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.
8 Since Exim version 4.90
11 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
12 Previously only the last row was returned.
14 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
15 we assumed that tags in the header were well-formed, and parsed the
16 element content after inspecting only the first char of the tag.
17 Assumptions at that stage could crash the receive process on malformed
20 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
21 While running the DKIM ACL we operate on the Permanent memory pool so that
22 variables created with "set" persist to the DATA ACL. Also (at any time)
23 DNS lookups that fail create cache records using the Permanent pool. But
24 expansions release any allocations made on the current pool - so a dnsdb
25 lookup expansion done in the DKIM ACL releases the memory used for the
26 DNS negative-cache, and bad things result. Solution is to switch to the
27 Main pool for expansions.
28 While we're in that code, add checks on the DNS cache during store_reset,
29 active in the testsuite.
30 Problem spotted, and debugging aided, by Wolfgang Breyha.
32 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
33 When none of the hosts presented to a transport match an already-open
34 connection, close it and proceed with the list. Previously we would
35 queue the message. Spotted by Lena with Yahoo, probably involving
38 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
39 Previously a spurious "250 OK id=" response was appended to the proper
42 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
43 rows affected is given instead).
45 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
46 SMTP connection. Previously, when one had more receipients than the
47 first, an abortive onward connection was made. Move to full support for
48 multiple onward connections in sequence, handling cutthrough connection
49 for all multi-message initiating connections.
51 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
52 routers. Previously, a multi-recipient message would fail to match the
53 onward-connection opened for the first recipient, and cause its closure.
55 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
56 a timeout on read on a GnuTLS initiating connection, resulting in the
57 initiating connection being dropped. This mattered most when the callout
58 was marked defer_ok. Fix to keep the two timeout-detection methods
61 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
62 metadata, resulting in a crash in free().
64 PP/01 Fix broken refactor of Heimdal GSSAPI authenticator init.
65 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
71 JH/01 Rework error string handling in TLS interface so that the caller in
72 more cases is responsible for logging. This permits library-sourced
73 string to be attached to addresses during delivery, and collapses
74 pairs of long lines into single ones.
76 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
77 during configuration. Wildcards are allowed and expanded.
79 JH/02 Rework error string handling in DKIM to pass more info back to callers.
80 This permits better logging.
82 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
83 do not close it down and have the child transport start it up again on
84 the passed-on TCP connection. Instead, proxy the child (and any
85 subsequent ones) for TLS via a unix-domain socket channel. Logging is
86 affected: the continued delivery log lines do not have any DNSSEC, TLS
87 Certificate or OCSP information. TLS cipher information is still logged.
89 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
90 identical IP addresses on different listening ports. Will also affect
93 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
94 add noisy ifdef guards to special-case this sillyness.
95 Patch from Bernd Kuhls.
97 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
98 than 255 are no longer allowed.
100 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
101 Disable the session-cache too, which might reduce our load. Since we
102 currrectly use a new context for every connection, both as server and
103 client, there is no benefit for these.
104 GnuTLS appears to not support tickets server-side by default (we don't
105 call gnutls_session_ticket_enable_server()) but client side is enabled
106 by default on recent versions (3.1.3 +) unless the PFS priority string
109 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
110 <https://reproducible-builds.org/specs/source-date-epoch/>.
112 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
113 the check for any unsuccessful recipients did not notice the limit, and
114 erroneously found still-pending ones.
116 JH/08 Pipeline CHUNKING command and data together, on kernels that support
117 MSG_MORE. Only in-clear (not on TLS connections).
119 JH/09 Avoid using a temporary file during transport using dkim. Unless a
120 transport-filter is involved we can buffer the headers in memory for
121 creating the signature, and read the spool data file once for the
122 signature and again for transmission.
124 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
125 4.77 as the kernel support then wasn't solid, having issues in 64bit
126 mode. Now, it's been long enough. Add support for FreeBSD also.
128 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
129 case where the routing stage had gathered several addresses to send to
130 a host before calling the transport for the first, we previously failed
131 to close down TLS in the old transport process before passing the TCP
132 connection to the new process. The new one sent a STARTTLS command
133 which naturally failed, giving a failed delivery and bloating the retry
134 database. Investigation and fix prototype from Wolfgang Breyha.
136 JH/12 Fix check on SMTP command input synchronisation. Previously there were
137 false-negatives in the check that the sender had not preempted a response
138 or prompt from Exim (running as a server), due to that code's lack of
139 awareness of the SMTP input buffering.
141 PP/04 Add commandline_checks_require_admin option.
142 Exim drops privileges sanely, various checks such as -be aren't a
143 security problem, as long as you trust local users with access to their
144 own account. When invoked by services which pass untrusted data to
145 Exim, this might be an issue. Set this option in main configuration
146 AND make fixes to the calling application, such as using `--` to stop
149 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
150 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
151 responses to those, into a single TLS record each way (this usually means
152 a single packet). As a side issue, smtp_enforce_sync now works on TLS
155 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
156 affects you only if you're dancing at the edge of the param size limits.
157 If you are, and this message makes sense to you, then: raise the
158 configured limit or use OpenSSL 1.1. Nothing we can do for older
161 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
162 cmdline element to get the documented default one. Previously it was
165 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
166 are unsuable for non-message-specific reasons - eg. connection timeout,
167 banner-time rejection.
169 JH/16 Fix logging of delivery remote port, when specified by router, under
172 PP/06 Repair manualroute's ability to take options in any order, even if one
173 is the name of a transport.
176 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
178 JH/17 Change the list-building routines interface to use the expanding-string
179 triplet model, for better allocation and copying behaviour.
181 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
182 Previously it was constructed the first time a possibly-matching string
183 was met in the configuration file input during startup; now it is done
186 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
187 compatible one, to avoid the (poorly documented) possibility of a config
188 file in the working directory redirecting the DB files, possibly correpting
189 some existing file. CVE-2017-10140 assigned for BDB.
191 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
192 cache-hot. Previously, although the result was properly cached, the
193 initial verify call returned a defer.
195 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
196 the main verify for receipient in uncached-mode.
198 JH/22 Retire historical build files to an "unsupported" subdir. These are
199 defined as "ones for which we have no current evidence of testing".
201 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
202 if present. Previously it was ignored.
204 JH/24 Start using specified-initialisers in C structure init coding. This is
205 a C99 feature (it's 2017, so now considered safe).
207 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
208 if was a fixed-sized field and bitmask ops via macros; it is now more
211 PP/07 GitHub PR 56: Apply MariaDB build fix.
212 Patch provided by Jaroslav Škarvada.
214 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
215 during Coverity cleanups [4.87 JH/47]
216 Diagnosis and fix provided by Michael Fischer v. Mollard.
218 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
219 the right size to place the terminating semicolon on its own folded
220 line, the header hash was calculated to an incorrect value thanks to
221 the (relaxed) space the fold became.
223 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
224 and confused the parent.
226 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
227 which could crash as a result. This could lead to undeliverable messages.
229 JH/28 Logging: "next input sent too soon" now shows where input was truncated
232 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
233 matters on fast-turnover and PID-randomising systems, which were getting
234 out-of-order delivery.
236 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
237 a possibly-overlapping copy. The symptom was that "Remote host closed
238 connection in response to HELO" was logged instead of the actual 4xx
241 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
242 Previously only that bufferd was discarded, resulting in SYMTP command
245 JH/32 DKIM: when a message has multiple signatures matching an identity given
246 in dkim_verify_signers, run the dkim acl once for each. Previously only
247 one run was done. Bug 2189.
249 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
250 "panic the current process" to "deliberately defer". The panic log is
251 still written with the problem list name; the mail and reject logs now
252 get a temp-reject line for the message that was being handled, saying
253 something like "domains check lookup or other defer". The SMTP 451
254 message is still "Temporary local problem".
256 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
257 A crafted sequence of BDAT commands could result in in-use memory beeing
258 freed. CVE-2017-16943.
260 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
261 from SMTP input. Previously it was always done; now only done for DATA
262 and not BDAT commands. CVE-2017-16944.
264 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
265 to the message (such as an overlong header line). Previously this was
266 not done and we did not exit BDAT mode. Followon from the previous item
267 though a different problem.
273 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
274 than -2003 did; needs libidn2 in addition to libidn.
276 JH/02 The path option on a pipe transport is now expanded before use.
278 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
279 Patch provided by "Björn", documentation fix added too.
281 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
282 missing a wire-to-host endian conversion.
284 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
285 close after a BDAT command line could be taken as a following command,
286 giving a synch failure. Fix by only checking for synch immediately
287 before acknowledging the chunk.
289 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
290 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
291 macro. Patches provided by Josh Soref.
293 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
294 Previously we did not; the RFC seems ambiguous and VRFY is not listed
295 by IANA as a service extension. However, John Klensin suggests that we
298 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
299 the dkim code may be unix-mode line endings rather than smtp wire-format
300 CRLF, so prepend a CR to any bare LF.
302 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
303 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
305 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
306 the first were themselves being wrongly included in the feed into dkim
307 processing; with most chunk sizes in use this resulted in an incorrect
308 body hash calculated value.
310 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
311 DKIM signature block, for verification. Although advised against by
312 standards it is specifically not ruled illegal.
314 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
316 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
317 missing a body hash (the bh= tag).
319 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
320 It seems that HAProxy sends the Proxy Protocol information in clear and
321 only then does a TLS startup, so do the same.
323 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
324 TCP connections (such as for Spamd) unless the daemon successfully set
325 Fast Open mode on its listening sockets. This fixes breakage seen on
326 too-old kernels or those not configured for Fast Open, at the cost of
327 requiring both directions being enabled for TFO, and TFO never being used
328 by non-daemon-related Exim processes.
330 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
331 endings, at least on the first header line. Try to canonify any that get
332 past that check, despite the cost.
334 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
335 now limited to an arbitrary five deep, while parsing addresses with the
336 strip_excess_angle_brackets option enabled.
338 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
339 instead leave the unprompted TLS handshake in socket buffer for the
340 TLS library to consume.
342 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
344 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
346 JH/16 Drop variables when they go out of scope. Memory management drops a whole
347 region in one operation, for speed, and this leaves assigned pointers
348 dangling. Add checks run only under the testsuite which checks all
349 variables at a store-reset and panics on a dangling pointer; add code
350 explicitly nulling out all the variables discovered. Fixes one known
351 bug: a transport crash, where a dangling pointer for $sending_ip_address
352 originally assigned in a verify callout, is re-used.
354 PP/06 Drop '.' from @INC in various Perl scripts.
356 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
358 PP/08 Reduce a number of compilation warnings under clang; building with
359 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
360 should be warning-free.
362 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
364 HS/01 Fix portability problems introduced by PP/08 for platforms where
365 realloc(NULL) is not equivalent to malloc() [SunOS et al].
367 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
368 chunk. This allows us to accept broken chunked messages. We need a more
369 general solution here.
371 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
372 already-broken messages in the queue.
374 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
376 JH/19 Fix reference counting bug in routing-generated-address tracking.
382 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
383 supports it and a size is available (ie. the sending peer gave us one).
385 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
386 years of being deprecated). The replacements are the ACLs
387 acl_smtp_mime and acl_not_smtp_mime.
389 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
390 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
391 or even an in-clear connection were permitted. Now, if the host lookup
392 was dnssec and dane was requested then the host is only used if the
393 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
394 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
395 if one fails this test.
396 This means that a poorly-configured remote DNS will make it incommunicado;
397 but it protects against a DNS-interception attack on it.
399 JH/04 Bug 1810: make continued-use of an open smtp transport connection
400 non-noisy when a race steals the message being considered.
402 JH/05 If main configuration option tls_certificate is unset, generate a
403 self-signed certificate for inbound TLS connections.
405 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
406 in rewrites and routers.
408 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
409 and logged a warning sing 4.83; now they are a configuration file error.
411 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
412 (lacking @domain). Apply the same qualification processing as RCPT.
414 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
416 JH/10 Support ${sha256:} applied to a string (as well as the previous
419 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
420 a cutthrough deliver is pending, as we always want to make a connection.
421 This also avoids re-routing the message when later placing the cutthrough
422 connection after a verify cache hit.
423 Do not update it with the verify result either.
425 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
426 when routing results in more than one destination address.
428 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
429 signing (which inhibits the cutthrough capability). Previously only
430 the presence of an option was tested; now an expansion evaluating as
431 empty is permissible (obviously it should depend only on data available
432 when the cutthrough connection is made).
434 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
435 the relevant preceding SMTP command did not note the pipelining mode.
437 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
438 Previously they were not counted.
440 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
441 as one having no matching records. Previously we deferred the message
442 that needed the lookup.
444 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
445 distinguished as "(=".
447 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
448 for missing MX records. Previously it only worked for missing A records.
450 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
452 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
453 after the data-go-ahead and data-ack. Patch from Jason Betts.
455 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
456 even for a "none" policy. Patch from Tony Meyer.
458 JH/22 Fix continued use of a connection for further deliveries. If a port was
459 specified by a router, it must also match for the delivery to be
462 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
463 When one of the recipients of a message was unsuitable for the connection
464 (has no matching addresses), we lost track of needing to mark it
465 deferred. As a result mail would be lost.
467 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
469 JH/25 Decoding ACL controls is now done using a binary search; the source code
470 takes up less space and should be simpler to maintain. Merge the ACL
471 condition decode tables also, with similar effect.
473 JH/26 Fix problem with one_time used on a redirect router which returned the
474 parent address unchanged. A retry would see the parent address marked as
475 delivered, so not attempt the (identical) child. As a result mail would
478 JH/27 Fix a possible security hole, wherein a process operating with the Exim
479 UID can gain a root shell. Credit to http://www.halfdog.net/ for
480 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
483 JH/28 Enable {spool,log} filesystem space and inode checks as default.
484 Main config options check_{log,spool}_{inodes,space} are now
485 100 inodes, 10MB unless set otherwise in the configuration.
487 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
488 Previously it only applied to the main-section connection policy
491 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
493 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
494 by me. Added RFC7919 DH primes as an alternative.
496 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
497 are not in the system include path.
499 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
500 GnuTLS, when a session startup failed (eg because the client disconnected)
501 Exim did stdio operations after fclose. This was exposed by a recent
502 change which nulled out the file handle after the fclose.
504 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
505 signed directly by the cert-signing cert, rather than an intermediate
506 OCSP-signing cert. This is the model used by LetsEncrypt.
508 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
510 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
511 an incoming connection.
513 HS/02 Bug 1802: Do not half-close the connection after sending a request
516 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
517 fallback to "prime256v1".
519 JH/34 SECURITY: Use proper copy of DATA command in error message.
520 Could leak key material. Remotely exploitable. CVE-2016-9963.
526 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
527 and 3.4.4 - once the server is enabled to respond to an OCSP request
528 it does even when not requested, resulting in a stapling non-aware
529 client dropping the TLS connection.
531 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
532 support variable-length bit vectors. No functional change.
534 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
535 The I= interface field on outgoing lines is now after the H= remote
536 host field, same as incoming lines. There is a separate
537 outgoing_interface log selector which allows you to disable the
540 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
541 If not running log_selector +smtp_connection the mainlog would be held
542 open indefinitely after a "too many connections" event, including to a
543 deleted file after a log rotate. Leave the per net connection logging
544 leaving it open for efficiency as that will be quickly detected by the
545 check on the next write.
547 HS/01 Bug 1671: Fix post transport crash.
548 Processing the wait-<transport> messages could crash the delivery
549 process if the message IDs didn't exist for some reason. When
550 using 'split_spool_directory=yes' the construction of the spool
551 file name failed already, exposing the same netto behaviour.
553 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
554 mime_regex ACL conditions.
556 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
557 to DSN fail messages (bounces): remote IP, remote greeting, remote response
558 to HELO, local diagnostic string.
560 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
561 log line to debug. Even when configured with a tls authenticator many
562 client connections are expected to not authenticate in this way, so
563 an authenticate fail is not an error.
565 HS/02 Add the Exim version string to the process info. This way exiwhat
566 gives some more detail about the running daemon.
568 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
569 matter for fast-change records such as DNSBLs.
571 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
572 retry record, even if constant. There may be multiple transports with
573 different interface settings and the retry behaviour needs to be kept
576 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
578 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
580 JH/10 Bug 840: fix log_defer_output option of pipe transport
582 JH/11 Bug 830: use same host for all RCPTS of a message, even under
583 hosts_randomize. This matters a lot when combined with mua_wrapper.
585 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
586 ${quote_pgsql:<string>} operator.
588 JH/13 Bug 1708: avoid misaligned access in cached lookup.
590 JH/14 Change header file name for freeradius-client. Relevant if compiling
591 with Radius support; from the Gentoo tree and checked under Fedora.
593 JH/15 Bug 1712: Introduce $prdr_requested flag variable
595 JH/16 Bug 1714: Permit an empty string as expansion result for transport
596 option transport_filter, meaning no filtering.
598 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
600 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
601 defaults to "*" (all hosts). The variable is now available when not built
602 with TLS, default unset, mainly to enable keeping the testsuite sane.
603 If a server certificate is not supplied (via tls_certificate) an error is
604 logged, and clients will find TLS connections fail on startup. Presumably
605 they will retry in-clear.
