipped.
2873 SC/07 Eximstats V1.49 (Alain Williams)
2874 Added the -emptyok flag.
2876 SC/08 Eximstats V1.50
2877 Fixes for obtaining the IP address from reject messages.
2879 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2880 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2881 whitespace changes from 4.64-PH/27
2883 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2886 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2887 are found, allow negative numbers in numeric criteria)
2889 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2891 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2892 to match changes made in 4.64-PH/43
2894 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2896 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2897 "domain" is actually an IP address, and force a failure. This locks out
2898 those revolvers/nameservers that support "A-for-A" lookups, in
2899 contravention of the specifications.
2901 PH/48 When a host name was looked up from an IP address, and the subsequent
2902 forward lookup of the name timed out, the host name was left in
2903 $sender_host_name, contrary to the specification.
2905 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2906 restricted to single-key lookups, Exim was not diagnosing an error if
2907 * or *@ was used with a query-style lookup.
2909 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2911 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2912 long ago, but noone interested enough thought of it. Let's just say that
2913 the "1.1" means that there are some new functions that weren't there at
2914 some point in the past.
2916 PH/51 Error processing for expansion failure of helo_data from an smtp
2917 transport during callout processing was broken.
2919 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2920 tested/used via the -bh/-bhc/-bs options.
2922 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2923 bug, fixed in subsequent PCRE releases).
2925 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2926 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2928 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2934 SC/01 Use a glob alias rather than an array ref in eximstats generated
2935 parser. This improves both readability and performance.
2937 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2938 Don't display local sender or destination tables in eximstats unless
2939 there is data to show.
2940 Added average volumes into the eximstats top table text output.
2942 SC/03 Collect data on the number of addresses (recipients) as well
2943 as the number of messages in eximstats.
2945 TF/01 Correct an error in the documentation for the redirect router. Exim
2946 does (usually) call initgroups() when daemonizing.
2948 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2949 with consistent privilege compared to when running as a daemon.
2951 TF/03 Note in the spec that $authenticated_id is not set for local
2952 submissions from trusted users.
2954 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2955 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2957 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2958 by adding some example configuration directives to the default
2959 configuration file. A little bit of work is required to uncomment the
2960 directives and define how usernames and passwords are checked, but
2961 there is now a framework to start from.
2963 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2964 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2965 without this. I don't know how relevant this is to other LDAP libraries.
2967 PH/02 Add the verb name to the "unknown ACL verb" error.
2969 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2971 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2973 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2974 directory not expanded when it should be if an expanded home directory
2975 was set for the address (which is overridden by the transport).
2977 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2980 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2981 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2982 because it is too late at that time, and has no effect.
2984 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2985 security issue with \' (bugzilla #107). I could not use the
2986 PQescapeStringConn() function, because it needs a PGconn value as one of
2989 PH/08 When testing addresses using -bt, indicate those final addresses that
2990 are duplicates that would not cause an additional delivery. At least one
2991 person was confused, thinking that -bt output corresponded to deliveries.
2992 (Suppressing duplicates isn't a good idea as you lose the information
2993 about possibly different redirections that led to the duplicates.)
2995 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2996 systems where poll() doesn't work, in particular OS X.
2998 PH/10 Added more information to debugging output for retry time not reached.
3000 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3001 operations in malware.c.
3003 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3006 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3007 syslog_duplication set false, log lines that would normally be written
3008 both the the main log and to the reject log were not written to syslog at
3011 PH/14 In the default configuration, change the use of "message" in ACL warn
3012 statements to "add_header".
3014 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3015 not followed by a command (e.g. "seen endif").
3017 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3018 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3021 PH/17 Added extra conditions to the default value of delay_warning_condition
3025 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3026 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3027 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3030 The Auto-Submitted: and various List- headers are standardised, whereas I
3031 don't think Precedence: ever was.
3033 PH/18 Refactored debugging code in route_finduser() to show more information,
3034 in particular, the error code if getpwnam() issues one.
3036 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3037 This is apparently needed in addition to the PH/07 change above to avoid
3038 any possible encoding problems.
3040 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3041 but not after initializing Perl.
3043 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3044 output them only if debugging. By default they are written stderr,
3045 apparently, which is not desirable.
3047 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3050 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3053 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3055 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3056 authenticated or an ident call has been made. Suppress the default
3057 values for $authenticated_id and $authenticated_sender (but permit -oMai
3058 and -oMas) when testing with -bh.
3060 PH/24 Re-jigged the order of the tests in the default configuration so that the
3061 tests for valid domains and recipients precede the DNS black list and CSA
3062 tests, on the grounds that those ones are more expensive.
3064 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3065 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3066 This bug exists in every version of Exim that I still have, right back to
3069 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3070 However, an attempt to turn on case-sensitivity in a regex key by
3071 including (?-i) didn't work because the subject string was already
3072 lowercased, and the effects were non-intuitive. It turns out that a
3073 one-line patch can be used to allow (?-i) to work as expected.
3079 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3080 other effects) broke the use of negated acl sub-conditions.
3082 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3085 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3086 "Deny" causes Exim to reject the incoming connection with a 554 error.
3087 Unfortunately, if there is a major crisis, such as a disk failure,
3088 tcp-wrappers gives "deny", whereas what one would like would be some
3089 kind of temporary error. A kludge has been added to help with this.
3090 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3091 554 error is used if errno is still zero or contains ENOENT (which occurs
3092 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3095 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3097 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3098 errors. Otherwise a message that provokes a temporary error (when other
3099 messages do not) can cause a whole host to time out.
3101 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3102 addresses were routed directly to files or pipes from a redirect router.
3103 File deliveries just didn't batch; pipe deliveries might have suffered
3106 PH/06 A failure to get a lock for a hints database would erroneously always say
3107 "Failed to get write lock", even when it was really a read lock.
3109 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3110 of 0600. This has been changed to use the value of the lockfile_mode
3111 option (which defaults to 0600).
3113 PH/08 Applied small patch from the Sieve maintainer.
3115 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3116 folder from quota calculations, a direct delivery into this folder messed
3117 up the contents of the maildirsize file. This was because the regex was
3118 used only to exclude .Trash (or whatever) when the size of the mailbox
3119 was calculated. There was no check that a delivery was happening into an
3120 excluded directory. This bug has been fixed by ignoring all quota
3121 processing for deliveries into excluded directories.
3123 PH/10 Added the maildirfolder_create_regex option to appendfile.
3129 PH/01 The code for finding all the local interface addresses on a FreeBSD
3130 system running IPv6 was broken. This may well have applied to all BSD
3131 systems, as well as to others that have similar system calls. The broken
3132 code found IPv4 interfaces correctly, but gave incorrect values for the
3133 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3134 that it would not match correctly against @[] and not recognize the IPv6
3137 PH/02 The ipliteral router was not recognizing addresses of the form user@
3138 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3140 PH/03 Added disable_ipv6.
3142 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3143 decoded form, because it is most often used to construct To: headers
3144 lines in autoreplies, and the decoded form may well be syntactically
3145 invalid. However, $reply_address has leading white space removed, and all
3146 newlines turned into spaces so that the autoreply transport does not
3149 PH/05 If group was specified without a user on a router, and no group or user
3150 was specified on a transport, the group from the router was ignored.
3152 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3153 for visible compile-time settings that can be used to change these
3154 numbers, for those that want even more. Backwards compatibility with old
3155 spool files has been maintained. However, going back to a previous Exim
3156 release will lost any variables that are in spool files.
3158 PH/07 Two small changes when running in the test harness: increase delay when
3159 passing a TCP/IP connection to a new process, in case the original
3160 process has to generate a bounce, and remove special handling of
3161 127.0.0.2 (sic), which is no longer necessary.
3163 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3164 be the same on different OS.
3166 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3169 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3170 whether --show-vars was specified or not
3172 JJ/02 exipick: Added support for new ACL variable spool format introduced
3175 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3176 syntactically invalid From: or Reply-to: line, and a filter used this to
3177 generate an autoreply, and therefore failed to obtain an address for the
3178 autoreply, Exim could try to deliver to a non-existent relative file
3179 name, causing unrelated and misleading errors. What now happens is that
3180 it logs this as a hard delivery error, but does not attempt to create a
3183 PH/11 The exinext utility has a -C option for testing purposes, but although
3184 the given file was scanned by exinext itself; it wasn't being passed on
3185 when Exim was called.
3187 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3188 an end-of-file indication when reading a command response.
3190 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3191 compiled. In many other places in Exim, IPv6 addresses are always
3192 recognized, so I have changed this. It also means that IPv4 domain
3193 literals of the form [IPV4:n.n.n.n] are now always recognized.
3195 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3196 used if the router is not running as root, for example, when verifying at
3197 ACL time, or when using -bh. The debugging output from this situation was
3198 non-existent - all you got was a failure to exec. I have made two
3201 (a) Failures to set uid/gid, the current directory, or a process leader
3202 in a subprocess such as that created by queryprogram now generate
3203 suitable debugging output when -d is set.
3205 (b) The queryprogram router detects when it is not running as root,
3206 outputs suitable debugging information if -d is set, and then runs
3207 the subprocess without attempting to change uid/gid.
3209 PH/15 Minor change to Makefile for building test_host (undocumented testing
3212 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3213 additional section of a DNS packet that returns MX or SRV records.
3214 Instead, it always explicitly searches for A/AAAA records. This avoids
3215 major problems that occur when a DNS server includes only records of one
3216 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3217 fixed another bug: if SRV records were looked up and the corresponding
3218 address records were *not* found in the additional section, the port
3219 values from the SRV records were lost.
3221 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3222 using the correct key (the original address) when searching the retry
3223 rules in order to find which one to use for generating the retry hint.
3225 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3226 adding the default one. Similarly, if it contains a Reply-To: header, the
3227 errors_reply_to option, if set, is not used.
3229 PH/19 When calculating a retry time, Exim used to measure the "time since
3230 failure" by looking at the "first failed" field in the retry record. Now
3231 it does not use this if it is later than than the arrival time of the
3232 message. Instead it uses the arrival time. This makes for better
3233 behaviour in cases where some deliveries succeed, thus re-setting the
3234 "first failed" field. An example is a quota failure for a huge message
3235 when small messages continue to be delivered. Without this change, the
3236 "time since failure" will always be short, possible causing more frequent
3237 delivery attempts for the huge message than are intended.
3238 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3240 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3241 $1, $2, $3) because the numerical variables can be reset during some
3242 expansion items (e.g. "match"), thereby losing the authentication data.
3244 PH/21 Make -bV show the size of off_t variables so that the test suite can
3245 decide whether to run tests for quotas > 2G.
3247 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3248 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3249 filecount value is greater than 2G or if a quota value is greater than 2G
3250 on a system where the size of off_t is not greater than 4, a panic error
3253 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3254 never match. The debug and -bh output now contains an explicit error
3255 message indicating a malformed IPv4 address or mask.
3257 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3258 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3259 PH/23 above applies.
3261 PH/25 Do not write to syslog when running in the test harness. The only
3262 occasion when this arises is a failure to open the main or panic logs
3263 (for which there is an explicit test).
3265 PH/26 Added the /no_tell option to "control=freeze".
3267 PH/27 If a host name lookup failed very early in a connection, for example, if
3268 the IP address matched host_lookup and the reverse lookup yielded a name
3269 that did not have a forward lookup, an error message of the form "no IP
3270 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3271 could be logged. Now it outputs the IP address instead of "NULL".
3273 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3274 allows the sender and the authenticated sender to be set when
3275 submitting a message from within Exim. Since child_open_exim() is
3276 documented for local_scan(), the new function should be too.
3278 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3279 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3280 results in an empty string is now treated as unset.
3282 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3284 PH/31 Added sender_verify_fail logging option.
3286 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3287 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3288 tidied the source and removed it altogether.
3290 PH/33 When a queue run was abandoned because the load average was too high, a
3291 log line was always written; now it is written only if the queue_run log
3292 selector is set. In addition, the log line for abandonment now contains
3293 information about the queue run such as the pid. This is always present
3294 in "start" and "stop" lines but was omitted from the "abandon" line.
3296 PH/34 Omit spaces between a header name and the colon in the error message that
3297 is given when verify = headers_syntax fails (if there are lots of them,
3298 the message gets confusing).
3300 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3301 names, as there are now some PTR records that contain slashes. This check
3302 is only to protect against broken name servers that fall over on strange
3303 characters, so the fact that it applies to all lookups doesn't matter.
3305 PH/36 Now that the new test suite is complete, we can remove some of the
3306 special code in Exim that was needed for the old test suite. For example,
3307 sorting DNS records because real resolvers return them in an arbitrary
3308 order. The new test suite's fake resolver always returns records in the
3311 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3312 bounces) except when queue_only is set, to avoid logging races between
3313 the different processes.
3315 PH/38 Panic-die if .include specifies a non-absolute path.
3317 PH/39 A tweak to the "H" retry rule from its user.
3319 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3320 a label. They prevented compilation on older perls.
3322 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3323 a warning to be raised on newish perls.
3325 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3326 on queue. Changes to match documented behaviour of showing count of
3327 messages matching specified criteria.
3329 PH/40 Changed the default ident timeout from 30s to 5s.
3331 PH/41 Added support for the use of login_cap features, on those BSD systems
3332 that have them, for controlling the resources used by pipe deliveries.
3334 PH/42 The content-scanning code uses fopen() to create files in which to put
3335 message data. Previously it was not paying any attention to the mode of
3336 the files. Exim runs with umask(0) because the rest of the code creates
3337 files with open(), and sets the required mode explicitly. Thus, these
3338 files were ending up world-writeable. This was not a big issue, because,
3339 being within the spool directory, they were not world-accessible. I have
3340 created a function called modefopen, which takes an additional mode
3341 argument. It sets umask(777), creates the file, chmods it to the required
3342 mode, then resets the umask. All the relevant calls to fopen() in the
3343 content scanning code have been changed to use this function.
3345 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3346 to 24 hours. This avoids potential overflow problems when processing G
3347 and H retry rules. I suspect nobody ever tinkers with this value.
3349 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3351 PH/45 When the plaintext authenticator is running as a client, the server's
3352 challenges are checked to ensure they are valid base64 strings. By
3353 default, the authentication attempt is cancelled if an invalid string is
3354 received. Setting client_ignore_invalid_base64 true ignores these errors.
3355 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3356 they are received. Thus, the responses can be made to depend on the
3357 challenges. If an invalid string is ignored, an empty string is placed in
3360 PH/46 Messages that are created by the autoreply transport now contains a
3361 References: header, in accordance with RFCs 2822 and 3834.
3363 PH/47 Added authenticated_sender_force to the smtp transport.
3365 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3367 PH/49 Installed latest patch from the Sieve maintainer.
3369 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3370 was also set, the appendfile transport was unnecessarily scanning a
3371 directory of message files (e.g. for maildir delivery) to find the count
3372 of files (along with the size), even though it did not need this
3373 information. It now does the scan only if it needs to find either the
3374 size of the count of files.
3376 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3378 PH/52 Two bugs concerned with error handling when the smtp transport is
3381 (i) Exim was not creating retry information for temporary errors given
3382 for individual recipients after the DATA command when the smtp transport
3383 was used in LMTP mode. This meant that they could be retried too
3384 frequently, and not timed out correctly.
3386 (ii) Exim was setting the flag that allows error details to be returned
3387 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3388 recipients that were returned after the DATA command.
3390 PH/53 This is related to PH/52, but is more general: for any failing address,
3391 when detailed error information was permitted to be returned to the
3392 sender, but the error was temporary, then after the final timeout, only
3393 "retry timeout exceeded" was returned. Now it returns the full error as
3394 well as "retry timeout exceeded".
3396 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3397 do this, and (what is worse) MTAs that accept it.
3399 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3400 will now be deprecated.
3402 PH/56 New os.c-cygwin from the Cygwin maintainer.
3404 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3405 formats (previously only available in exim formats via -bpr, -bpru,
3406 and -bpra. Now also available in native and exiqgrep formats)
3408 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3409 with very large, slow to parse queues
3411 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3413 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3415 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3416 responses to authentication challenges, though it was showing the
3417 challenges; (ii) I've removed the CR characters from the debug output for
3420 PH/58 Allow for the insertion of a newline as well as a space when a string
3421 is turned into more than one encoded-word during RFC 2047 encoding. The
3422 Sieve code now uses this.
3424 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3425 data_4xx, lost_connection, tls_required.
3427 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3428 message was being sent as an SMTP response.
3430 PH/61 Add -l and -k options to exicyclog.
3432 PH/62 When verifying, if an address was redirected to one new address, so that
3433 verification continued, and the new address failed or deferred after
3434 having set something in $address_data, the value of $address_data was not
3435 passed back to the ACL. This was different to the case when no
3436 redirection occurred. The value is now passed back in both cases.
3438 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3439 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3440 use login_cap.h, so on its own it isn't the distinguishing feature. The
3441 new name refers directly to the setclassresources() function.
3443 PH/65 Added configuration files for NetBSD3.
3445 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3447 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3448 is preferred over IPv4.
3450 PH/68 The bounce_return_message and bounce_return_body options were not being
3451 honoured for bounces generated during the reception of non-SMTP messages.