606 Packagers of Exim are strongly encouraged to create a server certificate
607 at installation time.
609 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
610 with the $config_file variable.
612 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
613 in transport context, after the attempt, and per-recipient. The latter type
614 is per host attempted. The event data is the error message, and the errno
615 information encodes the lookup type (A vs. MX) used for the (first) host,
616 and the trailing two digits of the smtp 4xx response.
618 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
619 to write to mainlog (or rejectlog, paniclog) in the window between file
620 creation and permissions/ownership being changed. Particularly affects
621 installations where exicyclog is run as root, rather than exim user;
622 result is that the running daemon panics and dies.
624 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
626 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
627 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
628 "pri" and "weight". Note that the previous implicit priority given by the
629 list order is no longer honoured.
631 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
634 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
635 by defining SUPPORT_SOCKS.
637 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
638 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
639 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
640 variables are renamed to proxy_{local,external}_{address,port}.
642 JH/27 Move Internationalisation support from Experimental to mainline, enabled
643 for a build by defining SUPPORT_I18N
645 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
646 of the query string, and make ${quote_redis:} do that quoting.
648 JH/29 Move Events support from Experimental to mainline, enabled by default
649 and removable for a build by defining DISABLE_EVENT.
651 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
653 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
654 cached by the daemon.
656 JH/32 Move Redis support from Experimental to mainline, enabled for a build
657 by defining LOOKUP_REDIS. The libhiredis library is required.
659 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
660 keys are given for lookup.
662 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
663 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
664 only supported when built with TLS support. The PolarSSL SHA routines
665 are still used when the TLS library is too old for convenient support.
667 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
668 openssl_options), for security. OpenSSL forces this from version 1.1.0
669 server-side so match that on older versions.
671 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
672 allocation for $value could be released as the expansion processing
673 concluded, but leaving the global pointer active for it.
675 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
676 and to use the domains and local_parts ACL conditions.
678 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
679 incorrectly not doubled on cutthrough transmission, hence seen as a
680 body-termination at the receiving system - resulting in truncated mails.
681 Commonly the sender saw a TCP-level error, and retransmitted the message
682 via the normal store-and-forward channel. This could result in duplicates
683 received - but deduplicating mailstores were liable to retain only the
684 initial truncated version.
686 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
688 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
690 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
691 we're in there, support oversigning also; bug 1309.
693 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
695 HS/04 Add support for keep_environment and add_environment options.
697 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
698 either intentional arithmetic overflow during PRNG, or testing config-
701 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
702 delivery resulted in actual delivery. Cancel cutthrough before DATA
705 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
706 rejects a recipient: pass the reject to the originator.
708 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
709 Many were false-positives and ignorable, but it's worth fixing the
712 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
713 for the new environment-manipulation done at startup. Move the routines
714 from being local to tls.c to being global via the os.c file.
716 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
717 an extract embedded as result-arg for a map, the first arg for extract
718 is unavailable so we cannot tell if this is a numbered or keyed
719 extraction. Accept either.
725 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
728 JH/02 The smtp transport option "multi_domain" is now expanded.
730 JH/03 The smtp transport now requests PRDR by default, if the server offers
733 JH/04 Certificate name checking on server certificates, when exim is a client,
734 is now done by default. The transport option tls_verify_cert_hostnames
735 can be used to disable this per-host. The build option
736 EXPERIMENTAL_CERTNAMES is withdrawn.
738 JH/05 The value of the tls_verify_certificates smtp transport and main options
739 default to the word "system" to access the system default CA bundle.
740 For GnuTLS, only version 3.0.20 or later.
742 JH/06 Verification of the server certificate for a TLS connection is now tried
743 (but not required) by default. The verification status is now logged by
744 default, for both outbound TLS and client-certificate supplying inbound
747 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
750 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
751 Status Notification (bounce) messages are now MIME format per RFC 3464.
752 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
753 under the control of the dsn_advertise_hosts option, and routers may
754 have a dsn_lasthop option.
756 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
757 default, modifiable by a malware= option. The list separator for
758 the options can now be changed in the usual way. Bug 68.
760 JH/10 The smtp_receive_timeout main option is now expanded before use.
762 JH/11 The incoming_interface log option now also enables logging of the
763 local interface on delivery outgoing connections.
765 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
766 if the interface and destination host and port all match.
768 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
771 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
772 Patch from Andrew Lewis.
774 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
775 now supports optional time-restrictions, weighting, and priority
776 modifiers per server. Patch originally by <rommer@active.by>.
778 JH/16 The spamd_address main option now supports a mixed list of local
779 and remote servers. Remote servers can be IPv6 addresses, and
780 specify a port-range.
782 JH/17 Bug 68: The spamd_address main option now supports an optional
783 timeout value per server.
785 JH/18 Bug 1581: Router and transport options headers_add/remove can
786 now have the list separator specified.
788 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
791 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
794 JH/21 Support for the A6 type of dns record is withdrawn.
796 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
797 rather than the verbs used.
799 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
800 from 255 to 1024 chars.
802 JH/24 Verification callouts now attempt to use TLS by default.
804 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
805 are generic router options now. The defaults didn't change.
807 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
808 Original patch from Alexander Shikoff, worked over by JH.
810 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
811 files if ZCAT_COMMAND is not executable.
813 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
815 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
817 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
818 Normally benign, it bites when the pair was led to by a CNAME;
819 modern usage is to not canonicalize the domain to a CNAME target
820 (and we were inconsistent anyway for A-only vs AAAA+A).
822 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
824 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
825 when evaluating $sender_host_dnssec.
827 JH/31 Check the HELO verification lookup for DNSSEC, adding new
828 $sender_helo_dnssec variable.
830 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
832 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
834 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
836 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
837 documented as working, but never had. Support all but $spam_report.
839 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
840 added for tls authenticator.
842 HS/03 Add perl_taintmode main config option
848 TL/01 When running the test suite, the README says that variables such as
849 no_msglog_check are global and can be placed anywhere in a specific
850 test's script, however it was observed that placement needed to be near
851 the beginning for it to behave that way. Changed the runtest perl
852 script to read through the entire script once to detect and set these
853 variables, reset to the beginning of the script, and then run through
854 the script parsing/test process like normal.
856 TL/02 The BSD's have an arc4random API. One of the functions to induce
857 adding randomness was arc4random_stir(), but it has been removed in
858 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
859 function when detected.
861 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
862 cause callback expansion.
864 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
865 syntax errors in an expansion can be treated as a string instead of
866 logging or causing an error, due to the internal use of bool_lax
867 instead of bool when processing it.
869 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
870 server certificates when making smtp deliveries.
872 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
874 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
876 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
878 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
879 Merged patch from Sebastian Wiedenroth.
881 JH/05 Fix results-pipe from transport process. Several recipients, combined
882 with certificate use, exposed issues where response data items split
883 over buffer boundaries were not parsed properly. This eventually
884 resulted in duplicates being sent. This issue only became common enough
885 to notice due to the introduction of connection certificate information,
886 the item size being so much larger. Found and fixed by Wolfgang Breyha.
888 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
889 size buffer was used, resulting in syntax errors when an expansion
892 JH/07 Add support for directories of certificates when compiled with a GnuTLS
893 version 3.3.6 or later.
895 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
896 is EXPERIMENTAL_EVENT, the main-configuration and transport options
897 both become "event_action", the variables become $event_name, $event_data
898 and $event_defer_errno. There is a new variable $verify_mode, usable in
899 routers, transports and related events. The tls:cert event is now also
900 raised for inbound connections, if the main configuration event_action
903 TL/06 In test suite, disable OCSP for old versions of openssl which contained
904 early OCSP support, but no stapling (appears to be less than 1.0.0).
906 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
907 server certificate names available under the smtp transport option
908 "tls_verify_cert_hostname" now do not permit multi-component wildcard
911 JH/10 Time-related extraction expansions from certificates now use the main
912 option "timezone" setting for output formatting, and are consistent
913 between OpenSSL and GnuTLS compilations. Bug 1541.
915 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
916 encoded parameter in the incoming message. Bug 1558.
918 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
919 include certificate info, eximon was claiming there were spoolfile
922 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
924 JH/14 Log delivery-related information more consistently, using the sequence
925 "H=<name> [<ip>]" wherever possible.
927 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
928 are problematic for Debian distribution, omit them from the release
931 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
933 JH/16 Fix string representation of time values on 64bit time_t architectures.
936 JH/17 Fix a null-indirection in certextract expansions when a nondefault
937 output list separator was used.
942 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
943 checkers that were complaining about end of non-void function with no
946 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
947 This was a regression introduced in 4.83 by another bugfix.
949 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
951 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
952 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
958 TF/01 Correctly close the server side of TLS when forking for delivery.
960 When a message was received over SMTP with TLS, Exim failed to clear up
961 the incoming connection properly after forking off the child process to
962 deliver the message. In some situations the subsequent outgoing
963 delivery connection happened to have the same fd number as the incoming
964 connection previously had. Exim would try to use TLS and fail, logging
965 a "Bad file descriptor" error.
967 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
968 utilities have not been installed.
970 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
971 temporary space as the ACL may create new global variables.
973 TL/01 LDAP support uses per connection or global context settings, depending
974 upon the detected version of the libraries at build time.
976 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
977 to extract and use the src ip:port in logging and expansions as if it
978 were a direct connection from the outside internet. PPv2 support was
979 updated based on HAProxy spec change in May 2014.
981 JH/02 Add ${listextract {number}{list}{success}{fail}}.
983 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
984 Properly escape header and check for NULL return.
986 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
989 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
991 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
992 characters in header names, implemented as a verify condition.
993 Contributed by Michael Fischer v. Mollard.
995 TL/05 Rename SPF condition results err_perm and err_temp to standardized
996 results permerror and temperror. Previous values are deprecated but
997 still accepted. In a future release, err_perm and err_temp will be
998 completely removed, which will be a backward incompatibility if the
999 ACL tests for either of these two old results. Patch contributed by
1000 user bes-internal on the mailing list.
1002 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1004 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1005 selectors, in both main and reject logs.
1007 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1010 JH/07 Add malware type "sock" for talking to simple daemon.
1012 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1014 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1015 routers/transports under cutthrough routing.
1017 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1018 numbers. Touch up "bool" conditional to keep the same definition.
1020 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1022 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1024 JH/12 Expand items in router/transport headers_add or headers_remove lists
1025 individually rather than the list as a whole. Bug 1452.
1027 Required for reasonable handling of multiple headers_ options when
1028 they may be empty; requires that headers_remove items with embedded
1029 colons must have them doubled (or the list-separator changed).
1031 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1032 view the policy declared in the DMARC record. Currently, $dmarc_status
1033 is a combined value of both the record presence and the result of the
1036 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1038 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1039 dnslookup router and the smtp transport (applying to the forward
1042 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1043 of ldap servers used for a specific lookup. Patch provided by Heiko
1046 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1047 New variable $lookup_dnssec_authenticated for observability.
1049 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1050 Patch submitted by Lars Timman.
1052 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1054 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1055 Requires trusted mode and valid format message id, aborts otherwise.
1056 Patch contributed by Heiko Schlichting.
1058 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1059 certextract with support for various fields. Bug 1358.
1061 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1062 is requested by default, modifiable by smtp transport option
1065 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1066 operate on certificate variables to give certificate fingerprints
1067 Also new ${sha256:cert_variable}.
1069 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1071 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1072 Christian Aistleitner.
1074 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1076 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1077 file. Patch from Wolfgang Breyha.
1079 JH/25 Expand the coverage of the delivery $host and $host_address to
1080 client authenticators run in verify callout. Bug 1476.
1082 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1083 align with daemon_smtp_ports. Bug 72.
1085 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1086 support and error reporting did not work properly.
1088 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1089 and is readable. Patch from Andrew Colin Kissa.
1091 TL/14 Enhance documentation of ${run expansion and how it parses the
1092 commandline after expansion, particularly in the case when an
1093 unquoted variable expansion results in an empty value.
1095 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1097 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1098 by renaming away from it. Addresses GH issue 15, reported by
1101 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1103 TL/15 SECURITY: prevent double expansion in math comparison functions
1104 (can expand unsanitized data). Not remotely exploitable.
1111 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1113 PP/02 Make -n do something, by making it not do something.
1114 When combined with -bP, the name of an option is not output.
1116 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1119 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1120 $sender_host_name and config options to manage this, and basic check
1123 PP/05 DSCP support for outbound connections and control modifier for inbound.
1125 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1126 (Only plugin which currently uses this is kerberos4, which nobody should
1127 be using, but we should make it available and other future plugins might
1128 conceivably use it, even though it would break NAT; stuff *should* be
1129 using channel bindings instead).
1131 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1132 name; added for Sendmail compatibility; requires admin caller.
1133 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1134 just ignore it); requires trusted caller.
1135 Also parse but ignore: -Ac -Am -X<logfile>
1138 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1140 TL/02 Add +smtp_confirmation as a default logging option.
1142 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1143 Patch by Magnus Holmgren from 2007-02-20.
1145 TL/04 Bugzilla 1281 - Spec typo.
1146 Bugzilla 1283 - Spec typo.
1147 Bugzilla 1290 - Spec grammar fixes.
1149 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1151 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1153 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1154 mailing list by by Dmitry Isaikin.
1156 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1158 JH/02 Support "G" suffix to numbers in ${if comparisons.
1160 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1162 NM/01 Bugzilla 1197 - Spec typo
1163 Bugzilla 1196 - Spec examples corrections
1165 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1167 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1168 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1171 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1172 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1174 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1175 "acl {{name}{arg}...}", and optional args on acl condition
1178 JH/05 Permit multiple router/transport headers_add/remove lines.
1180 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1182 JH/07 Avoid using a waiting database for a single-message-only transport.
1183 Performance patch from Paul Fisher. Bugzilla 1262.
1185 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1188 JH/09 Add $headers_added variable, with content from use of ACL modifier
1189 add_header (but not yet added to the message). Bugzilla 199.
1191 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1192 Pulled from Bugzilla 817 by Wolfgang Breyha.
1194 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1196 (nb: this is the same fix as in Exim 4.80.1)
1198 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1201 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1204 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1206 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1207 advertise SMTP AUTH mechanism to us, instead of a generic
1208 protocol violation error. Also, make Exim more robust to bad
1209 data from the Dovecot auth socket.
1211 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1213 When a queue runner is handling a message, Exim first routes the
1214 recipient addresses, during which it prunes them based on the retry
1215 hints database. After that it attempts to deliver the message to
1216 any remaining recipients. It then updates the hints database using
1219 So if a recipient address works intermittently, it can get repeatedly
1220 deferred at routing time. The retry hints record remains fresh so the
1221 address never reaches the final cutoff time.
1223 This is a fairly common occurrence when a user is bumping up against
1224 their storage quota. Exim had some logic in its local delivery code
1225 to deal with this. However it did not apply to per-recipient defers
1226 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1228 This change adds a proper retry rule check during routing so that the
1229 final cutoff time is checked against the message's age. We only do
1230 this check if there is an address retry record and there is not a
1231 domain retry record; this implies that previous attempts to handle
1232 the address had the retry_use_local_parts option turned on. We use
1233 this as an approximation for the destination being like a local
1234 delivery, as in LMTP.
1236 I suspect this new check makes the old local delivery cutoff check
1237 redundant, but I have not verified this so I left the code in place.
1239 TF/02 Correct gecos expansion when From: is a prefix of the username.
1241 Test 0254 submits a message to Exim with the header
1245 When I ran the test suite under the user fanf2, Exim expanded
1246 the header to contain my full name, whereas it should have added
1247 a Resent-Sender: header. It erroneously treats any prefix of the
1248 username as equal to the username.
1250 This change corrects that bug.
1252 GF/01 DCC debug and logging tidyup
1253 Error conditions log to paniclog rather than rejectlog.
1254 Debug lines prefixed by "DCC: " to remove any ambiguity.
1256 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1258 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1259 Bug spotted by Jeremy Harris; was flawed since initial commit.
1260 Would have resulted in OCSP responses post-SNI triggering an Exim
1261 NULL dereference and crash.
1263 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1265 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1266 Bug detection, analysis and fix by Samuel Thibault.
1267 Bugzilla 1331, Debian bug #698092.
1269 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1271 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1272 Server implementation by Todd Lyons, client by JH.
1273 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1274 config variable "prdr_enable" controls whether the server
1275 advertises the facility. If the client requests PRDR a new
1276 acl_data_smtp_prdr ACL is called once for each recipient, after
1277 the body content is received and before the acl_smtp_data ACL.
1278 The client is controlled by both of: a hosts_try_prdr option
1279 on the smtp transport, and the server advertisement.
1280 Default client logging of deliveries and rejections involving
1281 PRDR are flagged with the string "PRDR".
1283 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1284 fclose(). Diagnosis by Todd Lyons.
1286 PP/17 Update configure.default to handle IPv6 localhost better.
1287 Patch by Alain Williams (plus minor tweaks).
1290 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1291 This is now consistent with GnuTLS, and is now documented: the
1292 previous undocumented portable approach to treating the option as
1293 unset was to force an expansion failure. That still works, and
1294 an empty string is now equivalent.
1296 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1297 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1298 not performing validation itself.
1300 PP/20 Added force_command boolean option to pipe transport.
1301 Patch from Nick Koston, of cPanel Inc.
1303 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1306 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1308 PP/21 Fix eximon continuous updating with timestamped log-files.
1309 Broken in a format-string cleanup in 4.80, missed when I repaired the
1310 other false fix of the same issue.
1311 Report and fix from Heiko Schlichting.
1314 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1315 Report from Prashanth Katuri.
1317 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1318 It's SecureTransport, so affects any MacOS clients which use the
1319 system-integrated TLS libraries, including email clients.
1321 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1322 using a MIME ACL for non-SMTP local injection.
1323 Report and assistance in diagnosis by Warren Baker.
1325 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1327 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1329 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1330 last id that failed so it may be referenced in subsequent ACL's.
1332 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1335 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1336 ldap library initialization, allowing self-signed CA's to be
1337 used. Also properly sets require_cert option later in code by
1338 using NULL (global ldap config) instead of ldap handle (per
1339 session). Bug diagnosis and testing by alxgomz.
1341 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1342 the src/util/ subdirectory.
1344 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1345 renamed to Transport Post Delivery Action by Jeremy Harris, as
1348 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1349 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1350 redis_servers = needs to be configured which will be used by the redis
1351 lookup. Patch from Warren Baker, of The Packet Hub.
1353 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1355 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1356 hostname or reverse DNS when processing a host list. Used suggestions
1357 from multiple comments on this bug.
1359 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1361 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1362 Missed a few lines, added it to make the runtest require no keyboard
1365 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1366 contains upper case chars. Make router use caseful_local_part.
1368 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1369 support when GnuTLS has been built with p11-kit.
1375 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1377 This, or similar/improved, will also be change PP/11 of 4.82.
1383 PP/01 Handle short writes when writing local log-files.
1384 In practice, only affects FreeBSD (8 onwards).
1385 Bugzilla 1053, with thanks to Dmitry Isaikin.
1387 NM/01 Bugzilla 949 - Documentation tweak
1389 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1392 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1394 PP/02 Implemented gsasl authenticator.
1396 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1398 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1399 `pkg-config foo` for cflags/libs.
1401 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1402 with rest of GSASL and with heimdal_gssapi.
1404 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1405 `pkg-config foo` for cflags/libs for the TLS implementation.
1407 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1408 properties get this fed in as external SSF. A number of robustness
1409 and debugging improvements to the cyrus_sasl authenticator.
1411 PP/08 cyrus_sasl server now expands the server_realm option.
1413 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1414 Patch by Jeremy Harris.
1416 PP/10 Added dbmjz lookup type.
1418 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1420 PP/12 MAIL args handles TAB as well as SP, for better interop with
1421 non-compliant senders.
1422 Analysis and variant patch by Todd Lyons.
1424 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1425 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1426 Patch from Dirk Mueller <dmueller@suse.com>
1428 PP/13 tls_peerdn now print-escaped for spool files.
1429 Observed some $tls_peerdn in wild which contained \n, which resulted
1430 in spool file corruption.
1432 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1433 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1434 or write after TLS renegotiation, which otherwise led to messages
1437 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1438 as a tracking header (ie: a signed header comes before the signature).
1439 Patch from Wolfgang Breyha.
1441 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1442 comma-sep list; embedded commas doubled.
1444 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1446 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1448 Report and patch from Dmitry Banschikov.
1450 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1451 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1452 failures. We appear to now support TLS1.1+ with Exim.
1454 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1455 lets Exim select keys and certificates based upon TLS SNI from client.
1456 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1457 before an outbound SMTP session. New log_selector, +tls_sni.
1459 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1460 NULL dereference. Report and patch from Alun Jones.
1462 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1463 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1464 Not seeing resolver debug output on NetBSD, but suspect this is a
1465 resolver implementation change.
1467 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1468 Left warnings. Added "eximon gdb" invocation mode.
1470 PP/21 Defaulting "accept_8bitmime" to true, not false.
1472 PP/22 Added -bw for inetd wait mode support.
1474 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1475 locate the relevant includes and libraries. Made this the default.
1477 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1478 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1480 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1481 This may cause build issues on older platforms.
1483 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1484 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1485 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1486 Added SNI support via GnuTLS too.
1487 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1489 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1491 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1492 Applied second patch from Janne, implementing suggestion to default
1493 multiple-strings-in-record handling to match SPF spec.
1495 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1497 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1498 read-only, out of scope).
1499 Patch from Wolfgang Breyha, report from Stuart Northfield.
1501 PP/29 Fix three issues highlighted by clang analyser static analysis.
1502 Only crash-plausible issue would require the Cambridge-specific
1503 iplookup router and a misconfiguration.
1504 Report from Marcin Mirosław.
1506 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1508 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1509 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1510 As part of this, removing so much warning spew let me fix some minor
1511 real issues in debug logging.
1513 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1514 assignment on my part. Fixed.
1516 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1517 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1518 Janne Snabb (who went above and beyond: thank you).
1520 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1521 string otherwise requires a connection and a bunch more work and it's
1522 relatively easy to get wrong. Should also expose TLS library linkage
1525 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1526 64-bit ${eval} (JH/03).
1528 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1529 GNU libc to support some of the 64-bit stuff, should not lead to
1530 conflicts. Defined before os.h is pulled in, so if a given platform
1531 needs to override this, it can.
1533 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1534 protection layer was required, which is not implemented.
1535 Bugzilla 1254, patch from Wolfgang Breyha.
1537 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1538 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1539 tls_dhparam take prime identifiers. Also unbreak combination of
1540 OpenSSL+DH_params+TLSSNI.
1542 PP/39 Disable SSLv2 by default in OpenSSL support.
1548 PP/01 Solaris build fix for Oracle's LDAP libraries.
1549 Bugzilla 1109, patch from Stephen Usher.
1551 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1553 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1556 TF/02 Fix a couple more cases where we did not log the error message
1557 when unlink() failed. See also change 4.74-TF/03.
1559 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1560 lock up or crash if it happened to be inside a call to libc when it
1561 got a SIGUSR1 from exiwhat.
1563 The SIGUSR1 handler appends the current process status to the process
1564 log which is later printed by exiwhat. It used to use the general
1565 purpose logging code to do this, but several functions it calls are
1566 not safe for signals.
1568 The new output code in the SIGUSR1 handler is specific to the process
1569 log, and simple enough that it's easy to inspect for signal safety.
1570 Removing some special cases also simplifies the general logging code.
1571 Removing the spurious timestamps from the process log simplifies
1574 TF/04 Improved ratelimit ACL condition.
1576 The /noupdate option has been deprecated in favour of /readonly which
1577 has clearer semantics. The /leaky, /strict, and /readonly update modes
1578 are mutually exclusive. The update mode is no longer included in the
1579 database key; it just determines when the database is updated. (This
1580 means that when you upgrade Exim will forget old rate measurements.)
1582 Exim now checks that the per_* options are used with an update mode that
1583 makes sense for the current ACL. For example, when Exim is processing a
1584 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1585 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1586 must specify per_mail/readonly. If you omit the update mode it defaults to
1587 /leaky where that makes sense (as before) or /readonly where required.
1589 The /noupdate option is now undocumented but still supported for
1590 backwards compatibility. It is equivalent to /readonly except that in
1591 ACLs where /readonly is required you may specify /leaky/noupdate or
1592 /strict/noupdate which are treated the same as /readonly.
1594 A useful new feature is the /count= option. This is a generalization
1595 of the per_byte option, so that you can measure the throughput of other
1596 aggregate values. For example, the per_byte option is now equivalent
1597 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1599 The per_rcpt option has been generalized using the /count= mechanism
1600 (though it's more complicated than the per_byte equivalence). When it is
1601 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1602 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1603 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1604 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1605 Note that using per_rcpt with a non-readonly update mode in more than
1606 one ACL will cause the recipients to be double-counted. (The per_mail
1607 and per_byte options don't have this problem.)
1609 The handling of very low rates has changed slightly. If the computed rate
1610 is less than the event's count (usually one) then this event is the first
1611 after a long gap. In this case the rate is set to the same as this event's
1612 count, so that the first message of a spam run is counted properly.
1614 The major new feature is a mechanism for counting the rate of unique
1615 events. The new per_addr option counts the number of different
1616 recipients that someone has sent messages to in the last time period. It
1617 behaves like per_rcpt if all the recipient addresses are different, but
1618 duplicate recipient addresses do not increase the measured rate. Like
1619 the /count= option this is a general mechanism, so the per_addr option
1620 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1621 example, measure the rate that a client uses different sender addresses
1622 with the options per_mail/unique=$sender_address. There are further
1623 details in the main documentation.
1625 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1627 TF/06 Removed a few PCRE remnants.
1629 TF/07 Automatically extract Exim's version number from tags in the git
1630 repository when doing development or release builds.
1632 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1633 Bugzilla 879. Patch from Paul Fisher.
1635 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1636 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1639 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1641 PP/05 Variable $av_failed, true if the AV scanner deferred.
1642 Bugzilla 1078. Patch from John Horne.
1644 PP/06 Stop make process more reliably on build failure.
1645 Bugzilla 1087. Patch from Heiko Schlittermann.
1647 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1648 Bugzilla 1089. Patch from Heiko Schlittermann.
1650 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1651 Bugzilla 1131. Patch from Holger Weiß.
1653 PP/09 Handle IPv6 addresses with SPF.
1654 Bugzilla 860. Patch from Wolfgang Breyha.
1656 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1658 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1661 PP/11 match_* no longer expand right-hand-side by default.
1662 New compile-time build option, EXPAND_LISTMATCH_RHS.
1663 New expansion conditions, "inlist", "inlisti".
1665 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1667 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1669 PP/14 fix log_write() format string regression from TF/03.
1670 Bugzilla 1152. Patch from Dmitry Isaikin.
1676 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1678 PP/02 Harmonised TLS library version reporting; only show if debugging.
1679 Layout now matches that introduced for other libraries in 4.74 PP/03.
1681 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1683 PP/04 New "dns_use_edns0" global option.
1685 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1688 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1689 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1691 TK/01 Updated PolarSSL code to 0.14.2.
1692 Bugzilla 1097. Patch from Andreas Metzler.
1694 PP/07 Catch divide-by-zero in ${eval:...}.
1695 Fixes bugzilla 1102.
1697 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1700 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1701 format-string attack -- SECURITY: remote arbitrary code execution.
1703 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1704 time unintentionally subject to list matching rules, letting the header
1705 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1706 arbitrary string expansion). This allowed for information disclosure.
1708 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1709 INT_MIN/-1 -- value coerced to INT_MAX.
1715 NM/01 Workaround for PCRE version dependency in version reporting
1718 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1719 This fixes portability to compilers other than gcc, notably
1720 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1722 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1723 makefiles for portability to HP-UX and POSIX correctness.
1725 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1726 Also via indented variable definition in the Makefile.
1727 (Debugging by Oliver Heesakkers).
1729 PP/02 Restore caching of spamd results with expanded spamd_address.
1730 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1732 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1733 Improves build reliability. Fix from: Frank Elsner
1735 NM/02 Fix wide character breakage in the rfc2047 coding
1736 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1738 NM/03 Allow underscore in dnslist lookups
1739 Fixes bug 1026. Patch from Graeme Fowler
1741 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1742 Code patches from Adam Ciarcinski of NetBSD.
1744 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1747 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1748 is logged, to avoid truncation. Patch from John Horne.
1750 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1751 Patch from Jakob Hirsch.
1753 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1754 SQL string expansion failure details.
1755 Patch from Andrey Oktyabrski.
1757 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1758 Patch from Simon Arlott.
1760 PP/09 New lookups functionality failed to compile on old gcc which rejects
1761 extern declarations in function scope.
1762 Patch from Oliver Fleischmann
1764 PP/10 Use sig_atomic_t for flags set from signal handlers.
1765 Check getgroups() return and improve debugging.
1766 Fixed developed for diagnosis in bug 927 (which turned out to be
1769 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1770 Patch from Mark Zealey.
1772 PP/12 Bugzilla 1056: Improved spamd server selection.
1773 Patch from Mark Zealey.
1775 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1776 Based on patch from Heiko Schlittermann.
1778 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1779 Patch from Uwe Doering, sign-off by Michael Haardt.
1781 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1782 variable declaration deep within a block. Bug and patch from
1785 PP/15 lookups-Makefile IRIX compatibility coercion.
1787 PP/16 Make DISABLE_DKIM build knob functional.
1789 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1790 Patch by Simon Arlott
1792 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1793 variable argument macros. Our copy now differs from upstream.
1799 TF/01 Failure to get a lock on a hints database can have serious
1800 consequences so log it to the panic log.
1802 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1803 controlled using the smtp_confirmation log selector.
1805 TF/03 Include the error message when we fail to unlink a spool file.
1807 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1808 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1809 for maintaining out-of-tree patches for some time.
1811 PP/01 Bugzilla 139: Documentation and portability issues.
1812 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1813 Handle per-OS dynamic-module compilation flags.
1815 PP/02 Let /dev/null have normal permissions.
1816 The 4.73 fixes were a little too stringent and complained about the
1817 permissions on /dev/null. Exempt it from some checks.
1818 Reported by Andreas M. Kirchwitz.
1820 PP/03 Report version information for many libraries, including
1821 Exim version information for dynamically loaded libraries. Created
1822 version.h, now support a version extension string for distributors
1823 who patch heavily. Dynamic module ABI change.
1825 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1826 privilege escalation vulnerability whereby the Exim run-time user
1827 can cause root to append content of the attacker's choosing to
1830 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1833 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1834 If dropping privileges for untrusted macros, we disabled normal logging
1835 on the basis that it would fail; for the Exim run-time user, this is not
1836 the case, and it resulted in successful deliveries going unlogged.
1837 Fixed. Reported by Andreas Metzler.
1843 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1844 only prepend for the Resent-* case. Fixes regression introduced in
1845 Exim 4.70 by NM/22 for Bugzilla 607.
1847 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1848 increasing numbers of administrators be bitten by this.
1850 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1852 PP/03 Bugzilla 994: added openssl_options main configuration option.
1854 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1856 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1858 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1860 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1861 without a peer certificate, leading to a segfault because of an
1862 assumption that peers always have certificates. Be a little more
1863 paranoid. Problem reported by Martin Tscholak.
1865 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1866 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1867 NB: ClamAV planning to remove STREAM in "middle of 2010".
1868 CL also introduces -bmalware, various -d+acl logging additions and
1869 more caution in buffer sizes.
1871 PP/09 Implemented reverse_ip expansion operator.
1873 PP/10 Bugzilla 937: provide a "debug" ACL control.
1875 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1877 PP/12 Bugzilla 973: Implement --version.
1879 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1881 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1883 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1885 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1886 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1887 ignore trailing whitespace.
1889 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1891 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1894 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1895 Notification from Dr Andrew Aitchison.
1897 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1898 ExtendedDetectionInfo response format.
1899 Notification from John Horne.
1901 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1904 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1905 XSL and documented dependency on system catalogs, with examples of how
1908 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1911 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1912 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1913 configuration file which is writeable by the Exim user or group.
1915 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1916 of configuration files to cover files specified with the -C option if
1917 they are going to be used with root privileges, not just the default
1920 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1921 option (effectively making it always true).
1923 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1924 files to be used while preserving root privileges.
1926 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1927 that rogue child processes cannot use them.
1929 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1930 run-time user, instead of root.
1932 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1933 Exim run-time user without dropping privileges.
1935 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1936 result string, instead of calling string_vformat() twice with the same
1939 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1940 for other users. Others should always drop root privileges if they use
1941 -C on the command line, even for a whitelisted configure file.
1943 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1945 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1951 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1952 $data_path, and $header_path variables; fixed documentation bugs and
1955 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1956 exipick to access non-standard spools, including the "frozen" queue
1959 NM/01 Bugzilla 965: Support mysql stored procedures.
1960 Patch from Alain Williams
1962 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1964 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1965 Patch from Andreas Metzler
1967 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1968 Patch from Kirill Miazine
1970 NM/05 Bugzilla 671: Added umask to procmail example.
1972 JJ/03 installed exipick 20100323.0, fixing doc bug
1974 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1975 directory. Notification and patch from Dan Rosenberg.
1977 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1979 TK/02 Improve log output when DKIM signing operation fails.
1981 MH/01 Treat the transport option dkim_domain as a colon separated
1982 list, not as a single string, and sign the message with each element,
1983 omitting multiple occurences of the same signer.
1985 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1986 Bugzilla 985, 986. Patch by Simon Arlott
1988 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1989 Patch by Simon Arlott
1991 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1992 MBX locking. Notification from Dan Rosenberg.
1998 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2000 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2002 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2004 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2006 NM/04 Bugzilla 915: Flag broken perl installation during build.
2012 TK/01 Added patch by Johannes Berg that expands the main option
2013 "spamd_address" if it starts with a dollar sign.
2015 TK/02 Write list of recipients to X-Envelope-Sender header when building
2016 the mbox-format spool file for content scanning (suggested by Jakob
2019 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2020 (http://www.dcc-servers.net/) support via dccifd. Activated by
2021 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2023 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2024 by Mark Daniel Reidel <mr@df.eu>.