3452 In particular, this applied to messages rejected by the ACL. This bug has
3453 been fixed. However, if bounce_return_message is true and bounce_return_
3454 body is false, the headers that are returned for a non-SMTP message
3455 include only those that have been read before the error was detected.
3456 (In the case of an ACL rejection, they have all been read.)
3458 PH/69 The HTML version of the specification is now built in a directory called
3459 spec_html instead of spec.html, because the latter looks like a path with
3460 a MIME-type, and this confuses some software.
3462 PH/70 Catch two compiler warnings in sieve.c.
3464 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3465 function verify_get_ident() calls ip_connect() to connect a socket, but
3466 if the "connect()" function timed out, ip_connect() used to close the
3467 socket. However, verify_get_ident() also closes the socket later, and in
3468 between Exim writes to the log, which may get opened at this point. When
3469 the socket was closed in ip_connect(), the log could get the same file
3470 descriptor number as the socket. This naturally causes chaos. The fix is
3471 not to close the socket in ip_connect(); the socket should be closed by
3472 the function that creates it. There was only one place in the code where
3473 this was missing, in the iplookup router, which I don't think anybody now
3474 uses, but I've fixed it anyway.
3476 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3477 well as to direct DNS lookups. Otherwise the handling of names in host
3478 lists is inconsistent and therefore confusing.
3484 PH/01 Two changes to the default runtime configuration:
3486 (1) Move the checks for relay_from_hosts and authenticated clients from
3487 after to before the (commented out) DNS black list checks.
3489 (2) Add control=submission to the relay_from_hosts and authenticated
3490 clients checks, on the grounds that messages accepted by these
3491 statements are most likely to be submissions.
3493 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3495 (1) Generate an error if the third argument for the ${prvs expansion is
3498 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3501 (3) Reset the variables that are obtained from the first argument of
3502 ${prvscheck and used in the second argument before leaving the code,
3503 because their memory is reclaimed, so using them afterwards may do
3506 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3507 one (it's much easier than Tom thought :-).
3509 (5) Because of (4), we can now allow for the use of $prvscheck_result
3510 inside the third argument.
3512 PH/03 For some reason, the default setting of PATH when running a command from
3513 a pipe transport was just "/usr/bin". I have changed it to
3516 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3517 anything to be listed in the output from -bV.
3519 PH/05 When a filter generated an autoreply, the entire To: header line was
3520 quoted in the delivery log line, like this:
3522 => >A.N.Other <ano@some.domain> <original@ddress> ...
3524 This has been changed so that it extracts the operative address. There
3525 may be more than one such address. If so, they are comma-separated, like
3528 => >ano@some.domain,ona@other.domain <original@ddress> ...
3530 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3531 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3532 not being looked up in the rDNS to get a host name, Exim was showing the
3533 IP address twice in Received: lines, even though the IP addresses were
3534 identical. For example:
3536 Received: from [1.2.3.4] (helo=[1.2.3.4])
3538 However, if the real host name was known, it was omitting the HELO data
3539 if it matched the actual IP address. This has been tidied up so that it
3540 doesn't show the same IP address twice.
3542 PH/07 When both +timestamp and +memory debugging was on, the value given by
3543 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3544 called by the debug printing, thereby overwriting the timestamp buffer.
3545 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3547 PH/08 When the original message was included in an autoreply transport, it
3548 always said "this is a copy of the message, including all the headers",
3549 even if body_only or headers_only was set. It now gives an appropriate
3552 PH/09 Applied a patch from the Sieve maintainer which:
3554 o fixes some comments
3555 o adds the (disabled) notify extension core
3556 o adds some debug output for the result of if/elsif tests
3557 o points to the current vacation draft in the documentation
3558 and documents the missing references header update
3562 o fixes a bug in processing the envelope test (when testing
3563 multiple envelope elements, the last element determined the
3566 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3567 Electronic Mail") by including:
3569 Auto-submitted: auto-generated
3571 in the messages that it generates (bounce messages and others, such as
3572 warnings). In the case of bounce messages for non-SMTP messages, there was
3573 also a typo: it was using "Auto_submitted" (underscore instead of
3574 hyphen). Since every message generated by Exim is necessarily in response
3575 to another message, thes have all been changed to:
3577 Auto-Submitted: auto-replied
3579 in accordance with these statements in the RFC:
3581 The auto-replied keyword:
3583 - SHOULD be used on messages sent in direct response to another
3584 message by an automatic process,
3586 - MUST NOT be used on manually-generated messages,
3588 - MAY be used on Delivery Status Notifications (DSNs) and Message
3589 Disposition Notifications (MDNs),
3591 - MUST NOT be used on messages generated by automatic or periodic
3592 processes, except for messages which are automatic responses to
3595 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3596 to the default Received: header definition.
3598 PH/12 Added log selector acl_warn_skipped (default on).
3600 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3601 variables because (a) they are in the wrong storage pool and (b) even if
3602 they were copied, it wouldn't work properly because of the caching.
3604 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3605 checking when decoding. Apparently there are clients that generate
3606 overlong encoded strings. Why am I not surprised?
3608 PH/15 If the first argument of "${if match_address" was not empty, but did not
3609 contain an "@" character, Exim crashed. Now it writes a panic log message
3610 and treats the condition as false.
3612 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3614 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3615 extension "envelope-auth". The code is finished and in agreement with
3616 other implementations, but there is no documentation so far and in fact,
3617 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3618 not changing the active code.
3620 Print executed "if" and "elsif" statements when debugging is used. This
3621 helps a great deal to understand what a filter does.
3623 Document more things not specified clearly in RFC3028. I had all this
3624 sorted out, when out of a sudden new issues came to my mind. Oops."
3626 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3629 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3630 canonical form (as documented). However, after a host name lookup from
3631 the IP address, check_host() was doing a simple string comparison with
3632 addresses acquired from the DNS when checking that the found name did
3633 have the original IP as one of its addresses. Since any found IPv6
3634 addresses are likely to be in abbreviated form, the comparison could
3635 fail. Luckily, there already exists a function for doing the comparison
3636 by converting both addresses to binary, so now that is used instead of
3637 the text comparison.
3639 PH/20 There was another similar case to PH/19, when a complete host name was
3640 given in a host list; looking up its IP address could give an abbreviated
3641 form, whereas the current host's name might or might not be abbreviated.
3642 The same fix has been applied.
3648 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3649 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3652 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3653 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3655 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3657 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3658 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3659 Also, for "Auto-Submitted", treat anything other than "no" as
3660 non-personal, in accordance with RFC 3834. (Previously it treated
3661 anything starting "auto-" as non-personal.)
3663 TF/01 The control=submission/name=... option had a problem with syntax
3664 errors if the name included a slash character. The /name= option
3665 now slurps the rest of the string, so it can include any characters
3666 but it must come last in the list of options (after /sender_retain
3669 PH/05 Some modifications to the interface to the fake nameserver for the new
3677 TK/01 Added the "success_on_redirect" address verification option. See
3678 NewStuff for rationale and an example.
3680 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3682 PH/02 Patch to exigrep to allow it to work on syslog lines.
3684 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3685 fread() to skip over the body file's header line, because in Cygwin the
3686 header line is locked and is inaccessible.
3688 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3689 co-exist for some time) to make it clear that it is the Exim ID that is
3690 referenced, not the Message-ID: header line.
3692 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3693 string_format() function, because snprintf() does not exist on all
3696 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3697 ${stat: expansion item.
3699 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3700 protocol synchronization error", to keep the pedants happy.
3702 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3703 well as for IRIX systems, when gcc is being used. See the host.c source
3706 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3708 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3711 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3712 transport and to the smtp transport in LMTP mode.
3714 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3716 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3717 run a filter in a subprocess. This could lead to confusion in subsequent
3718 lookups in the parent process. There should also be a search_tidyup() at
3719 the end of the subprocess.
3721 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3722 only if the host matched helo_try_verify_hosts, which caused the
3723 verification to occur when the EHLO/HELO command was issued. The ACL just
3724 tested the remembered result. Now, if a previous verification attempt has
3725 not happened, "verify = helo" does it there and then.
3727 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3729 TK/03 Fix log output including CR from clamd.
3731 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3732 exist provoked a memory error which could cause a segfault.
3734 PH/15 Installed PCRE 6.2
3736 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3738 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3739 of the problem. Specifically, suggested +O2 rather than +O1 for the
3742 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3744 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3745 record, the port was not correctly logged when the outgoing_port log
3746 selector was set (it logged the transort's default port).
3748 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3749 fallback_hosts, and "hosts" in the smtp transport.
3751 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3752 host errors such as "Connection refused".
3754 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3755 authentication with radiusclient 0.4.9:
3757 - Error returned from rc_read_config was caught wrongly
3758 - Username/password not passed on to radius server due to wrong length.
3760 The presumption is that some radiusclient API changes for 4.51/PH/17
3761 were not taken care of correctly. The code is still untested by me (my
3762 Linux distribution still has 0.3.2 of radiusclient), but it was
3763 contributed by a Radius user.
3765 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3766 expanding the "port" option of the smtp transport.
3768 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3769 while reading a MIME header. Thanks to Tom Hughes for a patch.
3771 PH/24 Include config.h inside local_scan.h so that configuration settings are
3774 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3775 that in an ACL for RCPT (for example), you can examine exactly what was
3778 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3779 commands, but it was not correctly comparing the address with the actual
3780 client host address. Thus, it would show the EHLO address in Received:
3781 header lines when this was not necessary.
3783 PH/27 Added the % operator to ${eval:}.
3785 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3786 it should be ignoring failures (because with -C, for example, it has lost
3787 privilege). It wasn't ignoring creation failures other than "already
3790 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3793 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3794 because an input line was too long, either on its own, or by virtue of
3795 too many continuations, the temporary file was not being removed, and the
3796 return code was incorrect.
3798 PH/31 Missing "BOOL" in function definition in filtertest.c.
3800 PH/32 Applied Sieve patches from the maintainer.
3802 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3804 PH/33 Added "verify = not_blind".
3806 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3807 Local/Makefile (with some defaults set). These are used in built scripts
3808 such as exicyclog, but they have never been used in the exim_install
3809 script (though there are many overriding facilities there). I have
3810 arranged that the exim_install script now takes note of these two
3813 PH/35 Installed configuration files for Dragonfly.
3815 PH/36 When a locally submitted message by a trusted user did not contain a
3816 From: header, and the sender address was obtained from -f or from an SMTP
3817 MAIL command, and the trusted user did not use -F to supply a sender
3818 name, $originator_name was incorrectly used when constructing a From:
3819 header. Furthermore, $originator_name was used for submission mode
3820 messages from external hosts without From: headers in a similar way,
3821 which is clearly wrong.
3823 PH/37 Added control=suppress_local_fixups.
3825 PH/38 When log_selector = +received_sender was set, and the addition of the
3826 sender made the log line's construction buffer exactly full, or one byte
3827 less than full, an overflow happened when the terminating "\n" was
3830 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3831 when the result of a list match is failure because a DNS lookup failed.
3833 PH/40 RM_COMMAND is now used in the building process.
3835 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3836 the "build-* directories that it finds.
3838 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3839 address in a domain literal was a prefix of an interface address.
3841 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3842 when verifying a sender address, unless rewrite_headers is false.
3844 PH/44 Wrote a long comment about why errors_to addresses are verified as
3845 recipients, not senders.
3847 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3848 the ratelimit ACL was added.
3850 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3852 PH/46 Added extra information about PostgreSQL errors to the error string.
3854 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3855 avoiding the need to install special zones in a real server. This is
3856 backwards compatible; if it can't find the fake resolver, it drops back.
3857 Thus, both old and new test suites can be run.
3859 TF/02 Added util/ratelimit.pl
3861 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3862 clock is set back in time.
3864 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3865 Candler <B.Candler@pobox.com>.
3867 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3868 OK for addresses that are the result of redirections.
3870 PH/48 A number of further additions for the benefit of the new test suite,
3871 including a fake gethostbyname() that interfaces to the fake DNS resolver
3874 TF/06 The fix for widen_domains has also been applied to qualify_single and
3875 search_parents which are the other dnslookup options that can cause
3878 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3881 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3883 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3884 service" if it encountered an MX record with an empty target hostname.
3885 The message is now "an MX or SRV record indicated no SMTP service".
3887 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3888 if the DNS of the sending site is misconfigured. This is quite a
3889 common situation. This change restores the behaviour of treating a
3890 helo verification defer as a failure.
3892 PH/51 If self=fail was set on a router, the bounce message did not include the
3893 actual error message.
3899 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3901 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3902 command itself ended in error, the underlying message about the transport
3903 filter timeout was being overwritten with the pipe command error. Now the
3904 underlying error message should be appended to the second error message.
3906 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3908 PH/02 Reduce the amount of output that "make" produces by default. Full output
3909 can still be requested.
3911 PH/03 The warning log line about a condition test deferring for a "warn" verb
3912 was being output only once per connection, rather than after each
3913 occurrence (because it was using the same function as for successful
3914 "warn" verbs). This seems wrong, so I have changed it.
3916 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3917 it should not have, which might have caused a crash in the right
3918 circumstances, but probably never did.
3920 PH/04 Installed a modified version of Tony Finch's patch to make submission
3921 mode fix the return path as well as the Sender: header line, and to
3922 add a /name= option so that you can make the user's friendly name appear
3925 TF/03 Added the control = fakedefer ACL modifier.
3927 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3928 Mark Lowes for thorough testing.
3930 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3932 TK/03 Merged latest SRS patch from Miles Wilton.
3934 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3935 with the definition in sysexits.h (which is #included earlier).
3936 Fortunately, Exim does not actually use EX_OK. The code used to try to
3937 preserve the sysexits.h value, by assuming that macro definitions were
3938 scanned for macro replacements. I have been disabused of this notion,
3939 so now the code just undefines EX_OK before #including unistd.h.
3941 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3942 in the smtp transport. When a block could not be written in a single
3943 write() function, the timeout was being re-applied to each part-write.
3944 This seems wrong - if the receiver was accepting one byte at a time it
3945 would take for ever. The timeout is now adjusted when this happens. It
3946 doesn't have to be particularly precise.
3948 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3949 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3951 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3952 without a domain if the check to <postmaster@domain> fails.
3954 SC/01 Eximstats: added -xls and the ability to specify output files
3955 (patch written by Frank Heydlauf).
3957 SC/02 Eximstats: use FileHandles for outputting results.
3959 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3961 SC/04 Eximstats: fixed display of large numbers with -nvr option
3963 SC/05 Eximstats: fixed merging of reports with empty tables.
3965 SC/06 Eximstats: added the -include_original_destination flag
3967 SC/07 Eximstats: removed tabs and trailing whitespace.
3969 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3971 TK/06 MBOX spool code: Add real "From " MBOX separator line
3972 so the .eml file is really in mbox format (even though
3973 most programs do not really care). Patch from Alex Miller.
3975 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3976 The latter is generated from $received_to and is only set if the
3977 message has one envelope recipient. SA can use these headers,
3978 obviously out-of-the-box. Patch from Alex Miller.
3980 PH/08 The ${def test on a variable was returning false if the variable's
3981 value was "0", contrary to what the specification has always said!
3982 The result should be true unless the variable is empty.
3984 PH/09 The syntax error of a character other than { following "${if
3985 def:variable_name" (after optional whitespace) was not being diagnosed.
3986 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3987 accidental colon was present, for example, could give incorrect results.
3989 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3990 result is used (not including appendfile, where other changes are about
3993 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3994 This involved changing a lot of size variables from int to off_t. It
3995 should work with maildirs and everything.
3997 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3998 spamd dying while we are connected to it.
4000 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4003 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4004 function for BDB 4.3.
4006 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4008 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4009 the sources for PCRE have been re-arranged and more files are now
4012 PH/15 The code I had for printing potentially long long variables in PH/11
4013 above was not the best (it lost precision). The length of off_t variables
4014 is now inspected at build time, and an appropriate printing format (%ld
4015 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4016 to be "long long int" or "long int". This is needed for the internal
4017 formatting function string_vformat().
4019 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4020 the configuration file to be ":syslog", then the script "guesses" where
4021 the logs files are, rather than using the compiled in default. In our
4022 case the guess is not the same as the compiled default, so the script
4023 suddenly stopped working when I started to use syslog. The patch checks
4024 to see if log_file_path is "". If so, it attempts to read it from exim
4025 with no configuration file to get the compiled in version, before it
4026 falls back to the previous guessing code."
4028 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4029 implementing BATV in an Exim configuration. See NewStuff for the gory
4032 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4033 Makefile that are specific to HP-UX.
4035 PH/18 If the "use_postmaster" option was set for a recipient callout together
4036 with the "random" option, the postmaster address was used as the MAIL
4037 FROM address for the random test, but not for the subsequent recipient
4038 test. It is now used for both.
4040 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4041 patch removes a few documentation additions to RFC 3028, because the
4042 latest draft now contains them. It adds the new en;ascii-case comparator
4043 and a new error check for 8bit text in MIME parts. Comparator and
4044 require names are now matched exactly. I enabled the subaddress
4045 extension, but it is not well tested yet (read: it works for me)."