2026 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2027 When building exim an external PCRE library is now needed -
2028 PCRE is a system library on the majority of modern systems.
2029 See entry on PCRE_LIBS in EDITME file.
2031 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2032 conversation. Added nologin parameter to request.
2033 Patch contributed by Kirill Miazine.
2035 TF/01 Do not log submission mode rewrites if they do not change the address.
2037 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2039 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2040 log files in place. Contributed by Roberto Lima.
2042 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2044 TF/03 Bugzilla 615: When checking the local_parts router precondition
2045 after a local_part_suffix or local_part_prefix option, Exim now
2046 does not use the address's named list lookup cache, since this
2047 contains cached lookups for the whole local part.
2049 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2050 Robert Millan. Documentation is in experimental-spec.txt.
2052 TF/04 Bugzilla 668: Fix parallel build (make -j).
2054 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2056 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2057 Patch provided by Jan Srzednicki.
2059 TF/05 Leading white space used to be stripped from $spam_report which
2060 wrecked the formatting. Now it is preserved.
2062 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2063 that they are available at delivery time.
2065 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2067 TF/08 TLS error reporting now respects the incoming_interface and
2068 incoming_port log selectors.
2070 TF/09 Produce a more useful error message if an SMTP transport's hosts
2071 setting expands to an empty string.
2073 NM/06 Bugzilla 744: EXPN did not work under TLS.
2074 Patch provided by Phil Pennock.
2076 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2077 Patch provided by Richard Godbee.
2079 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2080 acl_smtp_notquit, added index entry.
2082 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2083 Patch provided by Eugene Bujak.
2085 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2086 accept(). Patch provided by Maxim Dounin.
2088 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2089 Patch provided by Phil Pennock.
2091 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2093 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2094 Patch provided by Brad "anomie" Jorsch.
2096 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2097 Patch provided by Dean Brooks.
2099 TK/05 Add native DKIM support (does not depend on external libraries).
2101 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2102 Patch provided by Graeme Fowler.
2104 NM/16 Bugzilla 851: Documentation example syntax fix.
2106 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2108 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2111 NM/19 Bugzilla 745: TLS version reporting.
2112 Patch provided by Phil Pennock.
2114 NM/20 Bugzilla 167: bool: condition support.
2115 Patch provided by Phil Pennock.
2117 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2118 clients. Patch provided by Phil Pennock.
2120 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2121 Patch provided by Brad "anomie" Jorsch.
2123 NM/23 Bugzilla 687: Fix misparses in eximstats.
2124 Patch provided by Heiko Schlittermann.
2126 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2127 Patch provided by Heiko Schlittermann.
2129 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2130 plus update to original patch.
2132 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2134 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2135 Patch provided by David Brownlee.
2137 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2139 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2141 NM/30 Bugzilla 888: TLS documentation bugfixes.
2143 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2145 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2146 Unlike the original bugzilla I have changed all shell scripts in src tree.
2148 NM/33 Bugzilla 898: Transport filter timeout fix.
2149 Patch by Todd Rinaldo.
2151 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2152 Patch by Serge Demonchaux.
2154 NM/35 Bugzilla 39: Base64 decode bug fixes.
2155 Patch by Jakob Hirsch.
2157 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2159 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2161 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2163 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2169 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2170 ALT-N's libdkim that I have put here:
2171 http://duncanthrax.net/exim-experimental/
2173 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2174 were called 'true' and it seems that C99 defines that as a reserved
2175 keyword to be used with 'bool' variable types. That means you could
2176 not include C99-style headers which use bools without triggering
2177 build errors in sieve.c.
2179 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2180 as mailq or other aliases. Changed the --help handling significantly
2181 to do whats expected. exim_usage() emits usage/help information.
2183 SC/01 Added the -bylocaldomain option to eximstats.
2185 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2187 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2189 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2195 PH/01 Another patch from the Sieve maintainer.
2197 PH/02 When an IPv6 address is converted to a string for single-key lookup
2198 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2199 dots are used instead of colons so that keys in lsearch files need not
2200 contain colons. This was done some time before quoting was made available
2201 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2202 (notated using the quote facility) so as to distinguish them from IPv4
2203 keys. This meant that lookups for IP addresses in host lists did not work
2204 for iplsearch lookups.
2206 This has been fixed by arranging for IPv6 addresses to be expressed with
2207 colons if the lookup type is iplsearch. This is not incompatible, because
2208 previously such lookups could never work.
2210 The situation is now rather anomalous, since one *can* have colons in
2211 ordinary lsearch keys. However, making the change in all cases is
2212 incompatible and would probably break a number of configurations.
2214 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2217 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2218 conversion specification without a maximum field width, thereby enabling
2219 a rogue spamd server to cause a buffer overflow. While nobody in their
2220 right mind would setup Exim to query an untrusted spamd server, an
2221 attacker that gains access to a server running spamd could potentially
2222 exploit this vulnerability to run arbitrary code as the Exim user.
2224 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2225 $primary_hostname instead of what libspf2 thinks the hosts name is.
2227 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2228 a directory entry by the name of the lookup key. Previously, if a
2229 symlink pointed to a non-existing file or a file in a directory that
2230 Exim lacked permissions to read, a lookup for a key matching that
2231 symlink would fail. Now it is enough that a matching directory entry
2232 exists, symlink or not. (Bugzilla 503.)
2234 PH/03 The body_linecount and body_zerocount variables are now exported in the
2237 PH/04 Added the $dnslist_matched variable.
2239 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2240 This means they are set thereafter only if the connection becomes
2243 PH/06 Added the client_condition to authenticators so that some can be skipped
2244 by clients under certain conditions.
2246 PH/07 The error message for a badly-placed control=no_multiline_responses left
2247 "_responses" off the end of the name.
2249 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2251 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2252 (without spaces) instead of just copying the configuration text.
2254 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2256 PH/11 Added $max_received_linelength.
2258 PH/12 Added +ignore_defer and +include_defer to host lists.
2260 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2261 way in which PCRE > 7.0 is built.
2263 PH/14 Implemented queue_only_load_latch.
2265 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2266 MAIL command. The effect was to mangle the value on 64-bit systems.
2268 PH/16 Another patch from the Sieve maintainer.
2270 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2272 PH/18 If a system quota error occurred while trying to create the file for
2273 a maildir delivery, the message "Mailbox is full" was not appended to the
2274 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2275 only to a quota excession during the actual writing of the file.
2277 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2278 characters?) which causes problems in log lines. The DN values are now
2279 passed through string_printing() before being added to log lines.
2281 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2282 and InterBase are left for another time.)
2284 PH/21 Added message_body_newlines option.
2286 PH/22 Guard against possible overflow in moan_check_errorcopy().
2288 PH/23 POSIX allows open() to be a macro; guard against that.
2290 PH/24 If the recipient of an error message contained an @ in the local part
2291 (suitably quoted, of course), incorrect values were put in $domain and
2292 $local_part during the evaluation of errors_copy.
2298 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2299 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2302 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2303 issue a MAIL command.
2305 PH/02 In an ACL statement such as
2307 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2309 if a client was not listed at all, or was listed with a value other than
2310 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2311 the condition was not true (as it should be), so access was not denied.
2312 The bug was that the ! inversion was incorrectly passed on to the second
2313 item. This has been fixed.
2315 PH/03 Added additional dnslists conditions == and =& which are different from
2316 = and & when the dns lookup returns more than one IP address.
2318 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2319 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2321 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2322 FSYNC, which compiles an option called disable_fsync that allows for
2323 bypassing fsync(). The documentation is heavily laced with warnings.
2325 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2327 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2328 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2329 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2330 including adding "make clean"; (3) Added -fPIC when compiling the test
2331 dynamically loaded module, to get rid of a warning.
2333 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2334 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2335 = 0s. The bug is otherwise harmless.
2337 PH/07 There was a bug in the dovecot authenticator such that the value of
2338 $auth1 could be overwritten, and so not correctly preserved, after a
2339 successful authentication. This usually meant that the value preserved by
2340 the server_setid option was incorrect.
2342 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2344 PH/09 Installed PCRE release 7.0.
2346 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2347 run for batched SMTP input. It is now run at the start of every message
2348 in the batch. While fixing this I discovered that the process information
2349 (output by running exiwhat) was not always getting set for -bs and -bS
2350 input. This is fixed, and it now also says "batched" for BSMTP.
2352 PH/11 Added control=no_pipelining.
2354 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2355 patch, slightly modified), and move the expansion of helo_data till after
2356 the connection is made in the smtp transport (so it can use these
2359 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2361 PH/14 Added log_selector = +pid.
2363 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2365 PH/16 Add ${if forany and ${if forall.
2367 PH/17 Added dsn_from option to vary the From: line in DSNs.
2369 PH/18 Flush SMTP output before performing a callout, unless control =
2370 no_callout_flush is set.
2372 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2373 was true (the default) a successful delivery failed to delete the retry
2374 item, thus causing premature timeout of the address. The bug is now
2377 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2379 PH/21 Long custom messages for fakedefer and fakereject are now split up
2380 into multiline responses in the same way that messages for "deny" and
2381 other ACL rejections are.
2383 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2384 with slight modification.
2386 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2387 draft, changing the syntax and factoring some duplicate code".
2389 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2390 for deliveries of the second and subsequent messages over the same SMTP
2393 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2394 ${reduce, with only minor "tidies".
2396 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2398 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2399 expansion side effects.
2401 PH/27 When a message times out after an over-quota error from an Exim-imposed
2402 quota, the bounce message says "mailbox is full". This message was not
2403 being given when it was a system quota that was exceeded. It now should
2406 MH/03 Made $recipients available in local_scan(). local_scan() already has
2407 better access to the recipient list through recipients_list[], but
2408 $recipients can be useful in postmaster-provided expansion strings.
2410 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2411 in the case of a MAIL command with additional options following the
2412 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2413 were accidentally chopped off.
2415 PH/29 SMTP synchronization checks are implemented when a command is read -
2416 there is a check that no more input is waiting when there shouldn't be
2417 any. However, for some commands, a delay in an ACL can mean that it is
2418 some time before the response is written. In this time, more input might
2419 arrive, invalidly. So now there are extra checks after an ACL has run for
2420 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2421 pipelining has not been advertised.
2423 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2425 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2426 returned in the SMTP response when smtp_return_error_details was false.
2427 This has been fixed.
2429 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2430 instead of the C I/O that was originally supplied, because problems were
2431 reported on Solaris.
2433 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2434 Exim which did not show up earlier: it was assuming that a call to
2435 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2436 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2437 was a macro that became an assignment, so it seemed to work. This has
2438 changed to a proper function call with a void return, hence the compile
2439 error. Exim's code has been fixed.
2441 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2444 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2446 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2448 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2449 because it turns out to be too big or there is a timeout) there is no
2450 "Completed" line in the log. When some messages of this type were
2451 selected by exigrep, they were listed as "not completed". Others were
2452 picked up by some special patterns. I have improved the selection
2453 criteria to be more general.
2455 PH/38 The host_find_failed option in the manualroute router can now be set
2456 to "ignore", to completely ignore a host whose IP address cannot be
2457 found. If all hosts are ignored, the behaviour is controlled by the new
2458 host_all_ignored option.
2460 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2461 homing or because of multiple MX records with /mx) generated more than
2462 one IP address, and the following item turned out to be the local host,
2463 all the secondary addresses of the first item were incorrectly removed
2464 from the list, along with the local host and any following hosts (which
2465 is what is supposed to happen).
2467 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2468 whoever called Exim into the -H file. In the case of the daemon it was
2469 behaving confusingly. When first started, it used values for whoever
2470 started the daemon, but after a SIGHUP it used the Exim user (because it
2471 calls itself on a restart). I have changed the code so that it now always
2474 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2475 message are rejected with the same error (e.g. no authentication or bad
2476 sender address), and a DATA command is nevertheless sent (as can happen
2477 with PIPELINING or a stupid MUA), the error message that was given to the
2478 RCPT commands is included in the rejection of the DATA command. This is
2479 intended to be helpful for MUAs that show only the final error to their
2482 PH/42 Another patch from the Sieve maintainer.
2484 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2485 Eximstats - Fixed some broken HTML links and added missing column headers
2487 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2488 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2490 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2492 PH/43 Yet another patch from the Sieve maintainer.
2494 PH/44 I found a way to check for a TCP/IP connection going away before sending
2495 the response to the final '.' that terminates a message, but only in the
2496 case where the client has not sent further data following the '.'
2497 (unfortunately, this is allowed). However, in many cases there won't be
2498 any further data because there won't be any more messages to send. A call
2499 to select() can be used: if it shows that the input is "ready", there is
2500 either input waiting, or the socket has been closed. An attempt to read
2501 the next input character can distinguish the two cases. Previously, Exim
2502 would have sent an OK response which the client would never have see.
2503 This could lead to message repetition. This fix should cure that, at
2504 least in a lot of common cases.
2506 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2507 advertised in response to EHLO.
2513 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2514 fixed by 4.65/MH/01 (is this a record?) are fixed:
2516 (i) An empty string was always treated as zero by the numeric comparison
2517 operators. This behaviour has been restored.
2519 (ii) It is documented that the numeric comparison operators always treat
2520 their arguments as decimal numbers. This was broken in that numbers
2521 starting with 0 were being interpreted as octal.
2523 While fixing these problems I realized that there was another issue that
2524 hadn't been noticed. Values of message_size_limit (both the global option
2525 and the transport option) were treated as octal if they started with 0.
2526 The documentation was vague. These values are now always treated as
2527 decimal, and I will make that clear in the documentation.
2533 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2534 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2537 MH/01 Don't check that the operands of numeric comparison operators are
2538 integers when their expansion is in "skipping" mode (fixes bug
2539 introduced by 4.64-PH/07).
2541 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2542 child addresses, Exim now panics and dies. Previously, because the count
2543 is held in a short int, deliveries were likely to be lost. As such a
2544 large number of recipients for a single message is ridiculous
2545 (performance will be very, very poor), I have chosen to impose a limit
2546 rather than extend the field.
2552 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2553 leftover -K file (the existence of which was triggered by #402).
2554 While we were at it, introduced process PID as part of the -K
2555 filename. This should rule out race conditions when creating
2558 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2559 processing considerably. Previous code took too long for large mails,
2560 triggering a timeout which in turn triggers #401.
2562 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2563 in the DK code in transports.c. sendfile() is not really portable,
2564 hence the _LINUX specificness.
2566 TF/01 In the add_headers option to the mail command in an Exim filter,
2567 there was a bug that Exim would claim a syntax error in any
2568 header after the first one which had an odd number of characters
2571 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2572 callout verification, Exim cached a "reject" for the entire domain. This
2573 is correct for most verifications, but it is not correct for a recipient
2574 verification with use_sender or use_postmaster set, because in that case
2575 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2576 case of MAIL FROM:<> rejection from other early rejections (e.g.
2577 rejection of HELO). When verifying a recipient using a non-null MAIL
2578 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2579 Whatever the result of the callout, the value of the domain cache is
2580 left unchanged (for any other kind of callout, getting as far as trying
2581 RCPT means that the domain itself is ok).
2583 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2586 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2587 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2588 session, the daemon ignores ECONNECT errors and logs others; it now
2589 ignores EPIPE as well.
2591 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2592 (quoted-printable decoding).
2594 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2595 later the small subsequent patch to fix an introduced bug.
2597 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2599 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2601 PH/08 An error is now given if message_size_limit is specified negative.
2603 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2604 to be given (somewhat) arbitrary names.
2606 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2609 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2610 miscellaneous code fixes
2612 PH/10 Added the log_reject_target ACL modifier to specify where to log
2615 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2616 hostname. This is wrong, because it relates to the incoming message (and
2617 probably the interface on which it is arriving) and not to the outgoing
2618 callout (which could be using a different interface). This has been
2619 changed to use the value of the helo_data option from the smtp transport
2620 instead - this is what is used when a message is actually being sent. If
2621 there is no remote transport (possible with a router that sets up host
2622 addresses), $smtp_active_hostname is used.
2624 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2625 tweaks were necessary in order to get it to work (see also 21 below):
2626 (a) The code assumed that strncpy() returns a negative number on buffer
2627 overflow, which isn't the case. Replaced with Exim's string_format()
2629 (b) There were several signed/unsigned issues. I just did the minimum
2630 hacking in of casts. There is scope for a larger refactoring.
2631 (c) The code used strcasecmp() which is not a standard C function.
2632 Replaced with Exim's strcmpic() function.
2633 (d) The code set only $1; it now sets $auth1 as well.
2634 (e) A simple test gave the error "authentication client didn't specify
2635 service in request". It would seem that Dovecot has changed its
2636 interface. Fortunately there's a specification; I followed it and
2637 changed what the client sends and it appears to be working now.
2639 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2642 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2643 address A is aliased to B and C, where B exists and C does not. Without
2644 -v the output is "A verified" because verification stops after a
2645 successful redirection if more than one address is generated. However,
2646 with -v the child addresses are also verified. Exim was outputting "A
2647 failed to verify" and then showing the successful verification for C,
2648 with its parentage. It now outputs "B failed to verify", showing B's
2649 parentage before showing the successful verification of C.
2651 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2652 look up a TXT record in a specific list after matching in a combined
2655 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2656 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2657 they consult the DNS. I had assumed they would set it the way they
2658 wanted; and indeed my experiments on Linux seem to show that in some
2659 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2660 To be on the safe side, however, I have now made the interface to
2661 host_find_byname() similar to host_find_bydns(), with an argument
2662 containing the DNS resolver options. The host_find_byname() function now
2663 sets these options at its start, just as host_find_bydns() does. The smtp
2664 transport options dns_qualify_single and dns_search_parents are passed to
2665 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2666 of host_find_byname() use the default settings of RES_DEFNAMES
2667 (qualify_single) but not RES_DNSRCH (search_parents).
2669 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2670 spool_read_header() do less string testing, by means of a preliminary
2671 switch on the second character of optional "-foo" lines. (This is
2672 overdue, caused by the large number of possibilities that now exist.
2673 Originally there were few.) While I was there, I also converted the
2674 str(n)cmp tests so they don't re-test the leading "-" and the first
2675 character, in the hope this might squeeze out yet more improvement.
2677 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2678 flag allowing group syntax was set by the header_syntax check but not
2679 turned off, possible causing trouble later; (2) The flag was not being
2680 set at all for the header_verify test, causing "group"-style headers to
2681 be rejected. I have now set it in this case, and also caused header_
2682 verify to ignore an empty address taken from a group. While doing this, I
2683 came across some other cases where the code for allowing group syntax
2684 while scanning a header line wasn't quite right (mostly, not resetting
2685 the flag correctly in the right place). These bugs could have caused
2686 trouble for malformed header lines. I hope it is now all correct.
2688 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2689 with the "reply" argument non-NULL. The code, however (which originally
2690 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2691 but it didn't always do it. This confused somebody who was copying the
2692 code for some other use. I have removed all the tests.
2694 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2695 feature that was used to support insecure browsers during the U.S. crypto
2696 embargo. It requires special client support, and Exim is probably the
2697 only MTA that supported it -- and would never use it because real RSA is
2698 always available. This code has been removed, because it had the bad
2699 effect of slowing Exim down by computing (never used) parameters for the
2700 RSA_EXPORT functionality.
2702 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2703 authenticator to fail if there's a tab character in the incoming data
2704 (there should never be unless someone is messing about, as it's supposed
2705 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2706 if the connection is using TLS or if the remote IP is the same as the
2707 local IP, and the "valid-client-cert option" if a client certificate has
2710 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2711 authenticators. This can be used for authorization after authentication
2712 succeeds. (In the case of plaintext, it servers for both authentication
2715 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2716 if any retry times were supplied.
2718 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2719 connection, where there is no client IP address to check. In this
2720 situation, the verify now always succeeds.
2722 PH/25 Applied John Jetmore's -Mset patch.
2724 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2726 PH/27 In a string expansion for a processed (not raw) header when multiple
2727 headers of the same name were present, leading whitespace was being
2728 removed from all of them, but trailing whitespace was being removed only
2729 from the last one. Now trailing whitespace is removed from each header
2730 before concatenation. Completely empty headers in a concatenation (as
2731 before) are ignored.
2733 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2734 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2736 PH/29 [Removed. This was a change that I later backed out, and forgot to
2737 correct the ChangeLog entry (that I had efficiently created) before
2738 committing the later change.]
2740 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2741 address errors (4xx response to RCPT) over the same connection as other
2742 messages routed to the same hosts. Such deliveries are always "forced",
2743 so retry times are not inspected. This resulted in far too many retries
2744 for the affected addresses. The effect occurred only when there were more
2745 hosts than the hosts_max_try setting in the smtp transport when it had
2746 the 4xx errors. Those hosts that it had tried were not added to the list
2747 of hosts for which the message was waiting, so if all were tried, there
2748 was no problem. Two fixes have been applied:
2750 (i) If there are any address or message errors in an SMTP delivery, none
2751 of the hosts (tried or untried) are now added to the list of hosts
2752 for which the message is waiting, so the message should not be a
2753 candidate for sending over the same connection that was used for a
2754 successful delivery of some other message. This seems entirely
2755 reasonable: after all the message is NOT "waiting for some host".
2756 This is so "obvious" that I'm not sure why it wasn't done
2757 previously. Hope I haven't missed anything, but it can't do any
2758 harm, as the worst effect is to miss an optimization.
2760 (ii) If, despite (i), such a delivery is accidentally attempted, the
2761 routing retry time is respected, so at least it doesn't keep
2762 hammering the server.
2764 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2765 in ${readsocket because some servers need this prod.
2767 PH/32 Added some extra debug output when updating a wait-xxx database.
2769 PH/33 The hint "could be header name not terminated by colon", which has been
2770 given for certain expansion errors for a long time, was not being given
2771 for the ${if def:h_colon_omitted{... case.
2773 PH/34 The spec says: "With one important exception, whenever a domain list is
2774 being scanned, $domain contains the subject domain." There was at least
2775 one case where this was not true.
2777 PH/35 The error "getsockname() failed: connection reset by peer" was being
2778 written to the panic log as well as the main log, but it isn't really
2779 panic-worthy as it just means the connection died rather early on. I have
2780 removed the panic log writing for the ECONNRESET error when getsockname()
2783 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2784 runs only) independently of the message's sender address. This meant
2785 that, if the 4xx error was in fact related to the sender, a different
2786 message to the same recipient with a different sender could confuse
2787 things. In particular, this can happen when sending to a greylisting
2788 server, but other circumstances could also provoke similar problems.
2789 I have changed the default so that the retry time for these errors is now
2790 based a combination of the sender and recipient addresses. This change
2791 can be overridden by setting address_retry_include_sender=false in the
2794 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2795 remote server are returned as part of bounce messages. This was not
2796 happening for LMTP over a pipe (the lmtp transport), but now it is the
2797 same for both kinds of LMTP.
2799 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2800 in header lines that were in fact CNAMEs. This is no longer the case.
2802 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2803 and queue runs started by the daemon processed all messages. This has
2804 been fixed so that -R and -S can now usefully be given with -q<time>.
2806 PH/40 Import PCRE release 6.7 (fixes some bugs).
2808 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2810 PH/42 Give an error if -q is specified more than once.
2812 PH/43 Renamed the variables $interface_address and $interface_port as
2813 $received_ip_address and $received_port, to make it clear that these
2814 values apply to message reception, and not to the outgoing interface when
2815 a message is delivered. (The old names remain recognized, of course.)
2817 PH/44 There was no timeout on the connect() call when using a Unix domain
2818 socket in the ${readsocket expansion. There now is.
2820 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2821 be meaningful with "accept".
2823 SC/01 Eximstats V1.43
2824 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2826 SC/02 Eximstats V1.44
2827 Use a glob alias rather than an array ref in the generated
2828 parser. This improves both readability and performance.
2830 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2831 Collect SpamAssassin and rejection statistics.
2832 Don't display local sender or destination tables unless
2833 there is data to show.
2834 Added average volumes into the top table text output.
2836 SC/04 Eximstats V1.46
2837 Collect data on the number of addresses (recipients)
2838 as well as the number of messages.
2840 SC/05 Eximstats V1.47
2841 Added 'Message too big' to the list of mail rejection
2842 reasons (thanks to Marco Gaiarin).
2844 SC/06 Eximstats V1.48
2845 Mainlog lines which have GMT offsets and are too short to
2846 have a flag are now skipped.
2848 SC/07 Eximstats V1.49 (Alain Williams)
2849 Added the -emptyok flag.
2851 SC/08 Eximstats V1.50
2852 Fixes for obtaining the IP address from reject messages.
2854 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2855 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2856 whitespace changes from 4.64-PH/27
2858 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2861 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2862 are found, allow negative numbers in numeric criteria)
2864 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2866 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2867 to match changes made in 4.64-PH/43
2869 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2871 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2872 "domain" is actually an IP address, and force a failure. This locks out
2873 those revolvers/nameservers that support "A-for-A" lookups, in
2874 contravention of the specifications.
2876 PH/48 When a host name was looked up from an IP address, and the subsequent
2877 forward lookup of the name timed out, the host name was left in
2878 $sender_host_name, contrary to the specification.
2880 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2881 restricted to single-key lookups, Exim was not diagnosing an error if
2882 * or *@ was used with a query-style lookup.
2884 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2886 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2887 long ago, but noone interested enough thought of it. Let's just say that
2888 the "1.1" means that there are some new functions that weren't there at
2889 some point in the past.
2891 PH/51 Error processing for expansion failure of helo_data from an smtp
2892 transport during callout processing was broken.
2894 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2895 tested/used via the -bh/-bhc/-bs options.
2897 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2898 bug, fixed in subsequent PCRE releases).
2900 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2901 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2903 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2909 SC/01 Use a glob alias rather than an array ref in eximstats generated
2910 parser. This improves both readability and performance.
2912 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2913 Don't display local sender or destination tables in eximstats unless
2914 there is data to show.
2915 Added average volumes into the eximstats top table text output.
2917 SC/03 Collect data on the number of addresses (recipients) as well
2918 as the number of messages in eximstats.
2920 TF/01 Correct an error in the documentation for the redirect router. Exim
2921 does (usually) call initgroups() when daemonizing.
2923 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2924 with consistent privilege compared to when running as a daemon.
2926 TF/03 Note in the spec that $authenticated_id is not set for local
2927 submissions from trusted users.
2929 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2930 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2932 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2933 by adding some example configuration directives to the default
2934 configuration file. A little bit of work is required to uncomment the
2935 directives and define how usernames and passwords are checked, but
2936 there is now a framework to start from.
2938 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2939 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2940 without this. I don't know how relevant this is to other LDAP libraries.
2942 PH/02 Add the verb name to the "unknown ACL verb" error.
2944 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2946 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2948 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2949 directory not expanded when it should be if an expanded home directory
2950 was set for the address (which is overridden by the transport).
2952 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2955 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2956 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2957 because it is too late at that time, and has no effect.
2959 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2960 security issue with \' (bugzilla #107). I could not use the
2961 PQescapeStringConn() function, because it needs a PGconn value as one of
2964 PH/08 When testing addresses using -bt, indicate those final addresses that
2965 are duplicates that would not cause an additional delivery. At least one
2966 person was confused, thinking that -bt output corresponded to deliveries.
2967 (Suppressing duplicates isn't a good idea as you lose the information
2968 about possibly different redirections that led to the duplicates.)
2970 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2971 systems where poll() doesn't work, in particular OS X.
2973 PH/10 Added more information to debugging output for retry time not reached.
2975 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2976 operations in malware.c.
2978 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2981 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2982 syslog_duplication set false, log lines that would normally be written
2983 both the the main log and to the reject log were not written to syslog at
2986 PH/14 In the default configuration, change the use of "message" in ACL warn
2987 statements to "add_header".
2989 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2990 not followed by a command (e.g. "seen endif").
2992 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2993 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2996 PH/17 Added extra conditions to the default value of delay_warning_condition
3000 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3001 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3002 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3005 The Auto-Submitted: and various List- headers are standardised, whereas I
3006 don't think Precedence: ever was.
3008 PH/18 Refactored debugging code in route_finduser() to show more information,
3009 in particular, the error code if getpwnam() issues one.
3011 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3012 This is apparently needed in addition to the PH/07 change above to avoid
3013 any possible encoding problems.
3015 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3016 but not after initializing Perl.
3018 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3019 output them only if debugging. By default they are written stderr,
3020 apparently, which is not desirable.
3022 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3025 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3028 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3030 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3031 authenticated or an ident call has been made. Suppress the default
3032 values for $authenticated_id and $authenticated_sender (but permit -oMai
3033 and -oMas) when testing with -bh.
3035 PH/24 Re-jigged the order of the tests in the default configuration so that the
3036 tests for valid domains and recipients precede the DNS black list and CSA
3037 tests, on the grounds that those ones are more expensive.
3039 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3040 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3041 This bug exists in every version of Exim that I still have, right back to
3044 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3045 However, an attempt to turn on case-sensitivity in a regex key by
3046 including (?-i) didn't work because the subject string was already
3047 lowercased, and the effects were non-intuitive. It turns out that a
3048 one-line patch can be used to allow (?-i) to work as expected.
3054 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3055 other effects) broke the use of negated acl sub-conditions.
3057 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3060 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3061 "Deny" causes Exim to reject the incoming connection with a 554 error.
3062 Unfortunately, if there is a major crisis, such as a disk failure,
3063 tcp-wrappers gives "deny", whereas what one would like would be some
3064 kind of temporary error. A kludge has been added to help with this.
3065 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3066 554 error is used if errno is still zero or contains ENOENT (which occurs
3067 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3070 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3072 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3073 errors. Otherwise a message that provokes a temporary error (when other
3074 messages do not) can cause a whole host to time out.
3076 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3077 addresses were routed directly to files or pipes from a redirect router.
3078 File deliveries just didn't batch; pipe deliveries might have suffered
3081 PH/06 A failure to get a lock for a hints database would erroneously always say
3082 "Failed to get write lock", even when it was really a read lock.
3084 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3085 of 0600. This has been changed to use the value of the lockfile_mode
3086 option (which defaults to 0600).
3088 PH/08 Applied small patch from the Sieve maintainer.
3090 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3091 folder from quota calculations, a direct delivery into this folder messed
3092 up the contents of the maildirsize file. This was because the regex was
3093 used only to exclude .Trash (or whatever) when the size of the mailbox
3094 was calculated. There was no check that a delivery was happening into an
3095 excluded directory. This bug has been fixed by ignoring all quota
3096 processing for deliveries into excluded directories.
3098 PH/10 Added the maildirfolder_create_regex option to appendfile.
3104 PH/01 The code for finding all the local interface addresses on a FreeBSD
3105 system running IPv6 was broken. This may well have applied to all BSD
3106 systems, as well as to others that have similar system calls. The broken
3107 code found IPv4 interfaces correctly, but gave incorrect values for the
3108 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3109 that it would not match correctly against @[] and not recognize the IPv6
3112 PH/02 The ipliteral router was not recognizing addresses of the form user@
3113 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3115 PH/03 Added disable_ipv6.
3117 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3118 decoded form, because it is most often used to construct To: headers
3119 lines in autoreplies, and the decoded form may well be syntactically
3120 invalid. However, $reply_address has leading white space removed, and all
3121 newlines turned into spaces so that the autoreply transport does not
3124 PH/05 If group was specified without a user on a router, and no group or user
3125 was specified on a transport, the group from the router was ignored.
3127 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3128 for visible compile-time settings that can be used to change these
3129 numbers, for those that want even more. Backwards compatibility with old
3130 spool files has been maintained. However, going back to a previous Exim
3131 release will lost any variables that are in spool files.
3133 PH/07 Two small changes when running in the test harness: increase delay when
3134 passing a TCP/IP connection to a new process, in case the original
3135 process has to generate a bounce, and remove special handling of
3136 127.0.0.2 (sic), which is no longer necessary.
3138 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3139 be the same on different OS.
3141 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3144 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3145 whether --show-vars was specified or not
3147 JJ/02 exipick: Added support for new ACL variable spool format introduced
3150 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3151 syntactically invalid From: or Reply-to: line, and a filter used this to
3152 generate an autoreply, and therefore failed to obtain an address for the
3153 autoreply, Exim could try to deliver to a non-existent relative file
3154 name, causing unrelated and misleading errors. What now happens is that
3155 it logs this as a hard delivery error, but does not attempt to create a
3158 PH/11 The exinext utility has a -C option for testing purposes, but although
3159 the given file was scanned by exinext itself; it wasn't being passed on
3160 when Exim was called.
3162 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3163 an end-of-file indication when reading a command response.
3165 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3166 compiled. In many other places in Exim, IPv6 addresses are always
3167 recognized, so I have changed this. It also means that IPv4 domain
3168 literals of the form [IPV4:n.n.n.n] are now always recognized.
3170 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3171 used if the router is not running as root, for example, when verifying at
3172 ACL time, or when using -bh. The debugging output from this situation was
3173 non-existent - all you got was a failure to exec. I have made two
3176 (a) Failures to set uid/gid, the current directory, or a process leader
3177 in a subprocess such as that created by queryprogram now generate
3178 suitable debugging output when -d is set.
3180 (b) The queryprogram router detects when it is not running as root,
3181 outputs suitable debugging information if -d is set, and then runs
3182 the subprocess without attempting to change uid/gid.
3184 PH/15 Minor change to Makefile for building test_host (undocumented testing
3187 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3188 additional section of a DNS packet that returns MX or SRV records.
3189 Instead, it always explicitly searches for A/AAAA records. This avoids
3190 major problems that occur when a DNS server includes only records of one
3191 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3192 fixed another bug: if SRV records were looked up and the corresponding
3193 address records were *not* found in the additional section, the port
3194 values from the SRV records were lost.
3196 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3197 using the correct key (the original address) when searching the retry
3198 rules in order to find which one to use for generating the retry hint.
3200 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3201 adding the default one. Similarly, if it contains a Reply-To: header, the
3202 errors_reply_to option, if set, is not used.