4047 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4048 rework some of the code of TK/09 above to avoid the hardwired use of
4049 "%lld" and "long long". Replaced the call to snprintf() with a call to
4052 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4053 records point to non-existent hosts", "retry timeout exceeded", and
4054 "retry time not reached for any host after a long failure period".
4056 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4057 experimental DomainKeys support:
4059 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4060 (2) On an error such as an illegally used "control", the wrong name for
4061 the control was given.
4063 These problems did NOT occur unless DomainKeys support was compiled.
4065 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4067 PH/24 Added ${if match_ip condition.
4069 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4070 databases so that it will be absolutely obvious if a crash occurs in the
4071 DB library. This is a regular occurrence (often caused by mis-matched
4074 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4075 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4076 picked up on a user's system that detects such things. There doesn't seem
4077 to be a gcc warning option for this - only an attribute that has to be
4078 put on the function's prototype. It seems that in Fedora Core 4 they have
4079 set this on a number of new functions. No doubt there will be more in due
4082 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4083 specify a transport. However, if an address that was verified by such a
4084 router was the subject of a callout, Exim crashed because it tried to
4085 read the rcpt_include_affixes from the non-existent transport. Now it
4086 just assumes that the setting of that option is false. This bug was
4087 introduced by 4.51/PH/31.
4089 PH/28 Changed -d+all to exclude +memory, because that information is very
4090 rarely of interest, but it makes the output a lot bigger. People tend to
4091 do -d+all out of habit.
4093 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4094 code in os-type was giving problems when libc.so lives in lib64, like on
4097 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4098 aren't the modern standard, and it seems that some systems' include files
4099 don't always have them. Exim was already checking for some of the newer
4100 ones like T_AAAA, and defining it itself. I've added checks for all the
4101 record types that Exim uses.
4103 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4104 not automatically generating a new one, as it is supposed to. This
4105 prevented TLS from working. If the file did exist, but contained invalid
4106 data, a new version was generated, as expected. It was only the case of a
4107 non-existent file that was broken.
4109 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4110 with a change in libdomainkeys > 0.64.
4112 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4113 from DNS. If the selector record carries the flag, it now has
4114 precedence over the domain-wide flag.
4116 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4118 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4119 the use of a port name that isn't defined in /etc/services) occurred, the
4120 message was deferred as in a normal delivery, and thus remained on the
4121 spool, instead of being failed because of the mua_wrapper setting. This
4122 is now fixed, and I tidied up some of the mua_wrapper messages at the
4125 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4126 the messages in a hash of arrays rather than using individual hashes.
4127 This is a bit cleaner and results in dramatic memory savings, albeit
4128 at a slight CPU cost.
4130 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4131 as requested by Marc Sherman.
4133 SC/10 Eximstats: added histograms for user specified patterns as requested
4136 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4138 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4139 fopen() in the content-scanning modules that did not already have it.
4145 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4146 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4148 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4150 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4152 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4153 to test Sieve filters that use "vacation".
4155 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4156 that changes the way the GnuTLS parameters are stored in the cache file.
4157 The new format can be generated externally. For backward compatibility,
4158 if the data in the cache doesn't make sense, Exim assumes it has read an
4159 old-format file, and it generates new data and writes a new file. This
4160 means that you can't go back to an older release without removing the
4163 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4164 work if there are any delivery delays because "one_time" forces the
4165 parent to be marked "delivered", so its unseen clone is never tried
4166 again. For this reason, Exim now forbids the simultaneous setting of
4169 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4170 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4171 entry if you want to know the details. The fix, however, neglected to
4172 consider the case when local delivery batching is involved. The test for
4173 "previously delivered" was not happening when checking to see if an
4174 address could be batched with a previous (undelivered) one; under
4175 certain circumstances this could lead to multiple deliveries to the same
4178 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4179 in its include files, and this causes problems building Exim.
4181 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4182 header_syntax) but Exim was just ignoring anything given after a slash.
4183 In particular, this caused confusion with an attempt to use "verify =
4184 reverse_host_lookup/defer_ok". An error is now given when options are
4185 supplied for verify items that do not have them. (Maybe reverse_host_
4186 lookup should have a defer_ok option, but that's a different point.)
4188 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4189 defined by RFC 821) to 2048, because there were problems with some AUTH
4190 commands, and RFC 1869 says the size should be increased for extended
4191 SMTP commands that take arguments.
4193 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4196 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4197 "unknown" error; now it says that the functionality isn't in the binary.
4199 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4200 an address' error message when a string expansion fails (syntax or
4201 whatever). Otherwise the password may appear in the log. Following change
4202 PH/42 below, there is no longer a chance of it appearing in a bounce
4205 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4207 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4208 of its addresses was ever tried. (Bugzilla bug #2.)
4210 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4211 the result incorrectly in the debug output. (It correctly added a newline
4212 to what was transported.)
4214 TF/01 Added $received_time.
4216 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4217 commented out examples of how to interface to a virus scanner and to
4218 SpamAssassin. Also added commented examples of av_scanner and
4219 spamd_address settings.
4221 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4222 and controls are allowed in which ACLs. There were a couple of minor
4223 errors. Some of the entries in the conditions table (which is a table of
4224 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4225 negation of where the condition IS allowed.
4227 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4229 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4230 header file does not have a version number, so I've had to invent a new
4231 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4232 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4233 radiusclient), but it was contributed by a Radius user.
4235 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4236 files or directories, for OpenSSL.
4238 PH/19 When an Exim process that is running as root has to create an Exim log
4239 file, it does so in a subprocess that runs as exim:exim so as to get the
4240 ownership right at creation (otherwise, other Exim processes might see
4241 the file with the wrong ownership). There was no test for failure of this
4242 fork() call, which would lead to the process getting stuck as it waited
4243 for a non-existent subprocess. Forks do occasionally fail when resources
4244 run out. I reviewed all the other calls to fork(); they all seem to check
4247 PH/20 When checking for unexpected SMTP input at connect time (before writing
4248 the banner), Exim was not dealing correctly with a non-positive return
4249 from the read() function. If the client had disconnected by this time,
4250 the result was a log entry for a synchronization error with an empty
4251 string after "input=" when read() returned zero. If read() returned -1
4252 (an event I could not check), uninitialized data bytes were printed.
4253 There were reports of junk text (parts of files, etc) appearing after
4256 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4258 PH/22 Added support for macro redefinition, and (re)definition in between
4259 driver and ACL definitions.
4261 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4262 forgetting to use the resulting value; it was using the unexpanded value.
4264 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4265 hadn't been configured. The fix is from Juergen Kreileder, who
4266 understands it better than I do:
4268 "Here's what I see happening with three configured cyrus_sasl
4269 authenticators configured (plain, login, cram-md5):
4271 On startup auth_cyrus_sasl_init() gets called for each of these.
4272 This means three calls to sasl_listmech() without a specified mech_list.
4273 => SASL tests which mechs of all available mechs actually work
4274 => three warnings about OTP not working
4275 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4277 With the patch, sasl_listmech() also gets called three times. But now
4278 SASL's mech_list option is set to the server_mech specified in the the
4279 authenticator. Or in other words, the answer from sasl_listmech()
4280 gets limited to just the mech you're testing for (which is different
4282 => the return list contains just 'plain' or 'login', 'cram-md5' or
4283 nothing depending on the value of ob->server_mech.
4285 I've just tested the patch: Authentication still works fine,
4286 unavailable mechs specified in the exim configuration are still
4287 caught, and the auth.log warnings about OTP are gone."
4289 PH/25 When debugging is enabled, the contents of the command line are added
4290 to the debugging output, even when log_selector=+arguments is not
4293 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4294 answer is "GNU", and only if the return is "GNU/something" is the answer
4297 PH/27 $acl_verify_message is now set immediately after the failure of a
4298 verification in an ACL, and so is available in subsequent modifiers. In
4299 particular, the message can be preserved by coding like this:
4301 warn !verify = sender
4302 set acl_m0 = $acl_verify_message
4304 Previously, $acl_verify_message was set only while expanding "message"
4305 and "log_message" when a very denied access.
4307 PH/28 Modified OS/os.c-Linux with
4309 -#ifndef OS_LOAD_AVERAGE
4310 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4312 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4313 nomenclature these days.)
4315 PH/29 Installed patch from the Sieve maintainer that adds the options
4316 sieve_useraddress and sieve_subaddress to the redirect router.
4318 PH/30 In these circumstances:
4319 . Two addresses routed to the same list of hosts;
4320 . First host does not offer TLS;
4321 . First host accepts first address;
4322 . First host gives temporary error to second address;
4323 . Second host offers TLS and a TLS session is established;
4324 . Second host accepts second address.
4325 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4326 and peerdn, if requested) that were in fact used only for the second
4329 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4330 attention to any local part prefix or suffix that was matched by the
4331 router that accepted the address. It now behaves in the same way as it
4332 does for delivery: the affixes are removed from the local part unless
4333 rcpt_include_affixes is set on the transport.
4335 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4336 timeout during the DATA phase of an incoming message.
4338 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4339 applied a patch sanctioned by the Sieve maintainer.
4341 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4342 the uid or gid is negative. A case of a negative gid caused this to be
4343 noticed. The fix allows for either to be negative.
4345 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4346 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4349 PH/36 The change PH/12 above was broken. Fixed it.
4351 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4352 the grounds that routing the same address twice would always produce the
4353 same answer. This might have been true once, but it is certainly no
4354 longer true now. Routing a child address may depend on the previous
4355 routing that produced that child. Some complicated redirection strategies
4356 went wrong when messages had multiple recipients, and made Exim's
4357 behaviour dependent on the order in which the addresses were given.
4359 I have moved the duplicate checking until after the routing is complete.
4360 Exim scans the addresses that are assigned to local and remote
4361 transports, and removes any duplicates. This means that more work will be
4362 done, as duplicates will always all be routed, but duplicates are
4363 presumably rare, so I don't expect this is of any significance.
4365 For deliveries to pipes, files, and autoreplies, the duplicate checking
4366 still happens during the routing process, since they are not going to be
4369 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4370 It corrects a timeout issue with spamd. This is Ian's comment: "The
4371 background is that sometimes spamd either never reads data from a
4372 connection it has accepted, or it never writes response data. The exiscan
4373 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4374 blindly assumes that writes won't block so it may never time out."
4376 PH/39 Allow G after quota size as well as K and M.
4378 PH/40 The value set for $authenticated_id in an authenticator may not contain
4379 binary zeroes or newlines because the value is written to log lines and
4380 to spool files. There was no check on this. Now the value is run through
4381 the string_printing() function so that such characters are converted to
4382 printable escape sequences.
4384 PH/41 $message_linecount is a new variable that contains the total number of
4385 lines in the message. Compare $body_linecount, which is the count for the
4388 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4389 bounce and delay warning messages, except in certain special cases, which
4392 (a) An SMTP error message from a remote host;
4393 (b) A message specified in a :fail: redirection;
4394 (c) A message specified in a "fail" command in a system filter;
4395 (d) A message specified in a FAIL return from the queryprogram router;
4396 (e) A message specified by the cannot_route_message router option.
4398 In these cases only, Exim does include the error details in bounce and
4399 warning messages. There are also a few cases where bland messages such
4400 as "unrouteable address" or "local delivery error" are given.
4402 PH/43 $value is now also set for the "else" part of a ${run expansion.
4404 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4405 being worked on, but at least Exim now implements the latest version to
4408 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4409 process to complete was treated as a delivery failure, a timeout while
4410 writing the message to the pipe was logged, but erroneously treated as a
4411 successful delivery. Such timeouts include transport filter timeouts. For
4412 consistency with the overall process timeout, these timeouts are now
4413 treated as errors, giving rise to delivery failures by default. However,
4414 there is now a new Boolean option for the pipe transport called
4415 timeout_defer, which, if set TRUE, converts the failures into defers for
4416 both kinds of timeout. A transport filter timeout is now identified in
4419 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4420 systems where "make" and "gmake" are different, calling "gmake" at top
4421 level broke things. I've arranged for the value of $(MAKE) to be passed
4422 from the Makefile to this script so that it can call the same version of
4426 A note about Exim versions 4.44 and 4.50
4427 ----------------------------------------
4429 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4430 changes of various kinds. As a consequence, a big documentation update was
4431 needed. This delayed the release for rather longer than seemed good, especially
4432 in the light of a couple of (minor) security issues. Therefore, the changes
4433 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4434 release. So 4.44 and 4.50 are in effect two different branches that both start
4437 I have left the 4.50 change log unchanged; it contains all the changes since
4438 4.43. The change log for 4.44 is below; many of its items are identical to
4439 those for 4.50. This seems to be the most sensible way to preserve the
4440 historical information.
4446 1. Minor wording change to the doc/README.SIEVE file.
4448 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4449 computation of the current number of files was incorrect.
4451 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4452 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4455 4. Give more explanation in the error message when the command for a transport
4456 filter fails to execute.
4458 5. There are several places where Exim runs a non-Exim command in a
4459 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4460 was being done only for the command run by the queryprogram router. It is
4461 now done for all such subprocesses. The other cases are: ${run, transport
4462 filters, and the commands run by the lmtp and pipe transports.
4464 6. Added CONFIGURE_GROUP build-time option.
4466 7. Some older OS have a limit of 256 on the maximum number of file
4467 descriptors. Exim was using setrlimit() to set 1000 as a large value
4468 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4469 systems. I've change it so that if it can't get 1000, it tries for 256.
4471 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4472 was an oversight, and furthermore, ever since the addition of extra
4473 controls (e.g. 4.43/32), the checks on when to allow different forms of
4474 "control" were broken. There should now be diagnostics for all cases when a
4475 control that does not make sense is encountered.
4477 9. Added the /retain_sender option to "control=submission".
4479 10. $recipients is now available in the predata ACL (oversight).
4481 11. Tidy the search cache before the fork to do a delivery from a message
4482 received from the command line. Otherwise the child will trigger a lookup
4483 failure and thereby defer the delivery if it tries to use (for example) a
4484 cached ldap connection that the parent has called unbind on.
4486 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4487 of $address_data from the recipient verification was clobbered by the
4488 sender verification.
4490 13. The value of address_data from a sender verification is now available in
4491 $sender_address_data in subsequent conditions in the ACL statement.
4493 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4495 15. Added a new option "connect=<time>" to callout options, to set a different
4498 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4499 was its contents. (It was OK if the option was not defined at all.)
4501 17. A "Completed" log line is now written for messages that are removed from
4502 the spool by the -Mrm option.
4504 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4505 information about exactly what failed.
4507 19. Added -dd to debug only the daemon process.
4509 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4510 handles timeouts, both on the server side and network timeouts. Renamed the
4511 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4513 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4514 It is now set to "smtps".
4516 22. $host_address is now set to the target address during the checking of
4517 ignore_target_hosts.
4519 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4520 being passed; this would have caused $sender_host_name to have been used if
4521 matching the list had actually called for a host name (not very likely,
4522 since this list is usually IP addresses). A host name is now passed as
4525 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4526 code that specifies a non-restarting handler (typically sigaction() in
4527 modern systems) in an attempt to fix a rare and obscure crash bug.
4529 25. Narrowed the window for a race in the daemon that could cause it to ignore
4530 SIGCHLD signals. This is not a major problem, because they are used only to
4531 wake it up if nothing else does.
4533 26. A malformed maildirsize file could cause Exim to calculate negative values
4534 for the mailbox size or file count. Odd effects could occur as a result.
4535 The maildirsize information is now recalculated if the size or filecount
4538 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4539 support for a long time. Removed HAVE_SYS_VFS_H.
4541 28. Installed the latest version of exipick from John Jetmore.
4543 29. In an address list, if the pattern was not a regular expression, an empty
4544 subject address (from a bounce message) matched only if the pattern was an
4545 empty string. Non-empty patterns were not even tested. This was the wrong
4546 because it is perfectly reasonable to use an empty address as part of a
4547 database query. An empty address is now tested by patterns that are
4548 lookups. However, all the other forms of pattern expect the subject to
4549 contain a local part and a domain, and therefore, for them, an empty
4550 address still always fails if the pattern is not itself empty.
4552 30. Exim went into a mad DNS loop when attempting to do a callout where the
4553 host was specified on an smtp transport, and looking it up yielded more
4554 than one IP address.
4556 31. Re-factored the code for checking spool and log partition space into a
4557 function that finds that data and another that does the check. The former
4558 is then used to implement four new variables: $spool_space, $log_space,
4559 $spool_inodes, and $log_inodes.
4561 32. The RFC2047 encoding function was originally intended for short strings
4562 such as real names; it was not keeping to the 75-character limit for
4563 encoded words that the RFC imposes. It now respects the limit, and
4564 generates multiple encoded words if necessary. To be on the safe side, I
4565 have increased the buffer size for the ${rfc2047: expansion operator from
4568 33. It is now permitted to omit both strings after an "if" condition; if the
4569 condition is true, the result is "true". As before, when the second string
4570 is omitted, a false condition yields an empty string. This makes it less
4571 cumbersome to write custom ACL and router conditions.