3204 PH/19 When calculating a retry time, Exim used to measure the "time since
3205 failure" by looking at the "first failed" field in the retry record. Now
3206 it does not use this if it is later than than the arrival time of the
3207 message. Instead it uses the arrival time. This makes for better
3208 behaviour in cases where some deliveries succeed, thus re-setting the
3209 "first failed" field. An example is a quota failure for a huge message
3210 when small messages continue to be delivered. Without this change, the
3211 "time since failure" will always be short, possible causing more frequent
3212 delivery attempts for the huge message than are intended.
3213 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3215 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3216 $1, $2, $3) because the numerical variables can be reset during some
3217 expansion items (e.g. "match"), thereby losing the authentication data.
3219 PH/21 Make -bV show the size of off_t variables so that the test suite can
3220 decide whether to run tests for quotas > 2G.
3222 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3223 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3224 filecount value is greater than 2G or if a quota value is greater than 2G
3225 on a system where the size of off_t is not greater than 4, a panic error
3228 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3229 never match. The debug and -bh output now contains an explicit error
3230 message indicating a malformed IPv4 address or mask.
3232 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3233 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3234 PH/23 above applies.
3236 PH/25 Do not write to syslog when running in the test harness. The only
3237 occasion when this arises is a failure to open the main or panic logs
3238 (for which there is an explicit test).
3240 PH/26 Added the /no_tell option to "control=freeze".
3242 PH/27 If a host name lookup failed very early in a connection, for example, if
3243 the IP address matched host_lookup and the reverse lookup yielded a name
3244 that did not have a forward lookup, an error message of the form "no IP
3245 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3246 could be logged. Now it outputs the IP address instead of "NULL".
3248 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3249 allows the sender and the authenticated sender to be set when
3250 submitting a message from within Exim. Since child_open_exim() is
3251 documented for local_scan(), the new function should be too.
3253 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3254 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3255 results in an empty string is now treated as unset.
3257 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3259 PH/31 Added sender_verify_fail logging option.
3261 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3262 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3263 tidied the source and removed it altogether.
3265 PH/33 When a queue run was abandoned because the load average was too high, a
3266 log line was always written; now it is written only if the queue_run log
3267 selector is set. In addition, the log line for abandonment now contains
3268 information about the queue run such as the pid. This is always present
3269 in "start" and "stop" lines but was omitted from the "abandon" line.
3271 PH/34 Omit spaces between a header name and the colon in the error message that
3272 is given when verify = headers_syntax fails (if there are lots of them,
3273 the message gets confusing).
3275 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3276 names, as there are now some PTR records that contain slashes. This check
3277 is only to protect against broken name servers that fall over on strange
3278 characters, so the fact that it applies to all lookups doesn't matter.
3280 PH/36 Now that the new test suite is complete, we can remove some of the
3281 special code in Exim that was needed for the old test suite. For example,
3282 sorting DNS records because real resolvers return them in an arbitrary
3283 order. The new test suite's fake resolver always returns records in the
3286 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3287 bounces) except when queue_only is set, to avoid logging races between
3288 the different processes.
3290 PH/38 Panic-die if .include specifies a non-absolute path.
3292 PH/39 A tweak to the "H" retry rule from its user.
3294 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3295 a label. They prevented compilation on older perls.
3297 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3298 a warning to be raised on newish perls.
3300 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3301 on queue. Changes to match documented behaviour of showing count of
3302 messages matching specified criteria.
3304 PH/40 Changed the default ident timeout from 30s to 5s.
3306 PH/41 Added support for the use of login_cap features, on those BSD systems
3307 that have them, for controlling the resources used by pipe deliveries.
3309 PH/42 The content-scanning code uses fopen() to create files in which to put
3310 message data. Previously it was not paying any attention to the mode of
3311 the files. Exim runs with umask(0) because the rest of the code creates
3312 files with open(), and sets the required mode explicitly. Thus, these
3313 files were ending up world-writeable. This was not a big issue, because,
3314 being within the spool directory, they were not world-accessible. I have
3315 created a function called modefopen, which takes an additional mode
3316 argument. It sets umask(777), creates the file, chmods it to the required
3317 mode, then resets the umask. All the relevant calls to fopen() in the
3318 content scanning code have been changed to use this function.
3320 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3321 to 24 hours. This avoids potential overflow problems when processing G
3322 and H retry rules. I suspect nobody ever tinkers with this value.
3324 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3326 PH/45 When the plaintext authenticator is running as a client, the server's
3327 challenges are checked to ensure they are valid base64 strings. By
3328 default, the authentication attempt is cancelled if an invalid string is
3329 received. Setting client_ignore_invalid_base64 true ignores these errors.
3330 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3331 they are received. Thus, the responses can be made to depend on the
3332 challenges. If an invalid string is ignored, an empty string is placed in
3335 PH/46 Messages that are created by the autoreply transport now contains a
3336 References: header, in accordance with RFCs 2822 and 3834.
3338 PH/47 Added authenticated_sender_force to the smtp transport.
3340 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3342 PH/49 Installed latest patch from the Sieve maintainer.
3344 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3345 was also set, the appendfile transport was unnecessarily scanning a
3346 directory of message files (e.g. for maildir delivery) to find the count
3347 of files (along with the size), even though it did not need this
3348 information. It now does the scan only if it needs to find either the
3349 size of the count of files.
3351 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3353 PH/52 Two bugs concerned with error handling when the smtp transport is
3356 (i) Exim was not creating retry information for temporary errors given
3357 for individual recipients after the DATA command when the smtp transport
3358 was used in LMTP mode. This meant that they could be retried too
3359 frequently, and not timed out correctly.
3361 (ii) Exim was setting the flag that allows error details to be returned
3362 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3363 recipients that were returned after the DATA command.
3365 PH/53 This is related to PH/52, but is more general: for any failing address,
3366 when detailed error information was permitted to be returned to the
3367 sender, but the error was temporary, then after the final timeout, only
3368 "retry timeout exceeded" was returned. Now it returns the full error as
3369 well as "retry timeout exceeded".
3371 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3372 do this, and (what is worse) MTAs that accept it.
3374 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3375 will now be deprecated.
3377 PH/56 New os.c-cygwin from the Cygwin maintainer.
3379 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3380 formats (previously only available in exim formats via -bpr, -bpru,
3381 and -bpra. Now also available in native and exiqgrep formats)
3383 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3384 with very large, slow to parse queues
3386 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3388 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3390 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3391 responses to authentication challenges, though it was showing the
3392 challenges; (ii) I've removed the CR characters from the debug output for
3395 PH/58 Allow for the insertion of a newline as well as a space when a string
3396 is turned into more than one encoded-word during RFC 2047 encoding. The
3397 Sieve code now uses this.
3399 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3400 data_4xx, lost_connection, tls_required.
3402 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3403 message was being sent as an SMTP response.
3405 PH/61 Add -l and -k options to exicyclog.
3407 PH/62 When verifying, if an address was redirected to one new address, so that
3408 verification continued, and the new address failed or deferred after
3409 having set something in $address_data, the value of $address_data was not
3410 passed back to the ACL. This was different to the case when no
3411 redirection occurred. The value is now passed back in both cases.
3413 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3414 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3415 use login_cap.h, so on its own it isn't the distinguishing feature. The
3416 new name refers directly to the setclassresources() function.
3418 PH/65 Added configuration files for NetBSD3.
3420 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3422 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3423 is preferred over IPv4.
3425 PH/68 The bounce_return_message and bounce_return_body options were not being
3426 honoured for bounces generated during the reception of non-SMTP messages.
3427 In particular, this applied to messages rejected by the ACL. This bug has
3428 been fixed. However, if bounce_return_message is true and bounce_return_
3429 body is false, the headers that are returned for a non-SMTP message
3430 include only those that have been read before the error was detected.
3431 (In the case of an ACL rejection, they have all been read.)
3433 PH/69 The HTML version of the specification is now built in a directory called
3434 spec_html instead of spec.html, because the latter looks like a path with
3435 a MIME-type, and this confuses some software.
3437 PH/70 Catch two compiler warnings in sieve.c.
3439 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3440 function verify_get_ident() calls ip_connect() to connect a socket, but
3441 if the "connect()" function timed out, ip_connect() used to close the
3442 socket. However, verify_get_ident() also closes the socket later, and in
3443 between Exim writes to the log, which may get opened at this point. When
3444 the socket was closed in ip_connect(), the log could get the same file
3445 descriptor number as the socket. This naturally causes chaos. The fix is
3446 not to close the socket in ip_connect(); the socket should be closed by
3447 the function that creates it. There was only one place in the code where
3448 this was missing, in the iplookup router, which I don't think anybody now
3449 uses, but I've fixed it anyway.
3451 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3452 well as to direct DNS lookups. Otherwise the handling of names in host
3453 lists is inconsistent and therefore confusing.
3459 PH/01 Two changes to the default runtime configuration:
3461 (1) Move the checks for relay_from_hosts and authenticated clients from
3462 after to before the (commented out) DNS black list checks.
3464 (2) Add control=submission to the relay_from_hosts and authenticated
3465 clients checks, on the grounds that messages accepted by these
3466 statements are most likely to be submissions.
3468 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3470 (1) Generate an error if the third argument for the ${prvs expansion is
3473 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3476 (3) Reset the variables that are obtained from the first argument of
3477 ${prvscheck and used in the second argument before leaving the code,
3478 because their memory is reclaimed, so using them afterwards may do
3481 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3482 one (it's much easier than Tom thought :-).
3484 (5) Because of (4), we can now allow for the use of $prvscheck_result
3485 inside the third argument.
3487 PH/03 For some reason, the default setting of PATH when running a command from
3488 a pipe transport was just "/usr/bin". I have changed it to
3491 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3492 anything to be listed in the output from -bV.
3494 PH/05 When a filter generated an autoreply, the entire To: header line was
3495 quoted in the delivery log line, like this:
3497 => >A.N.Other <ano@some.domain> <original@ddress> ...
3499 This has been changed so that it extracts the operative address. There
3500 may be more than one such address. If so, they are comma-separated, like
3503 => >ano@some.domain,ona@other.domain <original@ddress> ...
3505 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3506 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3507 not being looked up in the rDNS to get a host name, Exim was showing the
3508 IP address twice in Received: lines, even though the IP addresses were
3509 identical. For example:
3511 Received: from [1.2.3.4] (helo=[1.2.3.4])
3513 However, if the real host name was known, it was omitting the HELO data
3514 if it matched the actual IP address. This has been tidied up so that it
3515 doesn't show the same IP address twice.
3517 PH/07 When both +timestamp and +memory debugging was on, the value given by
3518 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3519 called by the debug printing, thereby overwriting the timestamp buffer.
3520 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3522 PH/08 When the original message was included in an autoreply transport, it
3523 always said "this is a copy of the message, including all the headers",
3524 even if body_only or headers_only was set. It now gives an appropriate
3527 PH/09 Applied a patch from the Sieve maintainer which:
3529 o fixes some comments
3530 o adds the (disabled) notify extension core
3531 o adds some debug output for the result of if/elsif tests
3532 o points to the current vacation draft in the documentation
3533 and documents the missing references header update
3537 o fixes a bug in processing the envelope test (when testing
3538 multiple envelope elements, the last element determined the
3541 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3542 Electronic Mail") by including:
3544 Auto-submitted: auto-generated
3546 in the messages that it generates (bounce messages and others, such as
3547 warnings). In the case of bounce messages for non-SMTP messages, there was
3548 also a typo: it was using "Auto_submitted" (underscore instead of
3549 hyphen). Since every message generated by Exim is necessarily in response
3550 to another message, thes have all been changed to:
3552 Auto-Submitted: auto-replied
3554 in accordance with these statements in the RFC:
3556 The auto-replied keyword:
3558 - SHOULD be used on messages sent in direct response to another
3559 message by an automatic process,
3561 - MUST NOT be used on manually-generated messages,
3563 - MAY be used on Delivery Status Notifications (DSNs) and Message
3564 Disposition Notifications (MDNs),
3566 - MUST NOT be used on messages generated by automatic or periodic
3567 processes, except for messages which are automatic responses to
3570 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3571 to the default Received: header definition.
3573 PH/12 Added log selector acl_warn_skipped (default on).
3575 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3576 variables because (a) they are in the wrong storage pool and (b) even if
3577 they were copied, it wouldn't work properly because of the caching.
3579 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3580 checking when decoding. Apparently there are clients that generate
3581 overlong encoded strings. Why am I not surprised?
3583 PH/15 If the first argument of "${if match_address" was not empty, but did not
3584 contain an "@" character, Exim crashed. Now it writes a panic log message
3585 and treats the condition as false.
3587 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3589 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3590 extension "envelope-auth". The code is finished and in agreement with
3591 other implementations, but there is no documentation so far and in fact,
3592 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3593 not changing the active code.
3595 Print executed "if" and "elsif" statements when debugging is used. This
3596 helps a great deal to understand what a filter does.
3598 Document more things not specified clearly in RFC3028. I had all this
3599 sorted out, when out of a sudden new issues came to my mind. Oops."
3601 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3604 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3605 canonical form (as documented). However, after a host name lookup from
3606 the IP address, check_host() was doing a simple string comparison with
3607 addresses acquired from the DNS when checking that the found name did
3608 have the original IP as one of its addresses. Since any found IPv6
3609 addresses are likely to be in abbreviated form, the comparison could
3610 fail. Luckily, there already exists a function for doing the comparison
3611 by converting both addresses to binary, so now that is used instead of
3612 the text comparison.
3614 PH/20 There was another similar case to PH/19, when a complete host name was
3615 given in a host list; looking up its IP address could give an abbreviated
3616 form, whereas the current host's name might or might not be abbreviated.
3617 The same fix has been applied.
3623 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3624 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3627 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3628 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3630 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3632 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3633 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3634 Also, for "Auto-Submitted", treat anything other than "no" as
3635 non-personal, in accordance with RFC 3834. (Previously it treated
3636 anything starting "auto-" as non-personal.)
3638 TF/01 The control=submission/name=... option had a problem with syntax
3639 errors if the name included a slash character. The /name= option
3640 now slurps the rest of the string, so it can include any characters
3641 but it must come last in the list of options (after /sender_retain
3644 PH/05 Some modifications to the interface to the fake nameserver for the new
3652 TK/01 Added the "success_on_redirect" address verification option. See
3653 NewStuff for rationale and an example.
3655 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3657 PH/02 Patch to exigrep to allow it to work on syslog lines.
3659 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3660 fread() to skip over the body file's header line, because in Cygwin the
3661 header line is locked and is inaccessible.
3663 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3664 co-exist for some time) to make it clear that it is the Exim ID that is
3665 referenced, not the Message-ID: header line.
3667 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3668 string_format() function, because snprintf() does not exist on all
3671 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3672 ${stat: expansion item.
3674 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3675 protocol synchronization error", to keep the pedants happy.
3677 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3678 well as for IRIX systems, when gcc is being used. See the host.c source
3681 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3683 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3686 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3687 transport and to the smtp transport in LMTP mode.
3689 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3691 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3692 run a filter in a subprocess. This could lead to confusion in subsequent
3693 lookups in the parent process. There should also be a search_tidyup() at
3694 the end of the subprocess.
3696 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3697 only if the host matched helo_try_verify_hosts, which caused the
3698 verification to occur when the EHLO/HELO command was issued. The ACL just
3699 tested the remembered result. Now, if a previous verification attempt has
3700 not happened, "verify = helo" does it there and then.
3702 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3704 TK/03 Fix log output including CR from clamd.
3706 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3707 exist provoked a memory error which could cause a segfault.
3709 PH/15 Installed PCRE 6.2
3711 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3713 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3714 of the problem. Specifically, suggested +O2 rather than +O1 for the
3717 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3719 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3720 record, the port was not correctly logged when the outgoing_port log
3721 selector was set (it logged the transort's default port).
3723 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3724 fallback_hosts, and "hosts" in the smtp transport.
3726 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3727 host errors such as "Connection refused".
3729 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3730 authentication with radiusclient 0.4.9:
3732 - Error returned from rc_read_config was caught wrongly
3733 - Username/password not passed on to radius server due to wrong length.
3735 The presumption is that some radiusclient API changes for 4.51/PH/17
3736 were not taken care of correctly. The code is still untested by me (my
3737 Linux distribution still has 0.3.2 of radiusclient), but it was
3738 contributed by a Radius user.
3740 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3741 expanding the "port" option of the smtp transport.
3743 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3744 while reading a MIME header. Thanks to Tom Hughes for a patch.
3746 PH/24 Include config.h inside local_scan.h so that configuration settings are
3749 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3750 that in an ACL for RCPT (for example), you can examine exactly what was
3753 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3754 commands, but it was not correctly comparing the address with the actual
3755 client host address. Thus, it would show the EHLO address in Received:
3756 header lines when this was not necessary.
3758 PH/27 Added the % operator to ${eval:}.
3760 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3761 it should be ignoring failures (because with -C, for example, it has lost
3762 privilege). It wasn't ignoring creation failures other than "already
3765 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3768 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3769 because an input line was too long, either on its own, or by virtue of
3770 too many continuations, the temporary file was not being removed, and the
3771 return code was incorrect.