4573 34. Failure to deliver a bounce message always caused it to be frozen, even if
4574 there was an errors_to setting on the router. The errors_to setting is now
4577 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4578 canonical form (fully expanded) before being placed in
4579 $sender_host_address.
4581 36. The table in the code that translates DNS record types into text (T_A to
4582 "A" for instance) was missing entries for NS and CNAME. It is just possible
4583 that this could have caused confusion if both these types were looked up
4584 for the same domain, because the text type is used as part of Exim's
4585 per-process caching. But the chance of anyone hitting this buglet seems
4588 37. The dnsdb lookup has been extended in a number of ways.
4590 (1) There is a new type, "zns", which walks up the domain tree until it
4591 finds some nameserver records. It should be used with care.
4593 (2) There is a new type, "mxh", which is like "mx" except that it returns
4594 just the host names, not the priorities.
4596 (3) It is now possible to give a list of domains (or IP addresses) to be
4597 looked up. The behaviour when one of the lookups defers can be
4598 controlled by a keyword.
4600 (4) It is now possible to specify the separator character for use when
4601 multiple records are returned.
4603 38. The dnslists ACL condition has been extended: it is now possible to supply
4604 a list of IP addresses and/or domains to be looked up in a particular DNS
4607 39. Added log_selector=+queue_time_overall.
4609 40. When running the queue in the test harness, wait just a tad after forking a
4610 delivery process, to get repeatability of debugging output.
4612 41. Include certificate and key file names in error message when GnuTLS fails
4613 to set them up, because the GnuTLS error message doesn't include the name
4614 of the failing file when there is a problem reading it.
4616 42. Allow both -bf and -bF in the same test run.
4618 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4620 44. The "Exiscan patch" is now merged into the mainline Exim source.
4622 45. Sometimes the final signoff response after QUIT could fail to get
4623 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4624 < 0 before doing a fflush(). This bug looks as though it goes back to the
4625 introduction of TLS in release 3.20, but "sometimes" must have been rare
4626 because the tests only now provoked it.
4628 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4629 (this can affect the format of dates).
4631 47. exim_tidydb, when checking for the continued existence of a message for
4632 which it has found a message-specific retry record, was not finding
4633 messages that were in split spool directories. Consequently, it was
4634 deleting retry records that should have stayed in existence.
4636 48. Steve fixed some bugs in eximstats.
4638 49. The SPA authentication driver was not abandoning authentication and moving
4639 on to the next authenticator when an expansion was forced to fail,
4640 contradicting the general specification for all authenticators. Instead it
4641 was generating a temporary error. It now behaves as specified.
4643 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4644 (the order specifies the preference for clients). The order is now AES256,
4645 AES128, 3DES, ARCFOUR128.
4647 51. Small patch to Sieve code - explicitly set From: when generating an
4650 52. Exim crashed if a remote delivery caused a very long error message to be
4651 recorded - for instance if somebody sent an entire SpamAssassin report back
4652 as a large number of 550 error lines. This bug was coincidentally fixed by
4653 increasing the size of one of Exim's internal buffers (big_buffer) that
4654 happened as part of the Exiscan merge. However, to be on the safe side, I
4655 have made the code more robust (and fixed the comments that describe what
4658 53. Now that there can be additional text after "Completed" in log lines (if
4659 the queue_time_overall log selector is set), a one-byte patch to exigrep
4660 was needed to allow it to recognize "Completed" as not the last thing in
4663 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4664 patch that reportedly fixes this has been added. I am not expert enough to
4665 create a test for it. This is what the patch creator wrote:
4667 "I found a little strange behaviour of ldap code when working with
4668 Windows 2003 AD Domain, where users was placed in more than one
4669 Organization Units. When I tried to give exim partial DN, the exit code
4670 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4671 But simultaneously result of request was absolutely normal ldap result,
4672 so I produce this patch..."
4674 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4675 REFERENCE, so I have modified the code to exclude the patch when that macro
4678 55. Some experimental protocols are using DNS PTR records for new purposes. The
4679 keys for these records are domain names, not reversed IP addresses. The
4680 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4681 leaves it alone. Component reversal etc. now happens only for IP addresses.
4684 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4686 57. Double the size of the debug message buffer (to 2048) so that more of very
4687 long debug lines gets shown.
4689 58. The exicyclog utility now does better if the number of log files to keep
4690 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4692 59. Two changes related to the smtp_active_hostname option:
4694 (1) $smtp_active_hostname is now available as a variable.
4695 (2) The default for smtp_banner uses $smtp_active_hostname instead
4696 of $primary_hostname.
4698 60. The host_aton() function is supposed to be passed a string that is known
4699 to be a valid IP address. However, in the case of IPv6 addresses, it was
4700 not checking this. This is a hostage to fortune. Exim now panics and dies
4701 if the condition is not met. A case was found where this could be provoked
4702 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4703 components; fortuitously, this particular loophole had already been fixed
4704 by change 4.50/55 above.
4706 If there are any other similar loopholes, the new check in host_aton()
4707 itself should stop them being exploited. The report I received stated that
4708 data on the command line could provoke the exploit when Exim was running as
4709 exim, but did not say which command line option was involved. All I could
4710 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4711 running as the user.
4714 61. There was a buffer overflow vulnerability in the SPA authentication code
4715 (which came originally from the Samba project). I have added a test to the
4716 spa_base64_to_bits() function which I hope fixes it.
4719 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4720 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4722 63. The daemon start-up calls getloadavg() while still root for those OS that
4723 need the first call to be done as root, but it missed one case: when
4724 deliver_queue_load_max is set with deliver_drop_privilege. This is
4725 necessary for the benefit of the queue runner, because there is no re-exec
4726 when deliver_drop_privilege is set.
4728 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4729 This has been fixed.
4731 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4732 was in use, was not putting the data itself into the right store pool;
4733 consequently, it could be overwritten for a subsequent message in the same
4734 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4737 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4739 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4740 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4741 the function were treating the return as a boolean value, which happened to
4742 work because 0=false and not-0=true, but is not correct code.
4744 68. The host_aton() function was not handling scoped IPv6 addresses (those
4745 with, for example, "%eth0" on the end) correctly.
4747 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4748 negated items (that is, ~something) in unsigned ints. Some compilers
4749 apparently mutter when there is no cast.
4751 70. If an address verification called from an ACL failed, and did not produce a
4752 user-specific message (i.e. there was only a "system" message), nothing was
4753 put in $acl_verify_message. In this situation, it now puts the system
4756 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4757 session; change 4.31/43 added the unwanted input to the log line - except
4758 that it did not do this in the start of session case. It now does.
4760 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4761 This is wrong and can cause the other end to generate a synchronization
4762 error if it is another Exim or anything else that does the synchronization
4763 check. A QUIT command is no longer sent after a timeout.
4765 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4766 during host lookups.
4768 74. The defer_ok option of callout verification was not working if it was used
4769 when verifying addresses in header lines, that is, for this case:
4771 verify = header_sender/callout=defer_ok
4773 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4774 those file descriptors could be used for SMTP connections. If anything
4775 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4776 could be sent to the SMTP client, causing chaos. The daemon now opens
4777 stdin, stdout, and stderr to /dev/null when it puts itself into the
4780 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4781 log by default. The user can override this with suitable Perl magic.
4783 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4784 the log message when discard triggers, was not working for the DATA ACL or
4785 for the non-SMTP ACL.
4787 78. Error message wording change in sieve.c.
4789 79. If smtp_accept_max_per_host was set, the number of connections could be
4790 restricted to fewer than expected, because the daemon was trying to set up
4791 a new connection before checking whether the processes handling previous
4792 connections had finished. The check for completed processes is now done
4793 earlier. On busy systems, this bug wouldn't be noticed because something
4794 else would have woken the daemon, and it would have reaped the completed
4797 80. If a message was submitted locally by a user whose login name contained one
4798 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4799 It caused a spool format error. I have fixed the spool reading code. A
4800 related problem was that the "from" clause in the Received: line became
4801 illegal because of the space(s). It is now covered by ${quote_local_part.
4803 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4806 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4808 83. Patch from Sieve maintainer to fix unterminated string problem in
4809 "vacation" handling.
4811 84. Some minor changes to the Linux configuration files to help with other
4812 OS variants using glibc.
4814 85. One more patch for Sieve to update vacation handling to latest spec.
4817 ----------------------------------------------------
4818 See the note above about the 4.44 and 4.50 releases.
4819 ----------------------------------------------------
4825 1. Change 4.43/35 introduced a bug that caused file counts to be
4826 incorrectly computed when quota_filecount was set in an appendfile
4829 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4830 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4833 3. Give more explanation in the error message when the command for a transport
4834 filter fails to execute.
4836 4. There are several places where Exim runs a non-Exim command in a
4837 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4838 was being done only for the command run by the queryprogram router. It is
4839 now done for all such subprocesses. The other cases are: ${run, transport
4840 filters, and the commands run by the lmtp and pipe transports.
4842 5. Some older OS have a limit of 256 on the maximum number of file
4843 descriptors. Exim was using setrlimit() to set 1000 as a large value
4844 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4845 systems. I've change it so that if it can't get 1000, it tries for 256.
4847 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4848 was an oversight, and furthermore, ever since the addition of extra
4849 controls (e.g. 4.43/32), the checks on when to allow different forms of
4850 "control" were broken. There should now be diagnostics for all cases when a
4851 control that does not make sense is encountered.
4853 7. $recipients is now available in the predata ACL (oversight).
4855 8. Tidy the search cache before the fork to do a delivery from a message
4856 received from the command line. Otherwise the child will trigger a lookup
4857 failure and thereby defer the delivery if it tries to use (for example) a
4858 cached ldap connection that the parent has called unbind on.
4860 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4861 of $address_data from the recipient verification was clobbered by the
4862 sender verification.
4864 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4865 was its contents. (It was OK if the option was not defined at all.)
4867 11. A "Completed" log line is now written for messages that are removed from
4868 the spool by the -Mrm option.
4870 12. $host_address is now set to the target address during the checking of
4871 ignore_target_hosts.
4873 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4874 being passed; this would have caused $sender_host_name to have been used if
4875 matching the list had actually called for a host name (not very likely,
4876 since this list is usually IP addresses). A host name is now passed as
4879 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4880 code that specifies a non-restarting handler (typically sigaction() in
4881 modern systems) in an attempt to fix a rare and obscure crash bug.
4883 15. Narrowed the window for a race in the daemon that could cause it to ignore
4884 SIGCHLD signals. This is not a major problem, because they are used only to
4885 wake it up if nothing else does.
4887 16. A malformed maildirsize file could cause Exim to calculate negative values
4888 for the mailbox size or file count. Odd effects could occur as a result.
4889 The maildirsize information is now recalculated if the size or filecount
4892 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4893 support for a long time. Removed HAVE_SYS_VFS_H.
4895 18. Updated exipick to current release from John Jetmore.
4897 19. Allow an empty sender to be matched against a lookup in an address list.
4898 Previously the only cases considered were a regular expression, or an
4901 20. Exim went into a mad DNS lookup loop when doing a callout where the
4902 host was specified on the transport, if the DNS lookup yielded more than
4905 21. The RFC2047 encoding function was originally intended for short strings
4906 such as real names; it was not keeping to the 75-character limit for
4907 encoded words that the RFC imposes. It now respects the limit, and
4908 generates multiple encoded words if necessary. To be on the safe side, I
4909 have increased the buffer size for the ${rfc2047: expansion operator from
4912 22. Failure to deliver a bounce message always caused it to be frozen, even if
4913 there was an errors_to setting on the router. The errors_to setting is now
4916 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4917 canonical form (fully expanded) before being placed in
4918 $sender_host_address.
4920 24. Updated eximstats to version 1.33
4922 25. Include certificate and key file names in error message when GnuTLS fails
4923 to set them up, because the GnuTLS error message doesn't include the name
4924 of the failing file when there is a problem reading it.
4926 26. Expand error message when OpenSSL has problems setting up cert/key files.
4929 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4930 (this can affect the format of dates).
4932 28. exim_tidydb, when checking for the continued existence of a message for
4933 which it has found a message-specific retry record, was not finding
4934 messages that were in split spool directories. Consequently, it was
4935 deleting retry records that should have stayed in existence.
4937 29. eximstats updated to version 1.35
4938 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4939 1.35 - bugfix such that pie charts by volume are generated correctly
4941 30. The SPA authentication driver was not abandoning authentication and moving
4942 on to the next authenticator when an expansion was forced to fail,
4943 contradicting the general specification for all authenticators. Instead it
4944 was generating a temporary error. It now behaves as specified.
4946 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4947 (the order specifies the preference for clients). The order is now AES256,
4948 AES128, 3DES, ARCFOUR128.
4950 31. Small patch to Sieve code - explicitly set From: when generating an
4953 32. Exim crashed if a remote delivery caused a very long error message to be
4954 recorded - for instance if somebody sent an entire SpamAssassin report back
4955 as a large number of 550 error lines. This bug was coincidentally fixed by
4956 increasing the size of one of Exim's internal buffers (big_buffer) that
4957 happened as part of the Exiscan merge. However, to be on the safe side, I
4958 have made the code more robust (and fixed the comments that describe what
4961 33. Some experimental protocols are using DNS PTR records for new purposes. The
4962 keys for these records are domain names, not reversed IP addresses. The
4963 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4964 leaves it alone. Component reversal etc. now happens only for IP addresses.
4967 34. The host_aton() function is supposed to be passed a string that is known
4968 to be a valid IP address. However, in the case of IPv6 addresses, it was
4969 not checking this. This is a hostage to fortune. Exim now panics and dies
4970 if the condition is not met. A case was found where this could be provoked
4971 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4972 components; fortuitously, this particular loophole had already been fixed
4973 by change 4.50/55 or 4.44/33 above.
4975 If there are any other similar loopholes, the new check in host_aton()
4976 itself should stop them being exploited. The report I received stated that
4977 data on the command line could provoke the exploit when Exim was running as
4978 exim, but did not say which command line option was involved. All I could
4979 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4980 running as the user.
4983 35. There was a buffer overflow vulnerability in the SPA authentication code
4984 (which came originally from the Samba project). I have added a test to the
4985 spa_base64_to_bits() function which I hope fixes it.
4988 36. The daemon start-up calls getloadavg() while still root for those OS that
4989 need the first call to be done as root, but it missed one case: when
4990 deliver_queue_load_max is set with deliver_drop_privilege. This is
4991 necessary for the benefit of the queue runner, because there is no re-exec
4992 when deliver_drop_privilege is set.
4994 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4995 was in use, was not putting the data itself into the right store pool;
4996 consequently, it could be overwritten for a subsequent message in the same
4997 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5000 38. Sometimes the final signoff response after QUIT could fail to get
5001 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5002 < 0 before doing a fflush(). This bug looks as though it goes back to the
5003 introduction of TLS in release 3.20, but "sometimes" must have been rare
5004 because the tests only now provoked it.
5010 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5011 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5012 Now it returns an integer. A number of calls were still expecting a T/F
5013 return. Fortuitously, in all cases, the tests worked in OK situations,
5014 which is the norm. However, things would have gone wrong on any write
5015 failures on the smtp file descriptor. This function is used when sending
5016 messages over SMTP and also when doing verify callouts.
5018 2. When Exim is called to do synchronous delivery of a locally submitted
5019 message (the -odf or -odi options), it no longer closes stderr before doing
5022 3. Implemented the mua_wrapper option.
5024 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5026 5. Implemented the functions header_remove(), header_testname(),
5027 header_add_at_position(), and receive_remove_recipient(), and exported them
5030 6. If an ACL "warn" statement specified the addition of headers, Exim already
5031 inserted X-ACL-Warn: at the start if there was no header name. However, it
5032 was not making this test for the second and subsequent header lines if
5033 there were newlines in the string. This meant that an invalid header could
5034 be inserted if Exim was badly configured.
5036 7. Allow an ACL "warn" statement to add header lines at the start or after all
5037 the Received: headers, as well as at the end.
5039 8. Added the rcpt_4xx retry error code.
5041 9. Added postmaster_mailfrom=xxx to callout verification option.
5043 10. Added mailfrom=xxxx to the callout verification option, for verify=
5046 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5047 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5049 12. Inserted some casts to stop certain compilers warning when using pointer
5050 differences as field lengths or precisions in printf-type calls (mostly
5051 affecting debugging statements).
5053 13. Added optional readline() support for -be (dynamically loaded).
5055 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5056 same clock tick as a message's arrival, so that its received time was the
5057 same as the "first fail" time on the retry record, and that message
5058 remained on the queue past the ultimate address timeout, every queue runner
5059 would try a delivery (because it was past the ultimate address timeout) but
5060 after another failure, the ultimate address timeout, which should have then
5061 bounced the address, did not kick in. This was a "< instead of <=" error;
5062 in most cases the first failure would have been in the next clock tick
5063 after the received time, and all would be well.
5065 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5066 being recognized when the domain list was tested by the match_domain
5067 condition in an expansion string.
5069 16. Added the ${str2b64: operator.
5071 17. Exim was always calling setrlimit() to set a large limit for the number of
5072 processes, without checking whether the existing limit was already
5073 adequate. (It did check for the limit on file descriptors.) Furthermore,
5074 errors from getrlimit() and setrlimit() were being ignored. Now they are
5075 logged to the main and panic logs, but Exim does carry on, to try to do its
5076 job under whatever limits there are.
5078 18. Imported PCRE 5.0.
5080 19. Trivial typo in log message " temporarily refused connection" (the leading
5083 20. If the log selector return_path_on_delivery was set and an address was
5084 redirected to /dev/null, the delivery process crashed because it assumed
5085 that a return path would always be set for a "successful" delivery. In this
5086 case, the whole delivery is bypassed as an optimization, and therefore no
5089 21. Internal re-arrangement: the function for sending a challenge and reading
5090 a response while authentication was assuming a zero-terminated challenge
5091 string. It's now changed to take a pointer and a length, to allow for
5092 binary data in such strings.
5094 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5096 23. Exim was not respecting finduser_retries when seeking the login of the
5097 uid under which it was called; it was always trying 10 times. (The default
5098 setting of finduser_retries is zero.) Also, it was sleeping after the final
5099 failure, which is pointless.
5101 24. Implemented tls_on_connect_ports.
5103 25. Implemented acl_smtp_predata.
5105 26. If the domain in control=submission is set empty, Exim assumes that the
5106 authenticated id is a complete email address when it generates From: or
5107 Sender: header lines.
5109 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5110 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5111 chown and chgrp in /bin and hostname in /usr/bin.
5113 28. Exim was keeping the "process log" file open after each use, just as it
5114 does for the main log. This opens the possibility of it remaining open for
5115 long periods when the USR1 signal hits a daemon. Occasional processlog
5116 errors were reported, that could have been caused by this. Anyway, it seems
5117 much more sensible not to leave this file open at all, so that is what now
5120 29. The long-running daemon process does not normally write to the log once it
5121 has entered its main loop, and it closes the log before doing so. This is
5122 so that log files can straightforwardly be renamed and moved. However,
5123 there are a couple of unusual error situations where the daemon does write
5124 log entries, and I had neglected to close the log afterwards.
5126 30. The text of an SMTP error response that was received during a remote
5127 delivery was being truncated at 512 bytes. This is too short for some of
5128 the long messages that one sometimes sees. I've increased the limit to
5131 31. It is now possible to make retry rules that apply only when a message has a
5132 specific sender, in particular, an empty sender.
5134 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5135 it possible to be selective about when SMTP synchronization is enforced.
5137 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5139 32. Implemented hosts_connection_nolog.
5141 33. Added an ACL for QUIT.
5143 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5146 35. Added mailbox_size and mailbox_filecount to appendfile.
5148 36. Added control = no_multiline_responses to ACLs.
5150 37. There was a bug in the logic of the code that waits for the clock to tick
5151 in the case where the clock went backwards by a substantial amount such
5152 that the microsecond fraction of "now" was more than the microsecond
5153 fraction of "then" (but the whole seconds number was less).
5155 38. Added support for the libradius Radius client library this is found on
5156 FreeBSD (previously only the radiusclient library was supported).
5162 1. When certain lookups returned multiple values in the form name=value, the
5163 quoting of the values was not always being done properly. Specifically:
5164 (a) If the value started with a double quote, but contained no whitespace,
5166 (b) If the value contained whitespace other than a space character (i.e.
5167 tabs or newlines or carriage returns) it was not quoted.
5168 This fix has been applied to the mysql and pgsql lookups by writing a
5169 separate quoting function and calling it from the lookup code. The fix
5170 should probably also be applied to nisplus, ibase and oracle lookups, but
5171 since I cannot test any of those, I have not disturbed their existing code.
5173 2. A hit in the callout cache for a specific address caused a log line with no
5174 reason for rejecting RCPT. Now it says "Previous (cached) callout
5175 verification failure".
5177 3. There was an off-by-one bug in the queryprogram router. An over-long
5178 return line was truncated at 256 instead of 255 characters, thereby
5179 overflowing its buffer with the terminating zero. As well as fixing this, I
5180 have increased the buffer size to 1024 (and made a note to document this).
5182 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5183 when Exim is waiting for an SMTP response from a remote server, Exim
5184 restarts its select() call on the socket, thereby resetting its timeout.
5185 This is not a problem when such interrupts are rare. Somebody set up a cron
5186 job to run exiwhat every 2 minutes, which is less than the normal select()
5187 timeout (5 or 10 minutes). This meant that the select() timeout never
5188 kicked in because it was always reset. I have fixed this by comparing the
5189 time when an interrupt arrives with the time at the start of the first call
5190 to select(). If more time than the timeout has elapsed, the interrupt is
5191 treated as a timeout.
5193 5. Some internal re-factoring in preparation for the addition of Sieve
5194 extensions (by MH). In particular, the "personal" test is moved to a
5195 separate function, and given an option for scanning Cc: and Bcc: (which is
5196 not set for Exim filters).
5198 6. When Exim created an email address using the login of the caller as the
5199 local part (e.g. when creating a From: or Sender: header line), it was not
5200 quoting the local part when it contained special characters such as @.
5202 7. Installed new OpenBSD configuration files.
5204 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5205 try to make them clearer.
5207 9. Callout options, other than the timeout value, were being ignored when
5208 verifying sender addresses in header lines. For example, when using
5210 verify = header_sender/callout=no_cache
5212 the cache was (incorrectly) being used.
5214 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5215 only the Cygwin environment.
5217 11. When return_path_on_delivery was set as a log selector, if different remote
5218 addresses in the same message used different return paths and parallel
5219 remote delivery occurred, the wrong values would sometimes be logged.
5220 (Whenever a remote delivery process finished, the return path value from
5221 the most recently started remote delivery process was logged.)
5223 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5224 lines when AUTH and/or TLS are in use. This is the "received protocol"
5225 field. Exim used to use "asmtp" for authenticated SMTP, without any
5226 indication (in the protocol name) for TLS use. Now it follows the RFC and
5227 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5228 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5229 names appear in log lines as well as in Received: header lines.
5231 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5232 extensions, and comparison tests, and to fix some bugs.
5234 14. Changes to the "personal" filter test:
5236 (1) The test was buggy in that it was just doing the equivalent of
5237 "contains" tests on header lines. For example, if a user's address was
5238 anne@some.where, the "personal" test would incorrectly be true for
5240 To: susanne@some.where
5242 This test is now done by extracting each address from the header in turn,
5243 and checking the entire address. Other tests that are part of "personal"
5244 are now done using regular expressions (for example, to check local parts
5245 of addresses in From: header lines).
5247 (2) The list of non-personal local parts in From: addresses has been
5248 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5249 taken from the Sieve specification recommendations.
5251 (3) If the message contains any header line starting with "List-" it is
5252 treated as non-personal.
5254 (4) The test for "circular" in the Subject: header line has been removed
5255 because it now seems ill-conceived.
5257 15. Minor typos in src/EDITME comments corrected.
5259 16. Installed latest exipick from John Jetmore.
5261 17. If headers_add on a router specified a text string that was too long for
5262 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5263 of string_sprintf() is now avoided.
5265 18. $message_body_size was not set (it was always zero) when running the DATA
5266 ACL and the local_scan() function.
5268 19. For the "mail" command in an Exim filter, no default was being set for
5269 the once_repeat time, causing a random time value to be used if "once" was
5270 specified. (If the value happened to be <= 0, no repeat happened.) The
5271 default is now 0s, meaning "never repeat". The "vacation" command was OK
5272 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5273 (I found it when inspecting the code).
5275 20. There is now an overall timeout for performing a callout verification. It
5276 defaults to 4 times the callout timeout, which applies to individual SMTP
5277 commands during the callout. The overall timeout applies when there is more
5278 than one host that can be tried. The timeout is checked before trying the
5279 next host. This prevents very long delays if there are a large number of
5280 hosts and all are timing out (e.g. when the network connections are timing
5281 out). The value of the overall timeout can be changed by specifying an
5282 additional sub-option for "callout", called "maxwait". For example:
5284 verify = sender/callout=5s,maxwait=20s
5286 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5287 to the end before writing, but this should make it even safer).
5289 22. Exim was forgetting that it had advertised PIPELINING for the second and
5290 subsequent messages on an SMTP connection. It was also not resetting its
5291 memory on STARTTLS and an internal HELO.
5293 23. When Exim logs an SMTP synchronization error within a session, it now
5294 records whether PIPELINING has been advertised or not.
5296 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5297 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5298 rather than long int.
5300 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5302 26. Added the never_mail option to autoreply.
5308 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5309 crash if the getsockname() call failed; this can happen if a connection is
5310 closed very soon after it is established. The problem was simply in the
5311 order in which certain operations were done, causing Exim to try to write
5312 to the SMTP stream before it had set up the file descriptor. The bug has
5313 been fixed by making things happen in the correct order.
5319 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5320 before the connection was closed, thus losing the rejection response.
5322 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5323 some early Solaris releases, but causes trouble in current releases where
5324 socklen_t is defined.
5326 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5329 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5332 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5333 that releasing the top of it at the end releases what was used for sub-
5334 expansions (unless the block got too big). However, discard this block if
5335 the first thing is a variable or header, so that we can use its block when
5336 it is dynamic (useful for very large $message_headers, for example).
5338 6. Lookups now cache *every* query, not just the most recent. A new, separate
5339 store pool is used for this. It can be recovered when all lookup caches are
5340 flushed. Lookups now release memory at the end of their result strings.
5341 This has involved some general refactoring of the lookup sources.
5343 7. Some code has been added to the store_xxx() functions to reduce the amount
5344 of flapping under certain conditions.
5346 8. log_incoming_interface used to affect only the <= reception log lines. Now
5347 it causes the local interface and port to be added to several more SMTP log
5348 lines, for example "SMTP connection from", and rejection lines.
5350 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5352 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5354 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5355 could be overwritten at the end of the current message (or the start of a
5356 new message if it was set in a HELO ACL). The value is now preserved for
5357 the duration of the SMTP connection.
5359 12. If a transport had a headers_rewrite setting, and a matching header line
5360 contained an unqualified address, that address was qualified, even if it
5361 did not match any rewriting rules. The underlying bug was that the values
5362 of the flags that permit the existence of unqualified sender and recipient
5363 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5364 non-local messages, and by -bnq for local messages) were not being
5365 preserved with the message after it was received.
5367 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5368 "next input=" as part of the text comprising the host identity instead of
5369 the correct text. The code was using the same buffer for two different
5370 strings. However, depending on which order the printing function evaluated
5371 its arguments, the bug did not always show up. Under Linux, for example, my
5372 test suite worked just fine.
5374 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5375 doesn't work with some older versions of Perl. It has been changed to "my",
5376 which in any case is probably the better facility to use.
5378 15. A really picky compiler found some instances of statements for creating
5379 error messages that either had too many or two few arguments for the format
5382 16. The size of the buffer for calls to the DNS resolver has been increased
5383 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5384 for addresses that have a lot of PTR records. This alleviates a problem; it
5385 does not fully solve it.
5387 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5388 buffer now truncates the list and logs the incident, which is the same
5389 action as happens when Exim is looking up a host name and its aliases.
5390 Previously in this situation something unpredictable would happen;
5391 sometimes it was "internal error: store_reset failed".
5393 18. If a server dropped the connection unexpectedly when an Exim client was
5394 using GnuTLS and trying to read a response, the client delivery process
5395 crashed while trying to generate an error log message.
5397 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5398 string, for example:
5400 warn message = H1: something\nH2: something
5402 the text was added as a single header line from Exim's point of view
5403 though it ended up OK in the delivered message. However, searching for the
5404 second and subsequent header lines using $h_h2: did not work. This has been
5405 fixed. Similarly, if a system filter added multiple headers in this way,
5406 the routers could not see them.
5408 20. Expanded the error message when iplsearch is called with an invalid key to
5409 suggest using net-iplsearch in a host list.
5411 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5412 ACLs are no longer actually imposed (and a message to that effect is
5415 22. If a "gecos" field in a passwd entry contained escaped characters, in
5416 particular, if it contained a \" sequence, Exim got it wrong when building
5417 a From: or a Sender: header from that name. A second bug also caused
5418 incorrect handling when an unquoted " was present following a character
5419 that needed quoting.
5421 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5422 was not being matched caselessly.
5424 24. Arranged for all hyphens in the exim.8 source to be escaped with
5427 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5428 properly. Recipient callout cache records were still being keyed to include
5429 the sender, even when use_sender was set false. This led to far more
5430 callouts that were necessary. The sender is no longer included in the key
5431 when use_sender is false.
5433 26. Added "control = submission" modifier to ACLs.
5435 27. Added the ${base62d: operator to decode base 62 numbers.
5437 28. dnsdb lookups can now access SRV records.
5439 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5440 the configuration file.
5442 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5443 (-v) mode. This makes the output for a verbose queue run more intelligible.
5445 31. Added a use_postmaster feature to recipient callouts.
5447 32. Added the $body_zerocount variable, containing the number of binary zero
5448 bytes in the message body.
5450 33. The time of last modification of the "new" subdirectory is now used as the
5451 "mailbox time last read" when there is a quota error for a maildir
5454 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5456 35. Added +ignore_unknown as a special item in host lists.
5458 36. Code for decoding IPv6 addresses in host lists is now included, even if
5459 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5460 address was recognized as an IP address, but was then not correctly decoded
5461 into binary, causing unexpected and incorrect effects when compared with
5468 1. Very minor rewording of debugging text in manualroute to say "list of
5469 hosts" instead of "hostlist".
5471 2. If verify=header_syntax was set, and a header line with an unqualified
5472 address (no domain) and a large number of spaces between the end of the
5473 name and the colon was received, the reception process suffered a buffer
5474 overflow, and (when I tested it) crashed. This was caused by some obsolete
5475 code that should have been removed. The fix is to remove it!
5477 3. When running in the test harness, delay a bit after writing a bounce
5478 message to get a bit more predictability in the log output.
5480 4. Added a call to search_tidyup() just before forking a reception process. In
5481 theory, someone could use a lookup in the expansion of smtp_accept_max_
5482 per_host which, without the tidyup, could leave open a database connection.
5484 5. Added the variables $recipient_data and $sender_data which get set from a
5485 lookup success in an ACL "recipients" or "senders" condition, or a router
5486 "senders" option, similar to $domain_data and $local_part_data.
5488 6. Moved the writing of debug_print from before to after the "senders" test
5491 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5492 problems for message scanning, either using a data ACL, or using
5493 local_scan() because the Received: header was not generated till after they
5494 were called (in order to set the time as the time of reception completion).
5495 I have revised the way this works. The header is now generated after the
5496 body is received, but before the ACL or local_scan() are called. After they
5497 are run, the timestamp in the header is updated.
5503 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5504 before starting a queue runner without re-exec. This happened only when
5505 deliver_drop_privilege was set or when the Exim user was set to root. The
5506 effect of the bug was that timeouts during subsequent deliveries caused
5507 crashes instead of being properly handled. The handler is now left at its
5508 default (and expected) setting.
5510 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5511 message, again when deliver_drop_privilege is set or Exim is run as root.
5512 The bug described in (1) was not present in this case, but the tidying up
5513 of the other signals was missing. I have made the two cases consistent.
5515 3. The ignore_target_hosts setting on a manualroute router was being ignored
5516 for hosts that were looked up using the /MX notation.
5518 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5521 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5522 operated on the sender address. After changing the $sender_address to <>
5523 for the sender address verify, Exim was re-instated it as the original
5524 (before rewriting) address, but remembering that it had rewritten it, so it
5525 wasn't rewriting it again. This bug also had the effect of breaking the
5526 sender address verification caching when the sender address was rewritten.
5528 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5529 This has been changed so that if the ip literal address matches
5530 ignore_target_hosts, the router declines.
5532 7. Added expansion conditions match_domain, match_address, and match_local_
5533 part (NOT match_host).
5535 8. The placeholder for the Received: header didn't have a length field set.
5537 9. Added code to Exim itself and to exim_lock to test for a specific race
5538 condition that could lead to file corruption when using MBX delivery. The
5539 issue is with the lockfile that is created in /tmp. If this file is removed
5540 after a process has opened it but before that process has acquired a lock,
5541 there is the potential for a second process to recreate the file and also
5542 acquire a lock. This could lead to two Exim processes writing to the file
5543 at the same time. The added code performs the same test as UW imapd; it
5544 checks after acquiring the lock that its file descriptor still refers to
5545 the same named file.
5547 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5548 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5551 11. Added the smtp_active_hostname option. If used, this will typically be made
5552 to depend on the incoming interface address. Because $interface_address is
5553 not set up until the daemon has forked a reception process, error responses
5554 that can happen earlier (such as "too many connections") no longer contain
5557 12. If an expansion in a condition on a "warn" statement fails because a lookup
5558 defers, the "warn" statement is abandoned, and the next ACL statement is
5559 processed. Previously this caused the whole ACL to be aborted.
5561 13. Added the iplsearch lookup type.
5563 14. Added ident_timeout as a log selector.
5565 15. Added tls_certificate_verified as a log selector.
5567 16. Added a global option tls_require_ciphers (compare the smtp transport
5568 option of the same name). This controls incoming TLS connections.