3773 PH/31 Missing "BOOL" in function definition in filtertest.c.
3775 PH/32 Applied Sieve patches from the maintainer.
3777 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3779 PH/33 Added "verify = not_blind".
3781 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3782 Local/Makefile (with some defaults set). These are used in built scripts
3783 such as exicyclog, but they have never been used in the exim_install
3784 script (though there are many overriding facilities there). I have
3785 arranged that the exim_install script now takes note of these two
3788 PH/35 Installed configuration files for Dragonfly.
3790 PH/36 When a locally submitted message by a trusted user did not contain a
3791 From: header, and the sender address was obtained from -f or from an SMTP
3792 MAIL command, and the trusted user did not use -F to supply a sender
3793 name, $originator_name was incorrectly used when constructing a From:
3794 header. Furthermore, $originator_name was used for submission mode
3795 messages from external hosts without From: headers in a similar way,
3796 which is clearly wrong.
3798 PH/37 Added control=suppress_local_fixups.
3800 PH/38 When log_selector = +received_sender was set, and the addition of the
3801 sender made the log line's construction buffer exactly full, or one byte
3802 less than full, an overflow happened when the terminating "\n" was
3805 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3806 when the result of a list match is failure because a DNS lookup failed.
3808 PH/40 RM_COMMAND is now used in the building process.
3810 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3811 the "build-* directories that it finds.
3813 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3814 address in a domain literal was a prefix of an interface address.
3816 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3817 when verifying a sender address, unless rewrite_headers is false.
3819 PH/44 Wrote a long comment about why errors_to addresses are verified as
3820 recipients, not senders.
3822 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3823 the ratelimit ACL was added.
3825 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3827 PH/46 Added extra information about PostgreSQL errors to the error string.
3829 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3830 avoiding the need to install special zones in a real server. This is
3831 backwards compatible; if it can't find the fake resolver, it drops back.
3832 Thus, both old and new test suites can be run.
3834 TF/02 Added util/ratelimit.pl
3836 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3837 clock is set back in time.
3839 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3840 Candler <B.Candler@pobox.com>.
3842 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3843 OK for addresses that are the result of redirections.
3845 PH/48 A number of further additions for the benefit of the new test suite,
3846 including a fake gethostbyname() that interfaces to the fake DNS resolver
3849 TF/06 The fix for widen_domains has also been applied to qualify_single and
3850 search_parents which are the other dnslookup options that can cause
3853 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3856 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3858 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3859 service" if it encountered an MX record with an empty target hostname.
3860 The message is now "an MX or SRV record indicated no SMTP service".
3862 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3863 if the DNS of the sending site is misconfigured. This is quite a
3864 common situation. This change restores the behaviour of treating a
3865 helo verification defer as a failure.
3867 PH/51 If self=fail was set on a router, the bounce message did not include the
3868 actual error message.
3874 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3876 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3877 command itself ended in error, the underlying message about the transport
3878 filter timeout was being overwritten with the pipe command error. Now the
3879 underlying error message should be appended to the second error message.
3881 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3883 PH/02 Reduce the amount of output that "make" produces by default. Full output
3884 can still be requested.
3886 PH/03 The warning log line about a condition test deferring for a "warn" verb
3887 was being output only once per connection, rather than after each
3888 occurrence (because it was using the same function as for successful
3889 "warn" verbs). This seems wrong, so I have changed it.
3891 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3892 it should not have, which might have caused a crash in the right
3893 circumstances, but probably never did.
3895 PH/04 Installed a modified version of Tony Finch's patch to make submission
3896 mode fix the return path as well as the Sender: header line, and to
3897 add a /name= option so that you can make the user's friendly name appear
3900 TF/03 Added the control = fakedefer ACL modifier.
3902 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3903 Mark Lowes for thorough testing.
3905 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3907 TK/03 Merged latest SRS patch from Miles Wilton.
3909 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3910 with the definition in sysexits.h (which is #included earlier).
3911 Fortunately, Exim does not actually use EX_OK. The code used to try to
3912 preserve the sysexits.h value, by assuming that macro definitions were
3913 scanned for macro replacements. I have been disabused of this notion,
3914 so now the code just undefines EX_OK before #including unistd.h.
3916 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3917 in the smtp transport. When a block could not be written in a single
3918 write() function, the timeout was being re-applied to each part-write.
3919 This seems wrong - if the receiver was accepting one byte at a time it
3920 would take for ever. The timeout is now adjusted when this happens. It
3921 doesn't have to be particularly precise.
3923 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3924 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3926 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3927 without a domain if the check to <postmaster@domain> fails.
3929 SC/01 Eximstats: added -xls and the ability to specify output files
3930 (patch written by Frank Heydlauf).
3932 SC/02 Eximstats: use FileHandles for outputting results.
3934 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3936 SC/04 Eximstats: fixed display of large numbers with -nvr option
3938 SC/05 Eximstats: fixed merging of reports with empty tables.
3940 SC/06 Eximstats: added the -include_original_destination flag
3942 SC/07 Eximstats: removed tabs and trailing whitespace.
3944 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3946 TK/06 MBOX spool code: Add real "From " MBOX separator line
3947 so the .eml file is really in mbox format (even though
3948 most programs do not really care). Patch from Alex Miller.
3950 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3951 The latter is generated from $received_to and is only set if the
3952 message has one envelope recipient. SA can use these headers,
3953 obviously out-of-the-box. Patch from Alex Miller.
3955 PH/08 The ${def test on a variable was returning false if the variable's
3956 value was "0", contrary to what the specification has always said!
3957 The result should be true unless the variable is empty.
3959 PH/09 The syntax error of a character other than { following "${if
3960 def:variable_name" (after optional whitespace) was not being diagnosed.
3961 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3962 accidental colon was present, for example, could give incorrect results.
3964 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3965 result is used (not including appendfile, where other changes are about
3968 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3969 This involved changing a lot of size variables from int to off_t. It
3970 should work with maildirs and everything.
3972 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3973 spamd dying while we are connected to it.
3975 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3978 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3979 function for BDB 4.3.
3981 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3983 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3984 the sources for PCRE have been re-arranged and more files are now
3987 PH/15 The code I had for printing potentially long long variables in PH/11
3988 above was not the best (it lost precision). The length of off_t variables
3989 is now inspected at build time, and an appropriate printing format (%ld
3990 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3991 to be "long long int" or "long int". This is needed for the internal
3992 formatting function string_vformat().
3994 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3995 the configuration file to be ":syslog", then the script "guesses" where
3996 the logs files are, rather than using the compiled in default. In our
3997 case the guess is not the same as the compiled default, so the script
3998 suddenly stopped working when I started to use syslog. The patch checks
3999 to see if log_file_path is "". If so, it attempts to read it from exim
4000 with no configuration file to get the compiled in version, before it
4001 falls back to the previous guessing code."
4003 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4004 implementing BATV in an Exim configuration. See NewStuff for the gory
4007 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4008 Makefile that are specific to HP-UX.
4010 PH/18 If the "use_postmaster" option was set for a recipient callout together
4011 with the "random" option, the postmaster address was used as the MAIL
4012 FROM address for the random test, but not for the subsequent recipient
4013 test. It is now used for both.
4015 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4016 patch removes a few documentation additions to RFC 3028, because the
4017 latest draft now contains them. It adds the new en;ascii-case comparator
4018 and a new error check for 8bit text in MIME parts. Comparator and
4019 require names are now matched exactly. I enabled the subaddress
4020 extension, but it is not well tested yet (read: it works for me)."
4022 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4023 rework some of the code of TK/09 above to avoid the hardwired use of
4024 "%lld" and "long long". Replaced the call to snprintf() with a call to
4027 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4028 records point to non-existent hosts", "retry timeout exceeded", and
4029 "retry time not reached for any host after a long failure period".
4031 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4032 experimental DomainKeys support:
4034 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4035 (2) On an error such as an illegally used "control", the wrong name for
4036 the control was given.
4038 These problems did NOT occur unless DomainKeys support was compiled.
4040 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4042 PH/24 Added ${if match_ip condition.
4044 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4045 databases so that it will be absolutely obvious if a crash occurs in the
4046 DB library. This is a regular occurrence (often caused by mis-matched
4049 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4050 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4051 picked up on a user's system that detects such things. There doesn't seem
4052 to be a gcc warning option for this - only an attribute that has to be
4053 put on the function's prototype. It seems that in Fedora Core 4 they have
4054 set this on a number of new functions. No doubt there will be more in due
4057 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4058 specify a transport. However, if an address that was verified by such a
4059 router was the subject of a callout, Exim crashed because it tried to
4060 read the rcpt_include_affixes from the non-existent transport. Now it
4061 just assumes that the setting of that option is false. This bug was
4062 introduced by 4.51/PH/31.
4064 PH/28 Changed -d+all to exclude +memory, because that information is very
4065 rarely of interest, but it makes the output a lot bigger. People tend to
4066 do -d+all out of habit.
4068 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4069 code in os-type was giving problems when libc.so lives in lib64, like on
4072 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4073 aren't the modern standard, and it seems that some systems' include files
4074 don't always have them. Exim was already checking for some of the newer
4075 ones like T_AAAA, and defining it itself. I've added checks for all the
4076 record types that Exim uses.
4078 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4079 not automatically generating a new one, as it is supposed to. This
4080 prevented TLS from working. If the file did exist, but contained invalid
4081 data, a new version was generated, as expected. It was only the case of a
4082 non-existent file that was broken.
4084 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4085 with a change in libdomainkeys > 0.64.
4087 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4088 from DNS. If the selector record carries the flag, it now has
4089 precedence over the domain-wide flag.
4091 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4093 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4094 the use of a port name that isn't defined in /etc/services) occurred, the
4095 message was deferred as in a normal delivery, and thus remained on the
4096 spool, instead of being failed because of the mua_wrapper setting. This
4097 is now fixed, and I tidied up some of the mua_wrapper messages at the
4100 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4101 the messages in a hash of arrays rather than using individual hashes.
4102 This is a bit cleaner and results in dramatic memory savings, albeit
4103 at a slight CPU cost.
4105 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4106 as requested by Marc Sherman.
4108 SC/10 Eximstats: added histograms for user specified patterns as requested
4111 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4113 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4114 fopen() in the content-scanning modules that did not already have it.
4120 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4121 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4123 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4125 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4127 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4128 to test Sieve filters that use "vacation".
4130 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4131 that changes the way the GnuTLS parameters are stored in the cache file.
4132 The new format can be generated externally. For backward compatibility,
4133 if the data in the cache doesn't make sense, Exim assumes it has read an
4134 old-format file, and it generates new data and writes a new file. This
4135 means that you can't go back to an older release without removing the
4138 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4139 work if there are any delivery delays because "one_time" forces the
4140 parent to be marked "delivered", so its unseen clone is never tried
4141 again. For this reason, Exim now forbids the simultaneous setting of
4144 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4145 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4146 entry if you want to know the details. The fix, however, neglected to
4147 consider the case when local delivery batching is involved. The test for
4148 "previously delivered" was not happening when checking to see if an
4149 address could be batched with a previous (undelivered) one; under
4150 certain circumstances this could lead to multiple deliveries to the same
4153 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4154 in its include files, and this causes problems building Exim.
4156 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4157 header_syntax) but Exim was just ignoring anything given after a slash.
4158 In particular, this caused confusion with an attempt to use "verify =
4159 reverse_host_lookup/defer_ok". An error is now given when options are
4160 supplied for verify items that do not have them. (Maybe reverse_host_
4161 lookup should have a defer_ok option, but that's a different point.)
4163 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4164 defined by RFC 821) to 2048, because there were problems with some AUTH
4165 commands, and RFC 1869 says the size should be increased for extended
4166 SMTP commands that take arguments.
4168 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4171 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4172 "unknown" error; now it says that the functionality isn't in the binary.
4174 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4175 an address' error message when a string expansion fails (syntax or
4176 whatever). Otherwise the password may appear in the log. Following change
4177 PH/42 below, there is no longer a chance of it appearing in a bounce
4180 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4182 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4183 of its addresses was ever tried. (Bugzilla bug #2.)
4185 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4186 the result incorrectly in the debug output. (It correctly added a newline
4187 to what was transported.)
4189 TF/01 Added $received_time.
4191 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4192 commented out examples of how to interface to a virus scanner and to
4193 SpamAssassin. Also added commented examples of av_scanner and
4194 spamd_address settings.
4196 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4197 and controls are allowed in which ACLs. There were a couple of minor
4198 errors. Some of the entries in the conditions table (which is a table of
4199 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4200 negation of where the condition IS allowed.
4202 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4204 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4205 header file does not have a version number, so I've had to invent a new
4206 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4207 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4208 radiusclient), but it was contributed by a Radius user.
4210 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4211 files or directories, for OpenSSL.
4213 PH/19 When an Exim process that is running as root has to create an Exim log
4214 file, it does so in a subprocess that runs as exim:exim so as to get the
4215 ownership right at creation (otherwise, other Exim processes might see
4216 the file with the wrong ownership). There was no test for failure of this
4217 fork() call, which would lead to the process getting stuck as it waited
4218 for a non-existent subprocess. Forks do occasionally fail when resources
4219 run out. I reviewed all the other calls to fork(); they all seem to check
4222 PH/20 When checking for unexpected SMTP input at connect time (before writing
4223 the banner), Exim was not dealing correctly with a non-positive return
4224 from the read() function. If the client had disconnected by this time,
4225 the result was a log entry for a synchronization error with an empty
4226 string after "input=" when read() returned zero. If read() returned -1
4227 (an event I could not check), uninitialized data bytes were printed.
4228 There were reports of junk text (parts of files, etc) appearing after
4231 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4233 PH/22 Added support for macro redefinition, and (re)definition in between
4234 driver and ACL definitions.
4236 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4237 forgetting to use the resulting value; it was using the unexpanded value.
4239 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4240 hadn't been configured. The fix is from Juergen Kreileder, who
4241 understands it better than I do:
4243 "Here's what I see happening with three configured cyrus_sasl
4244 authenticators configured (plain, login, cram-md5):
4246 On startup auth_cyrus_sasl_init() gets called for each of these.
4247 This means three calls to sasl_listmech() without a specified mech_list.
4248 => SASL tests which mechs of all available mechs actually work
4249 => three warnings about OTP not working
4250 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4252 With the patch, sasl_listmech() also gets called three times. But now
4253 SASL's mech_list option is set to the server_mech specified in the the
4254 authenticator. Or in other words, the answer from sasl_listmech()
4255 gets limited to just the mech you're testing for (which is different
4257 => the return list contains just 'plain' or 'login', 'cram-md5' or
4258 nothing depending on the value of ob->server_mech.
4260 I've just tested the patch: Authentication still works fine,
4261 unavailable mechs specified in the exim configuration are still
4262 caught, and the auth.log warnings about OTP are gone."
4264 PH/25 When debugging is enabled, the contents of the command line are added
4265 to the debugging output, even when log_selector=+arguments is not
4268 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4269 answer is "GNU", and only if the return is "GNU/something" is the answer
4272 PH/27 $acl_verify_message is now set immediately after the failure of a
4273 verification in an ACL, and so is available in subsequent modifiers. In
4274 particular, the message can be preserved by coding like this:
4276 warn !verify = sender
4277 set acl_m0 = $acl_verify_message
4279 Previously, $acl_verify_message was set only while expanding "message"
4280 and "log_message" when a very denied access.
4282 PH/28 Modified OS/os.c-Linux with
4284 -#ifndef OS_LOAD_AVERAGE
4285 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4287 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4288 nomenclature these days.)
4290 PH/29 Installed patch from the Sieve maintainer that adds the options
4291 sieve_useraddress and sieve_subaddress to the redirect router.
4293 PH/30 In these circumstances:
4294 . Two addresses routed to the same list of hosts;
4295 . First host does not offer TLS;
4296 . First host accepts first address;
4297 . First host gives temporary error to second address;
4298 . Second host offers TLS and a TLS session is established;
4299 . Second host accepts second address.
4300 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4301 and peerdn, if requested) that were in fact used only for the second
4304 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4305 attention to any local part prefix or suffix that was matched by the
4306 router that accepted the address. It now behaves in the same way as it
4307 does for delivery: the affixes are removed from the local part unless
4308 rcpt_include_affixes is set on the transport.
4310 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4311 timeout during the DATA phase of an incoming message.
4313 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4314 applied a patch sanctioned by the Sieve maintainer.
4316 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4317 the uid or gid is negative. A case of a negative gid caused this to be
4318 noticed. The fix allows for either to be negative.
4320 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4321 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4324 PH/36 The change PH/12 above was broken. Fixed it.
4326 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4327 the grounds that routing the same address twice would always produce the
4328 same answer. This might have been true once, but it is certainly no
4329 longer true now. Routing a child address may depend on the previous
4330 routing that produced that child. Some complicated redirection strategies
4331 went wrong when messages had multiple recipients, and made Exim's
4332 behaviour dependent on the order in which the addresses were given.
4334 I have moved the duplicate checking until after the routing is complete.
4335 Exim scans the addresses that are assigned to local and remote
4336 transports, and removes any duplicates. This means that more work will be
4337 done, as duplicates will always all be routed, but duplicates are
4338 presumably rare, so I don't expect this is of any significance.