5570 17. I finally figured out how to make tls_require_ciphers do a similar thing
5571 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5572 before starting the TLS session.
5574 18. Tabs are now shown as \t in -bP output.
5576 19. If the log selector return_path_on_delivery was set, Exim crashed when
5577 bouncing a message because it had too many Received: header lines.
5579 20. If two routers both had headers_remove settings, and the first one included
5580 a superfluous trailing colon, the final name in the first list and the
5581 first name in the second list were incorrectly joined into one item (with a
5582 colon in the middle).
5588 1. Added -C and -D options to the exinext utility, mainly to make it easier
5589 to include in the automated testing, but these could be helpful when
5590 multiple configurations are in use.
5592 2. The exinext utility was not formatting the output nicely when there was
5593 an alternate port involved in the retry record key, nor when there was a
5594 message id as well (for retries that were specific to a specific message
5595 and a specific host). It was also confused by IPv6 addresses, because of
5596 the additional colons they contain. I have fixed the IPv4 problem, and
5597 patched it up to do a reasonable job for IPv6.
5599 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5600 delivery, the log line now contains "pipelined" if PIPELINING was used.
5602 4. An SMTP transport process used to panic and die if the bind() call to set
5603 an explicit outgoing interface failed. This has been changed; it is now
5604 treated in the same way as a connect() failure.
5606 5. A reference to $sender_host_name in the part of a conditional expansion
5607 that was being skipped was still causing a DNS lookup. This no longer
5610 6. The def: expansion condition was not recognizing references to header lines
5611 that used bh_ and bheader_.
5613 7. Added the _cache feature to named lists.
5615 8. The code for checking quota_filecount in the appendfile transport was
5616 allowing one more file than it should have been.
5618 9. For compatibility with Sendmail, the command line option
5626 and sets the incoming protocol and host name (for trusted callers). The
5627 host name and its colon can be omitted when only the protocol is to be set.
5628 Note the Exim already has two private options, -pd and -ps, that refer to
5629 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5630 "s", but I don't think that's a major issue.
5632 10. A number of refactoring changes to the code, none of which should affect
5635 (a) The number of logging options was getting close to filling up the
5636 32-bit word that was used as a bit map. I have split them into two classes:
5637 those that are passed in the argument to log_write(), and those that are
5638 only ever tested independently outside of that function. These are now in
5639 separate 32-bit words, so there is plenty of room for expansion again.
5640 There is no change in the user interface or the logging behaviour.
5642 (b) When building, for example, log lines, the code previously used a
5643 macro that called string_cat() twice, in order to add two strings. This is
5644 not really sufficiently general. Furthermore, there was one instance where
5645 it was actually wrong because one of the argument was used twice, and in
5646 one call a function was used. (As it happened, calling the function twice
5647 did not affect the overall behaviour.) The macro has been replaced by a
5648 function that can join an arbitrary number of extra strings onto a growing
5651 (c) The code for expansion conditions now uses a table and a binary chop
5652 instead of a serial search (which was left over from when there were very
5653 few conditions). Also, it now recognizes conditions like "pam" even when
5654 the relevant support is not compiled in: a suitably worded error message is
5655 given if an attempt is made to use such a condition.
5657 11. Added ${time_interval:xxxxx}.
5659 12. A bug was causing one of the ddress fields not to be passed back correctly
5660 from remote delivery subprocesses. The field in question was not being
5661 subsequently used, so this caused to problems in practice.
5663 13. Added new log selectors queue_time and deliver_time.
5665 14. Might have fixed a bug in maildirsizefile handling that threw up
5666 "unexpected character" debug warnings, and recalculated the data
5667 unnecessarily. In any case, I expanded the warning message to give more
5670 15. Added the message "Restricted characters in address" to the statements in
5671 the default ACL that block characters like @ and % in local parts.
5673 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5674 Three changes have been made:
5676 (a) There was a serious bug; a negative response to MAIL caused the whole
5677 recipient domain to be cached as invalid, thereby blocking all messages
5678 to all local parts at the same domain, from all senders. This bug has
5679 been fixed. The domain is no longer cached after a negative response to
5680 MAIL if the sender used is not empty.
5682 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5685 (c) A new callout option, "use_sender" has been added for people who want
5686 the modified behaviour.
5692 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5695 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5696 indeed breaks things for older releases.
5698 3. Added additional logging to the case where there is a problem reading data
5699 from a filter that is running in a subprocess using a pipe, in order to
5700 try to track down a specific problem.
5702 4. Testing facility fudge: when running in the test harness and attempting
5703 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5704 getting "No route to host". Convert this to a timeout.
5706 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5709 6. Some OS don't have socklen_t but use size_t instead. This affects the
5710 fifth argument of getsockopt() amongst other things. This is now
5711 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5712 can be set for individual OS. I have set it for SunOS5, OSF1, and
5713 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5714 some earlier ones do not.
5716 7. Change 4.30/15 was not doing the test caselessly.
5718 8. The standard form for an IPv6 address literal was being rejected by address
5719 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5720 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5721 this, as well as the form without the "IPv6" on the front (but only when
5722 address literals are enabled, of course).
5724 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5726 10. Exim crashed if a message with an empty sender address specified by -f
5727 encountered a router with an errors_to setting. This could be provoked only
5728 by a command such as
5732 where an empty string was supplied; "<>" did not hit this bug.
5734 11. Installed PCRE release 4.5.
5736 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5737 remained set. It is now erased.
5739 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5740 times from message ids (which are base 36 rather than the normal 62).
5742 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5743 were being counted as actual protocol errors, and logged if the log
5744 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5745 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5746 and DATA following a set of rejected RCPTs do not count as protocol errors.
5747 In other words, Exim assumes they were pipelined, though this may not
5748 actually be the case. Of course, in all cases the client gets an
5749 appropriate error code.
5751 15. If a lookup fails in an ACL condition, a message about the failure may
5752 be available; it is used if testing the ACL cannot continue, because most
5753 such messages specify what the cause of the deferral is. However, some
5754 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5755 that caused an old message to be retained and used if a later statement
5756 caused a defer, replacing the real cause of the deferral.
5758 16. If an IP address had so many PTR records that the DNS lookup buffer
5759 was not large enough to hold them, Exim could crash while trying to process
5760 the truncated data. It now detects and logs this case.
5762 17. Further to 4.21/58, another change has been made: if (and only if) the
5763 first line of a message (the first header line) ends with CRLF, a bare LF
5764 in a subsequent header line has a space inserted after it, so as not to
5765 terminate the header.
5767 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5768 unnecessarily, used atoi() instead of strtol(), and didn't check the
5769 termination when getting file sizes from file names by regex.
5771 19. Completely re-implemented the support for maildirsize files, in the light
5772 of a number of problems with the previous contributed implementation
5773 (4.30/29). In particular:
5775 . If the quota is zero, the maildirsize file is maintained, but no quota is
5778 . If the maildir directory does not exist, it is created before any attempt
5779 to write a maildirsize file.
5781 . The quota value in the file is just a cache; if the quota is changed in
5782 the transport, the new value overrides.
5784 . A regular expression is available for excluding directories from the
5787 20. The autoreply transport checks the characters in options that define the
5788 message's headers; it allows continued headers, but it was checking with
5789 isspace() after an embedded newline instead of explicitly looking for a
5792 21. If all the "regular" hosts to which an address was routed had passed their
5793 expiry times, and had not reached their retry times, the address was
5794 bounced, even if fallback hosts were defined. Now Exim should go on to try
5797 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5798 equivalent code in the SMTP transport. Some hosts send humungous responses
5799 to HELO/EHLO, more than 1024 it seems.
5801 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5802 gives compiler warnings in some environments. I've now done it "properly",
5805 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5806 (because of problems with the built-in one) was declared to return uschar *
5807 instead of char *, causing compiler failure.
5809 25. Fixed a file descriptor leak when processing alias/forward files.
5811 26. Fixed a minor format string issue in dbfn.c.
5813 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5815 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5816 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5819 29. When a sender address is verified, it is cached, to save repeating the test
5820 when there is more than one recipient in a message. However, when the
5821 verification involves a callout, it is possible for different callout
5822 options to be set for different recipients. It is too complicated to keep
5823 track of this in the cache, so now Exim always runs a verification when a
5824 callout is required, relying on the callout cache for the optimization.
5825 The overhead is duplication of the address routing, but this should not be
5828 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5829 to be verified with callout=postmaster, and the main callout worked but the
5830 postmaster check failed, the verification correctly failed. However, if a
5831 subsequent RCPT command asked for sender verification *without* the
5832 postmaster check, incorrect caching caused this verification also to fail,
5835 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5836 it was not caching the DNS options (qualify_single, search_parents) that
5837 were used when the lookup failed. A subsequent lookup with different
5838 options therefore always gave the same answer, though there were cases
5839 where it should not have. (Example: a "domains = !$mx_any" option on a
5840 dnslookup router: the "domains" option is always processed without any
5841 widening, but the router might have qualify_single set.) Now Exim uses the
5842 cached value only when the same options are set.
5844 32. Added John Jetmore's "exipick" utility to the distribution.
5846 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5847 than a timeout (e.g. a certificate is required, and is not provided), an
5848 Exim server now closes the connection immediately. Previously it waited for
5849 the client to close - but if the client is SSL, it seems that they each
5850 wait for each other, leading to a delay before one of them times out.
5852 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5853 maintained 0.8.x compatibility because I don't think many are using it, and
5854 it is clearly obsolete.
5856 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5859 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5860 client certificate was expired. A simple patch fixes this, though I don't
5861 understand the full logic of why the verify callback is called multiple
5864 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5865 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5866 which causes problems with some clients (such as the Certicom SSL Plus
5867 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5868 disables the coutermeasure allowing Eudora to connect."
5870 38. Exim was not checking that a write() to a log file succeeded. This could
5871 lead to Bad Things if a log got too big, in particular if it hit a file
5872 size limit. Exim now panics and dies if it cannot write to a log file, just
5873 as it does if it cannot open a log file.
5875 39. Modified OS/Makefile-Linux so that it now contains
5877 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5879 The two -D definitions ensure that Exim is compiled with large file
5880 support, which makes it possible to handle log files that are bigger than
5883 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5884 instance) a domain was checked against a named list that involved a lookup,
5885 causing $domain_data to be set, then another domain was checked against the
5886 same list, then the first domain was re-checked, the value of $domain_data
5887 after the final check could be wrong. In particular, if the second check
5888 failed, it could be set empty. This bug probably also applied to
5891 41. The strip_trailing_dot option was not being applied to the address given
5892 with the -f command-line option.
5894 42. The code for reading a message's header from the spool was incrementing
5895 $received_count, but never initializing it. This meant that the value was
5896 incorrect (doubled) while delivering a message in the same process in which
5897 it was received. In the most common configuration of Exim, this never
5898 happens - a fresh exec is done - but it can happen when
5899 deliver_drop_privilege is set.
5901 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5902 it now includes up to 150 characters of the unexpected data in the log
5905 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5906 and building data strings. The size of both of these buffers was 10 000
5907 bytes - far larger than anybody would *ever* want, thought I. Needless to
5908 say, somebody hit the limit. I have increased the maximum line length to
5909 20 000 and the maximum data length of concatenated lines to 100 000. I have
5910 also fixed two bugs, because there was no checking on these buffers. Tsk,
5911 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5912 buffer is too small.
5914 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5915 lsearch lookups. Now it does.
5917 46. When parsing a route_list item in a manualroute router, a fixed-length
5918 buffer was used for the list of hosts. I made this 1024 bytes long,
5919 thinking that nobody would ever have a list of hosts that long. Wrong.
5920 Somebody had a whole pile of complicated expansion conditions, and the
5921 string was silently truncated, leading to an expansion error. It turns out
5922 that it is easier to change to an unlimited length (owing to other changes
5923 that have happened since this code was originally written) than to build
5924 structure for giving a limitation error. The length of the item that
5925 expands into the list of hosts is now unlimited.
5927 47. The lsearch lookup could not handle data where the length of text line was
5928 more than 4095 characters. Such lines were truncated, leading to shortened
5929 data being returned. It should now handle lines of any length.
5931 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5932 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5935 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5936 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5937 no longer happens in comment lines. A list of replacements is now placed
5938 at the head of all of the source files, except those whose only change is
5939 to replace PERL_COMMAND in the very first #! line.
5941 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5942 messages on the queue, with a bottom-up merge sort, using code contributed
5943 by Michael Haardt. This should make operations like -bp somewhat faster on
5944 large queues. It won't affect queue runners, except when queue_run_in_order
5947 51. Installed eximstats 1.31 in the distribution.
5949 52. Added support for SRV lookups to the dnslookup router.
5951 53. If an ACL referred to $message_body or $message_body_end, the value was not
5952 reset for any messages that followed in the same SMTP session.
5954 54. The store-handling optimization for building very long strings was not
5955 differentiating between the different store pools. I don't think this
5956 actually made any difference in practice, but I've tidied it.
5958 55. While running the routers to verify a sender address, $sender_address
5959 was still set to the sender address. This is wrong, because when routing to
5960 send a bounce to the sender, it would be empty. Therefore, I have changed
5961 it so that, while verifying a sender address, $sender_address is set to <>.
5962 (There is no change to what happens when verifying a recipient address.)
5964 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5965 target A or AAAA records (if not already returned) without resetting the
5966 qualify_single or search_parents options of the DNS resolver. These are
5967 inappropriate in this case because the targets of MX and SRV records must
5968 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5969 target that, when qualified, matched something in the local domain. These
5970 two options are now turned off when doing these lookups.
5972 57. It seems that at least some releases of Reiserfs (which does not have the
5973 concept of a fixed number of inodes) returns zero and not -1 for the
5974 number of available inodes. This interacted badly with check_spool_inodes,
5975 which assumed that -1 was the "no such thing" setting. What I have done is
5976 to check that the total number of inodes is greater than zero before doing
5977 the test of how many are available.
5979 58. When a "warn" ACL statement has a log_message modifier, the message is
5980 remembered, and not repeated. This is to avoid a lot of repetition when a
5981 message has many recipients that cause the same warning to be written.
5982 However, Exim was preserving the list of already written lines for an
5983 entire SMTP session, which doesn't seem right. The memory is now reset if a
5984 new message is started.
5986 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5987 debugging output unless log_rewrite was also set.
5989 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5990 of (int)(handle) when we know that handle contains (void *)(-1).
5992 61. The Exim daemon panic-logs an error return when it closes the incoming
5993 connection. However "connection reset by peer" seems to be common, and
5994 isn't really an error worthy of noting specially, so that particular error
5997 62. When Exim is trying to find all the local interfaces, it used to panic and
5998 die if the ioctl to get the interface flags failed. However, it seems that
5999 on at least one OS (Solaris 9) it is possible to have an interface that is
6000 included in the list of interfaces, but for which you get a failure error
6001 for this call. This happens when the interface is not "plumbed" into a
6002 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6003 failure of the "get flags" call assumes that the interface is down.
6005 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6006 makes life easier for people who are doing arithmetic on fields extracted
6007 from dates, where you often get leading zeros that should not be
6008 interpreted as octal.
6010 64. Added qualify_domain to the redirect router, to override the global
6013 65. If a pathologically long header line contained very many addresses (the
6014 report of this problem mentioned 10 000) and each of them was rewritten,
6015 Exim could use up a very large amount of memory. (It kept on making new
6016 copies of the header line as it rewrote, and never released the old ones.)
6017 At the expense of a bit more processing, the header rewriting function has
6018 been changed so that it no longer eats memory in this way.
6020 66. The generation of the Received: header has been moved from the time that a
6021 message starts to be received, to the time that it finishes. The timestamp
6022 in the Received: header should now be very close to that of the <= log
6023 line. There are two side-effects of this change:
6025 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6026 logged header lines no longer include the local Received: line, because
6027 it has not yet been created. The same applies to a copy of the message
6028 that is returned to a non-SMTP sender when a message is rejected.
6030 (b) When a filter file is tested using -bf, no additional Received: header
6031 is added to the test message. After some thought, I decided that this
6034 This change does not affect the value of $received_for. It is still set
6035 after address rewriting, but before local_scan() is called.
6037 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6039 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6040 gave an unhelpful panic error message, and a defer error. I have managed to
6041 change this behaviour so that it now rejects any supplied certificate,
6042 which seems right, as the list of acceptable certificates is empty.
6044 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6045 gave an unhelpful defer error. I have not managed to make this reject any
6046 supplied certificates, but the error message it gives is "no certificate
6047 supplied", which is not helpful.
6049 70. exigrep's output now also includes lines that are not associated with any
6050 message, but which match the given pattern. Implemented by a patch from
6051 Martin Sluka, which also tidied up the Perl a bit.
6053 71. Recipient callout verification, like sender verification, was using <> in
6054 the MAIL FROM command. This isn't really the right thing, since the actual
6055 sender may affect whether the remote host accepts the recipient or not. I
6056 have changed it to use the actual sender in the callout; this means that
6057 the cache record is now keyed on a recipient/sender pair, not just the
6058 recipient address. There doesn't seem to be a real danger of callout loops,
6059 since a callout by the remote host to check the sender would use <>.
6060 [SEE ABOVE: changed after hitting problems.]
6062 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6063 temporary errors. However, in the case of such a code being given after
6064 the end of a data transmission (i.e. after ".") Exim was failing to write
6065 a retry record for the message. (Yes, there was some broken host that was
6066 actually sending 8xx at this point.)
6068 73. An unknown lookup type in a host list could cause Exim to panic-die when
6069 the list was checked. (An example that provoked this was putting <; in the
6070 middle of a list instead of at the start.) If this happened during a DATA
6071 ACL check, a -D file could be left lying around. This kind of configuration
6072 error no longer causes Exim to die; instead it causes a defer error. The
6073 incident is still logged to the main and panic logs.
6075 74. Buglet left over from Exim 3 conversion. The message "too many messages
6076 in one connection" was written to the rejectlog but not the mainlog, except
6077 when address rewriting (yes!) was being logged.
6079 75. Added write_rejectlog option.
6081 76. When a system filter was run not as root (that is, when system_filter_user
6082 was set), the values of the $n variables were not being returned to the
6083 main process; thus, they were not subsequently available in the $sn
6086 77. Added +return_path_on_delivery log selector.
6088 78. A connection timeout was being treated differently from recipients deferred
6089 when testing hosts_max_try with a message that was older than the host's
6090 retry timeout. (The host should not be counted, thus allowing all hosts to
6091 be tried at least once before bouncing.) This may have been the cause of an
6092 occasionally reported bug whereby a message would remain on the queue
6093 longer than the retry timeout, but would be bounced if a delivery was
6094 forced. I say "may" because I never totally pinned down the problem;
6095 setting up timeout/retry tests is difficult. See also the next item.
6097 79. The ultimate address timeout was not being applied to errors that involved
6098 a combination of host plus message (for example, a timeout on a MAIL
6099 command). When an address resolved to a number of possible hosts, and they
6100 were not all tried for each delivery (e.g. because of hosts_max_try), a
6101 message could remain on the queue longer than the retry timeout.
6103 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6106 81. Fixed an obscure SMTP outgoing bug which required at least the following
6107 conditions: (a) there was another message waiting for the same server;
6108 (b) the server returned 5xx to all RCPT commands in the first message so
6109 that the message was not completed; (c) the server dropped the connection
6110 or gave a negative response to the RSET that Exim sends to abort the
6111 transaction. The observed case was a dropped connection after DATA that had
6112 been sent in pipelining mode. That is, the server had advertised PIPELINING
6113 but was not implementing it correctly. The effect of the bug was incorrect
6114 behaviour, such as trying another host, and this could lead to a crash.
6120 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6121 and daemon.c were passed as pointers to ints; they should have been
6122 pointers to socklen_t variables (which are typically unsigned ints).
6124 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6127 3. Fixed a really odd bug that affected only the testing scheme; patching a
6128 certain fixed string in the binary changed the value of another string that
6129 happened to be identical to the end of the original first string.
6131 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6132 name", it returns that address as the IP address. On some operating
6133 systems (e.g. Solaris), it also passes back the IP address string as the
6134 "host name". However, on others (e.g. Linux), it passes back an empty
6135 string. Exim wasn't checking for this, and was changing the host name to an
6136 empty string, assuming it had been canonicalized.
6138 5. Although rare, it is permitted to have more than one PTR record for a given
6139 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6140 all the names associated with an address, because they do in Solaris.
6141 However, it seems that they do not in Linux for data that comes from the
6142 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6143 I found this out when I moved to a new Linux workstation and tried to run
6144 the Exim test suite.
6146 To get round this problem I have changed the code so that it now does its
6147 own call to the DNS to look up PTR records when searching for a host name.
6148 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6149 addresses that are only in /etc/hosts are still found.
6151 This behaviour is, however, controlled by an option called host_lookup_
6152 order, which defaults to "bydns:byaddr". If people want to use the other
6153 order, or indeed, just use one or the other means of lookup, they can
6154 specify it in this variable.
6156 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6157 some operating systems, this comes back from gethostbyaddr() as an empty
6158 string, and this is what Exim used to test for. However, it seems that in
6159 other systems, "." is yielded. Exim now tests for this case too.
6161 7. The values of check_spool_space and check_log_space are now held internally
6162 as a number of kilobytes instead of an absolute number of bytes. If a
6163 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6164 kilobyte. This means that much larger values can be stored.
6166 8. Exim monitor: an attempt to get the action menu when not actually pointing
6167 at a message produces an empty menu entitled "No message selected". This
6168 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6169 no entries in it ("Shell widget menu has zero width and/or height"). So I
6170 have added a single, blank menu entry in this case.
6172 9. Added ${quote_local_part.
6174 10. MIME decoding is now applied to the contents of Subject: header lines when
6177 11. Now that a reference to $sender_host_address automatically causes a reverse
6178 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6179 host lookup before query-style lookups in lists that might use this
6180 variable. This has therefore been abolished, and the "net-" prefix is no
6181 longer necessary for query-style lookups.
6183 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6184 have been a typo for LFLAGS, so it has been changed.
6186 13. The install script calls Exim with "-C /dev/null" in order to find the
6187 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6188 to be output. However, since Exim outputs its version number before the
6189 error, it didn't break the script. It just looked ugly. I fixed this by
6190 always allowing "-C /dev/null" if the caller is root.
6192 14. Ignore overlarge ACL variable number when reading spool file - insurance
6193 against a later release with more variables having written the file.
6195 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6196 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6197 this, as well as the form without the "IPv6" on the front.
6199 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6200 OS/Makefile-Darwin file.
6202 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6203 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6205 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6206 variable. However, some OS use ints for inodes. I've added cast to long int
6207 to get rid of the compiler warning.
6209 19. I had forgotten to lock out "/../" in configuration file names when
6210 ALT_CONFIG_PREFIX was set.
6212 20. Routers used for verification do not need to specify transports. However,
6213 if such a router generated a host list, and callout was configured, Exim
6214 crashed, because it could not find a port number from the (non-existent)
6215 transport. It now assumes port 25 in this circumstance.
6217 21. Added the -t option to exigrep.
6219 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6220 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6221 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6222 them is set, LOOKUP_LSEARCH is forced.
6224 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6226 24. Added sender and host information to the "rejected by local_scan()" log
6227 line; previously there was no indication of these.
6229 25. Added .include_if_exists.
6231 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6232 Linux. It turns out that not all file systems support this. Apparently some
6233 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6234 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6235 not supported on the file descriptor, is now ignored when Exim is trying to
6236 sync a directory. This applies only to Linux.
6238 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6240 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6241 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6244 29. Added support for maildirsize files from supplied patch (modified a bit).
6246 30. The use of :fail: followed by an empty string could lead Exim to respond to
6247 sender verification failures with (e.g.):
6249 550 Verification failed for <xxx>
6250 550 Sender verify failed
6252 where the first response line was missing the '-' that indicates it is not
6253 the final line of the response.
6255 31. The loop for finding the name of the user that called Exim had a hardwired
6256 limit of 10; it now uses the value of finduser_retries, which is used for
6257 all other user lookups.
6259 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6262 33. Exim was neglecting to zero errno before one call of strtol() when
6263 expanding a string and expecting an integer value. On some systems this
6264 resulted in spurious "integer overflow" errors. Also, it was casting the
6265 result into an int without checking.
6267 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6268 did not work. The code looks as if it has *never* worked, though it appears
6269 to have been documented since at least release 1.62. I have made it work.
6271 35. The "timeout_DNS" error in retry rules, also documented since at least
6272 1.62, also never worked. As it isn't clear exactly what this means, and
6273 clearly it isn't a major issue, I have abolished the feature by treating it
6274 as "timeout", and writing a warning to the main and panic logs.
6276 36. The display of retry rules for -brt wasn't always showing the error code
6279 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6280 timeout_connect_A, timeout_connect_MX.
6282 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6283 to the empty sender.
6285 39. The daemon was not analysing the content of -oX till after it had closed
6286 stderr and disconnected from the controlling terminal. This meant that any
6287 syntax errors were only noted on the panic log, and the return code from
6288 the command was 0. By re-arranging the code a little, I've made the
6289 decoding happen first, so such errors now appear on stderr, and the return
6290 code is 1. However, the actual setting up of the sockets still happens in
6291 the disconnected process, so errors there are still only recorded on the
6294 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6295 connections (as happens on some IP stacks) was logged at start up time as
6296 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6297 it from "IPv6 and IPv4", which means that two separate sockets are being
6300 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6301 says whether AF_INET or AF_INET6 was passed as an argument.
6303 42. Exiwhat output was messed up when time zones were included in log
6306 43. Exiwhat now gives more information about the daemon's listening ports,
6307 and whether -tls-on-connect was used.
6309 44. The "port" option of the smtp transport is now expanded.
6311 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6312 silently ignored. Now an error message is written to the main and panic
6315 46. There's a new ACL modifier called "logwrite" which writes to a log file
6316 as soon as it is encountered.
6318 47. Added $local_user_uid and $local_user_gid at routing time.
6320 48. Exim crashed when trying to verify a sender address that was being
6323 49. Exim was recognizing only a space character after ".include". It now also
6324 recognizes a tab character.
6326 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6327 extracting the relevant information from the specification. The man page no
6328 longer contains scrambled data for the -d option, and I've added a section
6329 at the front about calling Exim under different names.
6331 51. Added "extra_headers" argument to the "mail" command in filter files.
6333 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6336 53. Installed eximstats 1.29.
6338 54. Added transport_filter_timeout as a generic transport option.
6340 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6341 Cc: header lines. This was required by RFC 822, but it not required by RFC
6344 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6345 incoming messages that did not have them. Now it does so only if the
6346 message originates locally, that is, if there is no associated remote host
6347 address. When Resent- header lines are present, this applies to the Resent-
6348 lines rather than the non-Resent- lines.
6350 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6351 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6353 58. Messages for configuration errors now include the name of the main
6354 configuration file - useful now that there may be more than one file in a
6355 list (.included file names were always shown).
6357 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6358 for those rare installations that do not start the daemon as root or run it
6359 setuid root. I've cut out the call to initgroups() if the daemon is not
6362 60. The Exim user and group can now be bound into the binary as text strings
6363 that are looked up at the start of Exim's processing.
6365 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6367 62. Added $mailstore_basename variable.
6369 63. Installed patch to sieve.c from Michael Haardt.
6371 64. When Exim failed to open the panic log after failing to open the main log,
6372 the original message it was trying to log was written to stderr and debug
6373 output, but if they were not available (the usual case in production), it
6374 was lost. Now it is written to syslog before the two lines that record the
6375 failures to open the logs.
6377 65. Users' Exim filters run in subprocesses under the user's uid. It is
6378 possible for a "deliver" command or an alias in a "personal" command to
6379 provoke an address rewrite. If logging of address rewriting is configured,
6380 this fails because the process is not running as root or exim. There may be
6381 a better way of dealing with this, but for the moment (because 4.30 needs
6382 to be released), I have disabled address rewrite logging when running a
6383 filter in a non-root, non-exim process.
6389 1. The buildconfig auxiliary program wasn't quoting the value set for
6390 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6391 not defined. This bug was masked in 4.22 by the effect that was fixed in
6394 2. Some messages that were rejected after a message id was allocated were
6395 shown as "incomplete" by exigrep. It no longer does this for messages that
6396 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6398 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6399 have allow_domain_literals set, the ID did not get logged in the <= line.
6400 Domain literals are now always recognized in Message-ID: header lines.
6402 4. The first argument for a ${extract expansion item is the key name or field
6403 number. Leading and trailing spaces in this item were not being ignored,
6404 causing some misleading effects.
6406 5. When deliver_drop_privilege was set, single queue runner processes started
6407 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6408 same command in the process it spins off) were not dropping privilege.
6410 6. When the daemon running as "exim" started a queue runner, it always
6411 re-executed Exim in the spun-off process. This is a waste of effort when
6412 deliver_drop_privilege is set. The new process now just calls the
6413 queue-runner function directly.
6419 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6422 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6423 ignored. Though the use of -oP was forcing the writing of a pid file, it
6424 was always written to the default place.
6426 3. If the message "no IP address found for host xxxx" is generated during
6427 incoming verification, it is now followed by identification of the incoming
6428 connection (so you can more easily find what provoked it).
6430 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6432 5. Added some features to "harden" Exim a bit more against certain attacks:
6434 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6435 be put in Local/Makefile. This is like the never_users runtime option,
6436 but it cannot be overridden. The default setting is "root".
6438 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6439 prefix string with which any file named in a -C command line option
6442 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6443 is retained for -C and -D only if the caller of Exim is root. Without
6444 it, the exim user may also use -C and -D and retain privilege.
6446 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6447 command line option is disabled.
6449 6. Macro names set by the -D option must start with an upper case letter, just
6450 like macro names defined in the configuration file.
6452 7. Added "dereference=" facility to LDAP.
6454 8. Two instances of the typo "uknown" in the source files are fixed.
6456 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6457 the Configure-Makefile script screwed up while processing it.
6459 10. Incorporated PCRE 4.4.
6461 11. The SMTP synchronization check was not operating right at the start of an
6462 SMTP session. For example, it could not catch a HELO sent before the client
6463 waited for the greeting. There is now a check for outstanding input at the
6464 point when the greeting is written. Because of the duplex, asynchronous
6465 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6466 way, but not yet received, when the check is performed.
6468 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6469 on TCP/IP sockets, because this apparently causes some broken clients to
6472 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6473 unchanged) from the Cygwin maintainer.
6475 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6476 was supported instead of testing for mbx. Effectively a typo.
6478 15. The spa authenticator server code was not checking that the input it
6479 received was valid base64.
6481 16. The debug output line for the "set" modifier in ACLs was not showing the
6482 name of the variable that was being set.
6484 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6486 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6487 lookup on its own. Something else was needed to trigger the lookup. For
6488 example, a match in host_lookup or the need for a host name in a host list.
6489 Now, if $sender_host_name is referenced and the host name has not yet been
6490 looked up, a lookup is performed. If the lookup fails, the variable remains
6491 empty, and $host_lookup_failed is set to "1".
6493 19. Added "eqi" as a case-independent comparison operator.
6495 20. The saslauthd authentication condition could segfault if neither service
6496 nor realm was specified.
6498 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6499 error message that was logged was misleading, and incoming SMTP
6500 connections were dropped. The message is now more accurate, and temporary
6501 errors are given to SMTP connections.
6503 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6504 (except RSET) with a 421 error, until QUIT is received. However, it was
6505 failing to send a response to QUIT.
6507 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6508 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6509 ACL is now run after the helo_try_verify_hosts code.
6511 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6512 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6513 case-independent, so other case variants are also recognized). Apparently
6514 some systems use these upper case variants.
6516 25. If more than two messages were waiting for the same host, and a transport
6517 filter was specified for the transport, Exim sent two messages over the
6518 same TCP/IP connection, and then failed with "socket operation on non-
6519 socket" when it tried to send the third.
6521 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6523 27. The extern definition of crypt16() in expand.c was not being excluded when
6524 the OS had its own crypt16() function.
6526 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6527 as a preferred synonym for return_size_limit, both as an option and as an
6530 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6532 30. Changed the default configuration ACL to relax the local part checking rule
6533 for addresses that are not in any local domains. For these addresses,
6534 slashes and pipe symbols are allowed within local parts, but the sequence
6535 /../ is explicitly forbidden.
6537 31. SPA server authentication was not clearing the challenge buffer before
6540 32. log_message in a "warn" ACL statement was writing to the reject log as
6541 well as to the main log, which contradicts the documentation and doesn't
6542 seem right (because no rejection is happening). So I have stopped it.
6544 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6545 However, I am unable to do any testing of this.
6547 34. Fixed an infelicity in the appendfile transport. When checking directories
6548 for a mailbox, to see if any needed to be created, it was accidentally
6549 using path names with one or more superfluous leading slashes; tracing
6550 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6552 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6553 contents are added to the log line that is written for every discarded
6554 recipient. (Previously a log_message setting was ignored.)
6556 36. The ${quote: operator now quotes the string if it is empty.
6558 37. The install script runs exim in order to find its version number. If for
6559 some reason other than non-existence or emptiness, which it checks, it
6560 could not run './exim', it was installing it with an empty version number,
6561 i.e. as "exim-". This error state is now caught, and the installation is
6564 38. An argument was missing from the function that creates an error message
6565 when Exim fails to connect to the socket for saslauthd authentication.
6566 This could cause Exim to crash, or give a corrupted message.
6568 39. Added isip, isip4, and isip6 to ${if conditions.
6570 40. The ACL variables $acl_xx are now saved with the message, and can be
6571 accessed later in routers, transports, and filters.
6573 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6574 strings in the file are not string-expanded.
6576 42. If a MAIL command specified a SIZE value that was too large to fit into an
6577 int variable, the check against message_size_limit failed. Such values are
6578 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6579 day this will have to be increased, but I don't think I want to be around
6580 when emails are that large.
6587 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6588 iconv() is not standard in FreeBSD.
6590 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6591 IPv6 enabled. The observed symptom was a segmentation fault on return from
6592 the function os_common_find_running_interfaces() in src/os.c.
6594 3. In the check_special_case() function in daemon.c I had used "errno" as an
6595 argument name, which causes warnings on some systems. This was basically a
6596 typo, since it was named "eno" in the comments!
6598 4. The code that waits for the clock to tick (at a resolution of some fraction
6599 of a second) so as to ensure message-id uniqueness was always waiting for
6600 at least one whole tick, when it could have waited for less. [This is
6601 almost certainly not relevant at current processor speeds, where it is
6602 unlikely to ever wait at all. But we try to future-proof.]
6604 5. The function that sleeps for a time interval that includes fractions of a
6605 second contained a race. It did not block SIGALRM between setting the
6606 timer, and suspending (a couple of lines later). If the interval was short
6607 and the sigsuspend() was delayed until after it had expired, the suspension
6608 never ended. On busy systems this could lead to processes getting stuck for
6611 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6612 process, before it forks any delivery processes. The open lookup caching
6613 mechanism meant that the open file or database connection was passed into
6614 the delivery process. The problem was that delivery processes always tidy
6615 up cached lookup data. This could cause a problem for the next delivery
6616 process started by the queue runner, because the external queue runner
6617 process does not know about the closure. So the next delivery process
6618 still has data in the lookup cache. In the case of a file lookup, there was
6619 no problem because closing a file descriptor in a subprocess doesn't affect
6620 the parent. However, if the lookup was caching a connection to a database,
6621 the connection was closed, and the second delivery process was likely to
6622 see errors such as "PGSQL: query failed: server closed the connection
6623 unexpectedly". The problem has been fixed by closing all cached lookups
6624 in a queue runner before running a delivery process.
6626 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6627 seem to have the "const" qualifier which it has on other OS. I've
6630 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6631 *of the same type* (client or server) with the same public name that an
6632 error should be diagnosed.
6634 9. When Exim looked up a host name for an IP address, but failed to find the
6635 original IP address when looking up the host name (a safety check), it
6636 output the message "<ip address> does not match any IP for NULL", which was
6637 confusing, to say the least. The bug was that the host name should have
6638 appeared instead of "NULL".
6640 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6641 user that is built into the binary, and the -C or -D options is used, root
6642 privilege is dropped before the configuration file is read. In addition,
6643 logging is switched to stderr instead of the normal log files. If the
6644 configuration then re-defines the Exim user, the unprivileged environment
6645 is probably not what is expected, so Exim logs a panic warning message (but
6648 However, if deliver_drop_privilege is set, the unprivileged state may well
6649 be exactly what is intended, so the warning has been cut out in that case,
6650 and Exim is allowed to try to write to its normal log files.
6656 1. smtp_return_error_details was not giving details for temporary sender
6657 or receiver verification errors.
6659 2. Diagnose a configuration error if two authenticators have the same public
6662 3. Exim used not to create the message log file for a message until the first
6663 delivery attempt. This could be confusing when incoming messages were held
6664 for policy or load reasons. The message log file is now created at the time
6665 the message is received, and an initial "Received" line is written to it.
6667 4. The automatically generated man page for command line options had a minor
6668 bug that caused no ill effects; however, a more serious problem was that
6669 the procedure for building the man page automatically didn't always
6670 operate. Consequently, release 4.20 contains an out-of-date version. This
6671 shouldn't happen again.
6673 5. When building Exim with embedded Perl support, the script that builds the
6674 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6675 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6677 6. The freeze_tell option was not being used for messages that were frozen on
6678 arrival, either by an ACL or by local_scan().
6680 7. Added the smtp_incomplete_transaction log selector.
6682 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6683 was accepting AUTH without a new EHLO.
6685 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6686 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6689 10. Logging of TCP/IP connections (when configured) now happens in the main
6690 daemon process instead of the child process, so that the TCP/IP connection
6691 count is more accurate (but it can never be perfect).
6693 11. The use of "drop" in a nested ACL was not being handled correctly in the
6694 outer ACL. Now, if condition failure induced by the nested "drop" causes
6695 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6696 or "require"), the connection is dropped.
6698 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6699 that yield "discard" can now be used with an "accept" or a "discard" verb,
6700 but an error is generated for any others (because I can't see a useful way
6701 to define what should happen).
6703 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6704 are now processed in the same way as lines in the Exim configuration file.
6705 In particular, continuation lines are supported.
6707 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6709 15. Added -ti meaning -t -i.
6711 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6712 domains, and warn by logging if others are found.
6714 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6715 returned from GIFCONF when Exim is trying to find the list of interfaces on
6716 a host. The code in os.c has been modified to copy each ifreq to an aligned
6717 structure in all cases.
6719 Also, in some cases, the returned ifreq's were being copied to a 'struct
6720 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6721 means the last couple of bytes of an IPv6 address could be chopped if the
6722 ifreq contained only a normal sockaddr (14 bytes storage).
6724 18. Named domain lists were not supported in the hosts_treat_as_local option.
6725 An entry such as +xxxx was not recognized, and was treated as a literal
6728 19. Ensure that header lines added by a DATA ACL are included in the reject log
6729 if the ACL subsequently rejects the message.
6731 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6732 MD5 (which is deprecated).
6734 21. When testing a filter file using -bf, Exim was writing a message when it
6735 took the sender from a "From " line in the message, but it was not doing so
6736 when it took $return_path from a Return-Path: header line. It now does.
6738 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6739 with a valid header line field name (a series of printing characters
6740 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6742 23. Changed "disc" in the source to "disk" to conform to the documentation and
6743 the book and for uniformity.
6745 24. Ignore Sendmail's -Ooption=value command line item.
6747 25. When execve() failed while trying to run a command in a pipe transport,
6748 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6749 could be confused with a return value of 69 from the command itself. This
6750 has been changed to 127, the value the shell returns if it is asked to run
6751 a non-existent command. The wording for the related log line suggests a
6752 non-existent command as the problem.
6754 26. If received_header_text expands to an empty string, do not add a Received:
6755 header line to the message. (Well, it adds a token one on the spool, but
6756 marks it "old" so that it doesn't get used or transmitted.)
6758 27. Installed eximstats 1.28 (addition of -nt option).
6760 28. There was no check for failure on the call to getsockname() in the daemon
6761 code. This can fail if there is a shortage of resources on the system, with
6762 ENOMEM, for example. A temporary error is now given on failure.
6764 29. Contrary to the C standard, it seems that in some environments, the
6765 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6766 program. Exim now does this explicitly; it affects the formatting of
6767 timestamps using strftime().
6769 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6770 complaints. I've now initialized all the variables, to avoid this.
6772 32. Header lines added by a system filter were not being "seen" during
6773 transport-time rewrites.
6775 33. The info_callback() function passed to OpenSSL is set up with type void
6776 (*)(SSL *, int, int), as described somewhere. However, when calling the
6777 function (actually a macro) that sets it up, the type void(*)() is
6778 expected. I've put in a cast to prevent warnings from picky compilers.
6780 34. If a DNS black list lookup found a CNAME record, but there were no A
6781 records associated with the domain it pointed at, Exim crashed.
6783 35. If a DNS black list lookup returned more than one A record, Exim ignored
6784 all but the first. It now scans all returned addresses if a particular IP
6785 value is being sought. In this situation, the contents of the
6786 $dnslist_value variable are a list of all the addresses, separated by a
6789 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6790 to accept a host name and all its aliases if the forward lookup for any of
6791 them yielded the IP address of the incoming connection. Now it accepts only
6792 those names whose forward lookup yields the correct IP address. Any other
6793 names are discarded. This closes a loophole whereby a rogue DNS
6794 administrator could create reverse DNS records to break through a
6795 wildcarded host restriction in an ACL.
6797 37. If a user filter or a system filter that ran in a subprocess used any of
6798 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6799 the wrong values were passed to the pipe command ($thisaddress had the
6800 value of $0, $0 had the value of $1, etc). This bug was introduced by
6801 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6803 38. Improved the line breaking for long SMTP error messages from ACLs.
6804 Previously, if there was no break point between 40 and 75 characters, Exim
6805 left the rest of the message alone. Two changes have been made: (a) I've
6806 reduced the minimum length to 35 characters; (b) if it can't find a break
6807 point between 35 and 75 characters, it looks ahead and uses the first one
6808 that it finds. This may give the occasional overlong line, but at least the
6809 remaining text gets split now.
6811 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6812 file descriptors might be low, and that setting 1000 would always raise it.
6813 It turns out that in some environments, the limit is already over 1000 and
6814 that lowering it causes trouble. So now Exim takes care not to decrease it.
6816 40. When delivering a message, the value of $return_path is set to $sender_
6817 address at the start of routing (routers may change the value). By an
6818 oversight, this default was not being set up when an address was tested by
6819 -bt or -bv, which affected the outcome if any router or filter referred to
6822 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6823 log only when "message" or "log_message" are set. However, if one of the
6824 conditions was an address verification, or a call to a nested ACL, the
6825 messages generated by the underlying test were being passed through. This
6826 no longer happens. The underlying message is available in $acl_verify_
6827 message for both "message" and "log_message" expansions, so it can be
6828 passed through if needed.
6830 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6831 new expansion $bh_ to give the encoded byte string without charset
6832 translation. Translation happens only if iconv() is available; HAVE_ICONV
6833 indicates this at build time. HEADERS_CHARSET gives the charset to
6834 translate to; headers_charset can change it in the configuration, and
6835 "headers charset" can change it in an individual filter file.
6837 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6838 that creates RFC 2047 encoded "words" labels them as that charset instead
6839 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6840 expansion operator; (ii) when Exim creates a From: line for a local
6841 message; (iii) when a header line is rewritten to include a "phrase" part.
6843 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6844 buggy, causing it to skip the first lines of messages whose message ID
6845 ended in 'D'. This would not have bitten before Exim release 4.14, because
6846 message IDs were unlikely to end in 'D' before then. The effect was to have
6847 incorrect size information for certain domains.
6849 45. #include "config.h" was missing at the start of the crypt16.c module. This
6850 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6853 46. If there was a timeout during a "random" callout check, Exim treated it as
6854 a failure of the random address, and carried on sending RSET and the real
6855 address. If the delay was just some slowness somewhere, the response to the
6856 original RCPT would be taken as a response to RSET and so on, causing
6857 mayhem of various kinds.
6859 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6860 when I implemented it. It didn't allow for the fact that some option values
6861 may legitimately be negative (e.g. size_addition), and it didn't even do
6862 the right test for positive values.
6864 48. Domain names in DNS records are case-independent. Exim always looks them up
6865 in lower case. Some resolvers return domain names in exactly the case they
6866 appear in the zone file, that is, they may contain uppercase letters. Not
6867 all resolvers do this - some return always lower case. Exim was treating a
6868 change of case by a resolver as a change of domain, similar to a widening
6869 of a domain abbreviation. This triggered its re-routing code and so it was
6870 trying to route what was effectively the same domain again. This normally
6871 caused routing to fail (because the router wouldn't handle the domain
6872 twice). Now Exim checks for this case specially, and just changes the
6873 casing of the domain that it ultimately uses when it transmits the message
6876 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6879 50. If a filter generated a file delivery with a non-absolute name (possible if
6880 no home directory exists for the router), the forbid_file option was not
6883 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6884 the existing equality matching.
6886 52. Exim was using ints instead of ino_t variables in some places where it was
6887 dealing with inode numbers.
6889 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6890 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6891 called TMPDIR, and if it finds it is different, it changes its value.
6893 54. The smtp_printf() function is now made available to local_scan() so
6894 additional output lines can be written before returning. There is also an
6895 smtp_fflush() function to enable the detection of a dropped connection.
6896 The variables smtp_input and smtp_batched_input are exported to
6899 55. Changed the default runtime configuration: the message "Unknown user"
6900 has been removed from the ACL, and instead placed on the localuser router,
6901 using the cannot_route_message feature. This means that any verification
6902 failures that generate their own messages won't get overridden. Similarly,
6903 the "Unrouteable address" message that was in the ACL for unverifiable
6904 relay addresses has also been removed.
6906 56. Added hosts_avoid_esmtp to the smtp transport.
6908 57. The exicyclog script was not checking for the esoteric option
6909 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6910 will work only if exicyclog is run under the appropriate euid.
6912 58. Following a discussion on the list, the rules by which Exim recognises line
6913 endings on incoming messages have been changed. The -dropcr and drop_cr
6914 options are now no-ops, retained only for backwards compatibility. The
6915 following line terminators are recognized: LF CRLF CR. However, special
6916 processing applies to CR:
6918 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6919 nor a local message in the state where . is a terminator.
6921 (ii) If a bare CR is encountered in a header line, an extra space is added
6922 after the line terminator so as not to end the header. The reasoning
6923 behind this is that bare CRs in header lines are most likely either
6924 to be mistakes, or people trying to play silly games.
6926 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6927 was being incorrectly given as 18 bytes larger than it should have been.
6928 This is a VOB (very old bug).
6930 60. This may never have affected anything current, but just in case it has:
6931 When the local host is found other than at the start of a list of hosts,
6932 the local host, those with the same MX, and any that follow, are discarded.
6933 When the list in question was part of a longer list of hosts, the following
6934 hosts (not currently being processed) were also being discarded. This no
6935 longer happens. I'm not sure if this situation could ever has previously
6938 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6941 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6942 header. This is something that is recommended in a new Internet Draft, and
6943 is something that is documented as being done by Sendmail. There are two
6944 possible values. For messages generated by the autoreply transport, Exim
6947 Auto-Submitted: auto-replied
6949 whereas for all other generated messages (e.g. bounces) it adds
6951 Auto-Submitted: auto-generated
6953 63. The "personal" condition in filters now includes a test for the
6954 Auto-Submitted: header. If it contains the string "auto-" the message it
6955 not considered personal.
6957 64. Added rcpt_include_affixes as a generic transport option.
6959 65. Added queue_only_override (default true).
6961 66. Added the syslog_duplication option.
6963 67. If what should have been the first header line of a message consisted of
6964 a space followed by a colon, Exim was mis-interpreting it as a header line.
6965 It isn't of course - it is syntactically invalid and should therefore be
6966 treated as the start of the message body. The misbehaviour could have
6967 caused a number of strange effects, including loss of data in subsequent
6968 header lines, and spool format errors.
6970 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6971 client host had authenticated. This control can now be exercised by an ACL
6972 for more flexibility.
6974 69. By default, callouts do not happen when testing with -bh. There is now a
6975 variant, -bhc, which does actually run the callout code, including
6976 consulting and updating the callout cache.
6978 70. Added support for saslauthd authentication, courtesy of Alexander
6981 71. If statvfs() failed on the spool or log directories while checking their
6982 size for availability, Exim confusingly gave the error "space shortage".
6983 Furthermore, in debugging mode it crashed with a floating point exception.
6984 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6985 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6986 serious problem, Exim now writes to the main and panic logs when this
6987 happens, with details of the failure. It then refuses to accept the
6988 incoming message, giving the message "spool directory problem" or "log
6989 directory problem" with a 421 code for SMTP messages.
6991 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6992 0, 1, and 2 exist, because some OS complain for execs without them (see
6993 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6994 descriptors. However, the code omitted to check that the open succeeded,
6995 causing mysterious errors if for some reason the permissions on /dev/null
6996 got screwed. Now Exim writes a message to the main and panic logs, and
6997 bombs out if it can't open /dev/null.
6999 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7000 interact so that it is all more flexible. It is supposed to remain
7001 backwards compatible. Also added extra_local_interfaces.
7003 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7004 to bomb out with an assertion failure - to the client this appears as a
7005 connection drop. This problem occurs in the part of the code that was taken
7006 from the Samba project. Fortunately, the assertion is in a very simple
7007 function, so I have fixed this by reproducing the function inline in the
7008 one place where it is called, and arranging for authentication to fail
7009 instead of killing the process with assert().
7011 75. The SPA client code was not working when the server requested OEM rather
7012 than Unicode encoding.
7014 76. Added code to make require_files with a specific uid setting more usable in
7015 the case where statting the file as root fails - usually a non-root-mounted
7016 NFS file system. When this happens and the failure is EACCES, Exim now
7017 forks a subprocess and does the per-uid checking as the relevant uid.
7019 77. Added process_log_path.
7021 78. If log_file_path was not explicitly set, a setting of check_log_space or
7022 check_log_inodes was ignored.
7024 79. If a space check for the spool or log partitions fails, the incident is now
7025 logged. Of course, in the latter case the data may get lost...
7027 80. Added the %p formatting code to string_format() so that it can be used to
7028 print addresses in debug_print(). Adjusted all the address printing in the
7029 debugging in store.c to use %p rather than %d.
7031 81. There was a concern that a line of code in smtp_in.c could overflow a
7032 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7033 initially expressed, the concern was not well-founded, because trailing
7034 spaces are removed early. However, if the trailing spaces were followed by
7035 a NULL, they did not get removed, so the overflow was possible. Two fixes
7038 (a) I re-wrote the offending code in a cleaner fashion.
7039 (b) If an incoming SMTP command contains a NULL character, it is rejected
7042 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7043 runs initgroups(), so that if the Exim user is in any additional groups,
7044 they will be used during message reception.
7050 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / heiko / exim.git / blob