4340 For deliveries to pipes, files, and autoreplies, the duplicate checking
4341 still happens during the routing process, since they are not going to be
4344 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4345 It corrects a timeout issue with spamd. This is Ian's comment: "The
4346 background is that sometimes spamd either never reads data from a
4347 connection it has accepted, or it never writes response data. The exiscan
4348 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4349 blindly assumes that writes won't block so it may never time out."
4351 PH/39 Allow G after quota size as well as K and M.
4353 PH/40 The value set for $authenticated_id in an authenticator may not contain
4354 binary zeroes or newlines because the value is written to log lines and
4355 to spool files. There was no check on this. Now the value is run through
4356 the string_printing() function so that such characters are converted to
4357 printable escape sequences.
4359 PH/41 $message_linecount is a new variable that contains the total number of
4360 lines in the message. Compare $body_linecount, which is the count for the
4363 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4364 bounce and delay warning messages, except in certain special cases, which
4367 (a) An SMTP error message from a remote host;
4368 (b) A message specified in a :fail: redirection;
4369 (c) A message specified in a "fail" command in a system filter;
4370 (d) A message specified in a FAIL return from the queryprogram router;
4371 (e) A message specified by the cannot_route_message router option.
4373 In these cases only, Exim does include the error details in bounce and
4374 warning messages. There are also a few cases where bland messages such
4375 as "unrouteable address" or "local delivery error" are given.
4377 PH/43 $value is now also set for the "else" part of a ${run expansion.
4379 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4380 being worked on, but at least Exim now implements the latest version to
4383 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4384 process to complete was treated as a delivery failure, a timeout while
4385 writing the message to the pipe was logged, but erroneously treated as a
4386 successful delivery. Such timeouts include transport filter timeouts. For
4387 consistency with the overall process timeout, these timeouts are now
4388 treated as errors, giving rise to delivery failures by default. However,
4389 there is now a new Boolean option for the pipe transport called
4390 timeout_defer, which, if set TRUE, converts the failures into defers for
4391 both kinds of timeout. A transport filter timeout is now identified in
4394 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4395 systems where "make" and "gmake" are different, calling "gmake" at top
4396 level broke things. I've arranged for the value of $(MAKE) to be passed
4397 from the Makefile to this script so that it can call the same version of
4401 A note about Exim versions 4.44 and 4.50
4402 ----------------------------------------
4404 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4405 changes of various kinds. As a consequence, a big documentation update was
4406 needed. This delayed the release for rather longer than seemed good, especially
4407 in the light of a couple of (minor) security issues. Therefore, the changes
4408 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4409 release. So 4.44 and 4.50 are in effect two different branches that both start
4412 I have left the 4.50 change log unchanged; it contains all the changes since
4413 4.43. The change log for 4.44 is below; many of its items are identical to
4414 those for 4.50. This seems to be the most sensible way to preserve the
4415 historical information.
4421 1. Minor wording change to the doc/README.SIEVE file.
4423 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4424 computation of the current number of files was incorrect.
4426 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4427 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4430 4. Give more explanation in the error message when the command for a transport
4431 filter fails to execute.
4433 5. There are several places where Exim runs a non-Exim command in a
4434 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4435 was being done only for the command run by the queryprogram router. It is
4436 now done for all such subprocesses. The other cases are: ${run, transport
4437 filters, and the commands run by the lmtp and pipe transports.
4439 6. Added CONFIGURE_GROUP build-time option.
4441 7. Some older OS have a limit of 256 on the maximum number of file
4442 descriptors. Exim was using setrlimit() to set 1000 as a large value
4443 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4444 systems. I've change it so that if it can't get 1000, it tries for 256.
4446 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4447 was an oversight, and furthermore, ever since the addition of extra
4448 controls (e.g. 4.43/32), the checks on when to allow different forms of
4449 "control" were broken. There should now be diagnostics for all cases when a
4450 control that does not make sense is encountered.
4452 9. Added the /retain_sender option to "control=submission".
4454 10. $recipients is now available in the predata ACL (oversight).
4456 11. Tidy the search cache before the fork to do a delivery from a message
4457 received from the command line. Otherwise the child will trigger a lookup
4458 failure and thereby defer the delivery if it tries to use (for example) a
4459 cached ldap connection that the parent has called unbind on.
4461 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4462 of $address_data from the recipient verification was clobbered by the
4463 sender verification.
4465 13. The value of address_data from a sender verification is now available in
4466 $sender_address_data in subsequent conditions in the ACL statement.
4468 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4470 15. Added a new option "connect=<time>" to callout options, to set a different
4473 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4474 was its contents. (It was OK if the option was not defined at all.)
4476 17. A "Completed" log line is now written for messages that are removed from
4477 the spool by the -Mrm option.
4479 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4480 information about exactly what failed.
4482 19. Added -dd to debug only the daemon process.
4484 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4485 handles timeouts, both on the server side and network timeouts. Renamed the
4486 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4488 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4489 It is now set to "smtps".
4491 22. $host_address is now set to the target address during the checking of
4492 ignore_target_hosts.
4494 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4495 being passed; this would have caused $sender_host_name to have been used if
4496 matching the list had actually called for a host name (not very likely,
4497 since this list is usually IP addresses). A host name is now passed as
4500 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4501 code that specifies a non-restarting handler (typically sigaction() in
4502 modern systems) in an attempt to fix a rare and obscure crash bug.
4504 25. Narrowed the window for a race in the daemon that could cause it to ignore
4505 SIGCHLD signals. This is not a major problem, because they are used only to
4506 wake it up if nothing else does.
4508 26. A malformed maildirsize file could cause Exim to calculate negative values
4509 for the mailbox size or file count. Odd effects could occur as a result.
4510 The maildirsize information is now recalculated if the size or filecount
4513 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4514 support for a long time. Removed HAVE_SYS_VFS_H.
4516 28. Installed the latest version of exipick from John Jetmore.
4518 29. In an address list, if the pattern was not a regular expression, an empty
4519 subject address (from a bounce message) matched only if the pattern was an
4520 empty string. Non-empty patterns were not even tested. This was the wrong
4521 because it is perfectly reasonable to use an empty address as part of a
4522 database query. An empty address is now tested by patterns that are
4523 lookups. However, all the other forms of pattern expect the subject to
4524 contain a local part and a domain, and therefore, for them, an empty
4525 address still always fails if the pattern is not itself empty.
4527 30. Exim went into a mad DNS loop when attempting to do a callout where the
4528 host was specified on an smtp transport, and looking it up yielded more
4529 than one IP address.
4531 31. Re-factored the code for checking spool and log partition space into a
4532 function that finds that data and another that does the check. The former
4533 is then used to implement four new variables: $spool_space, $log_space,
4534 $spool_inodes, and $log_inodes.
4536 32. The RFC2047 encoding function was originally intended for short strings
4537 such as real names; it was not keeping to the 75-character limit for
4538 encoded words that the RFC imposes. It now respects the limit, and
4539 generates multiple encoded words if necessary. To be on the safe side, I
4540 have increased the buffer size for the ${rfc2047: expansion operator from
4543 33. It is now permitted to omit both strings after an "if" condition; if the
4544 condition is true, the result is "true". As before, when the second string
4545 is omitted, a false condition yields an empty string. This makes it less
4546 cumbersome to write custom ACL and router conditions.
4548 34. Failure to deliver a bounce message always caused it to be frozen, even if
4549 there was an errors_to setting on the router. The errors_to setting is now
4552 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4553 canonical form (fully expanded) before being placed in
4554 $sender_host_address.
4556 36. The table in the code that translates DNS record types into text (T_A to
4557 "A" for instance) was missing entries for NS and CNAME. It is just possible
4558 that this could have caused confusion if both these types were looked up
4559 for the same domain, because the text type is used as part of Exim's
4560 per-process caching. But the chance of anyone hitting this buglet seems
4563 37. The dnsdb lookup has been extended in a number of ways.
4565 (1) There is a new type, "zns", which walks up the domain tree until it
4566 finds some nameserver records. It should be used with care.
4568 (2) There is a new type, "mxh", which is like "mx" except that it returns
4569 just the host names, not the priorities.
4571 (3) It is now possible to give a list of domains (or IP addresses) to be
4572 looked up. The behaviour when one of the lookups defers can be
4573 controlled by a keyword.
4575 (4) It is now possible to specify the separator character for use when
4576 multiple records are returned.
4578 38. The dnslists ACL condition has been extended: it is now possible to supply
4579 a list of IP addresses and/or domains to be looked up in a particular DNS
4582 39. Added log_selector=+queue_time_overall.
4584 40. When running the queue in the test harness, wait just a tad after forking a
4585 delivery process, to get repeatability of debugging output.
4587 41. Include certificate and key file names in error message when GnuTLS fails
4588 to set them up, because the GnuTLS error message doesn't include the name
4589 of the failing file when there is a problem reading it.
4591 42. Allow both -bf and -bF in the same test run.
4593 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4595 44. The "Exiscan patch" is now merged into the mainline Exim source.
4597 45. Sometimes the final signoff response after QUIT could fail to get
4598 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4599 < 0 before doing a fflush(). This bug looks as though it goes back to the
4600 introduction of TLS in release 3.20, but "sometimes" must have been rare
4601 because the tests only now provoked it.
4603 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4604 (this can affect the format of dates).
4606 47. exim_tidydb, when checking for the continued existence of a message for
4607 which it has found a message-specific retry record, was not finding
4608 messages that were in split spool directories. Consequently, it was
4609 deleting retry records that should have stayed in existence.
4611 48. Steve fixed some bugs in eximstats.
4613 49. The SPA authentication driver was not abandoning authentication and moving
4614 on to the next authenticator when an expansion was forced to fail,
4615 contradicting the general specification for all authenticators. Instead it
4616 was generating a temporary error. It now behaves as specified.
4618 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4619 (the order specifies the preference for clients). The order is now AES256,
4620 AES128, 3DES, ARCFOUR128.
4622 51. Small patch to Sieve code - explicitly set From: when generating an
4625 52. Exim crashed if a remote delivery caused a very long error message to be
4626 recorded - for instance if somebody sent an entire SpamAssassin report back
4627 as a large number of 550 error lines. This bug was coincidentally fixed by
4628 increasing the size of one of Exim's internal buffers (big_buffer) that
4629 happened as part of the Exiscan merge. However, to be on the safe side, I
4630 have made the code more robust (and fixed the comments that describe what
4633 53. Now that there can be additional text after "Completed" in log lines (if
4634 the queue_time_overall log selector is set), a one-byte patch to exigrep
4635 was needed to allow it to recognize "Completed" as not the last thing in
4638 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4639 patch that reportedly fixes this has been added. I am not expert enough to
4640 create a test for it. This is what the patch creator wrote:
4642 "I found a little strange behaviour of ldap code when working with
4643 Windows 2003 AD Domain, where users was placed in more than one
4644 Organization Units. When I tried to give exim partial DN, the exit code
4645 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4646 But simultaneously result of request was absolutely normal ldap result,
4647 so I produce this patch..."
4649 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4650 REFERENCE, so I have modified the code to exclude the patch when that macro
4653 55. Some experimental protocols are using DNS PTR records for new purposes. The
4654 keys for these records are domain names, not reversed IP addresses. The
4655 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4656 leaves it alone. Component reversal etc. now happens only for IP addresses.
4659 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4661 57. Double the size of the debug message buffer (to 2048) so that more of very
4662 long debug lines gets shown.
4664 58. The exicyclog utility now does better if the number of log files to keep
4665 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4667 59. Two changes related to the smtp_active_hostname option:
4669 (1) $smtp_active_hostname is now available as a variable.
4670 (2) The default for smtp_banner uses $smtp_active_hostname instead
4671 of $primary_hostname.
4673 60. The host_aton() function is supposed to be passed a string that is known
4674 to be a valid IP address. However, in the case of IPv6 addresses, it was
4675 not checking this. This is a hostage to fortune. Exim now panics and dies
4676 if the condition is not met. A case was found where this could be provoked
4677 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4678 components; fortuitously, this particular loophole had already been fixed
4679 by change 4.50/55 above.
4681 If there are any other similar loopholes, the new check in host_aton()
4682 itself should stop them being exploited. The report I received stated that
4683 data on the command line could provoke the exploit when Exim was running as
4684 exim, but did not say which command line option was involved. All I could
4685 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4686 running as the user.
4689 61. There was a buffer overflow vulnerability in the SPA authentication code
4690 (which came originally from the Samba project). I have added a test to the
4691 spa_base64_to_bits() function which I hope fixes it.
4694 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4695 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4697 63. The daemon start-up calls getloadavg() while still root for those OS that
4698 need the first call to be done as root, but it missed one case: when
4699 deliver_queue_load_max is set with deliver_drop_privilege. This is
4700 necessary for the benefit of the queue runner, because there is no re-exec
4701 when deliver_drop_privilege is set.
4703 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4704 This has been fixed.
4706 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4707 was in use, was not putting the data itself into the right store pool;
4708 consequently, it could be overwritten for a subsequent message in the same
4709 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4712 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4714 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4715 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4716 the function were treating the return as a boolean value, which happened to
4717 work because 0=false and not-0=true, but is not correct code.
4719 68. The host_aton() function was not handling scoped IPv6 addresses (those
4720 with, for example, "%eth0" on the end) correctly.
4722 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4723 negated items (that is, ~something) in unsigned ints. Some compilers
4724 apparently mutter when there is no cast.
4726 70. If an address verification called from an ACL failed, and did not produce a
4727 user-specific message (i.e. there was only a "system" message), nothing was
4728 put in $acl_verify_message. In this situation, it now puts the system
4731 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4732 session; change 4.31/43 added the unwanted input to the log line - except
4733 that it did not do this in the start of session case. It now does.
4735 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4736 This is wrong and can cause the other end to generate a synchronization
4737 error if it is another Exim or anything else that does the synchronization
4738 check. A QUIT command is no longer sent after a timeout.
4740 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4741 during host lookups.
4743 74. The defer_ok option of callout verification was not working if it was used
4744 when verifying addresses in header lines, that is, for this case:
4746 verify = header_sender/callout=defer_ok
4748 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4749 those file descriptors could be used for SMTP connections. If anything
4750 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4751 could be sent to the SMTP client, causing chaos. The daemon now opens
4752 stdin, stdout, and stderr to /dev/null when it puts itself into the
4755 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4756 log by default. The user can override this with suitable Perl magic.
4758 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4759 the log message when discard triggers, was not working for the DATA ACL or
4760 for the non-SMTP ACL.
4762 78. Error message wording change in sieve.c.
4764 79. If smtp_accept_max_per_host was set, the number of connections could be
4765 restricted to fewer than expected, because the daemon was trying to set up
4766 a new connection before checking whether the processes handling previous
4767 connections had finished. The check for completed processes is now done
4768 earlier. On busy systems, this bug wouldn't be noticed because something
4769 else would have woken the daemon, and it would have reaped the completed
4772 80. If a message was submitted locally by a user whose login name contained one
4773 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4774 It caused a spool format error. I have fixed the spool reading code. A
4775 related problem was that the "from" clause in the Received: line became
4776 illegal because of the space(s). It is now covered by ${quote_local_part.
4778 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4781 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4783 83. Patch from Sieve maintainer to fix unterminated string problem in
4784 "vacation" handling.
4786 84. Some minor changes to the Linux configuration files to help with other
4787 OS variants using glibc.
4789 85. One more patch for Sieve to update vacation handling to latest spec.
4792 ----------------------------------------------------
4793 See the note above about the 4.44 and 4.50 releases.
4794 ----------------------------------------------------
4800 1. Change 4.43/35 introduced a bug that caused file counts to be
4801 incorrectly computed when quota_filecount was set in an appendfile
4804 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4805 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4808 3. Give more explanation in the error message when the command for a transport
4809 filter fails to execute.
4811 4. There are several places where Exim runs a non-Exim command in a
4812 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4813 was being done only for the command run by the queryprogram router. It is
4814 now done for all such subprocesses. The other cases are: ${run, transport
4815 filters, and the commands run by the lmtp and pipe transports.
4817 5. Some older OS have a limit of 256 on the maximum number of file
4818 descriptors. Exim was using setrlimit() to set 1000 as a large value
4819 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4820 systems. I've change it so that if it can't get 1000, it tries for 256.
4822 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4823 was an oversight, and furthermore, ever since the addition of extra
4824 controls (e.g. 4.43/32), the checks on when to allow different forms of
4825 "control" were broken. There should now be diagnostics for all cases when a
4826 control that does not make sense is encountered.
4828 7. $recipients is now available in the predata ACL (oversight).
4830 8. Tidy the search cache before the fork to do a delivery from a message
4831 received from the command line. Otherwise the child will trigger a lookup
4832 failure and thereby defer the delivery if it tries to use (for example) a
4833 cached ldap connection that the parent has called unbind on.
4835 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4836 of $address_data from the recipient verification was clobbered by the
4837 sender verification.
4839 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4840 was its contents. (It was OK if the option was not defined at all.)
4842 11. A "Completed" log line is now written for messages that are removed from
4843 the spool by the -Mrm option.
4845 12. $host_address is now set to the target address during the checking of
4846 ignore_target_hosts.
4848 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4849 being passed; this would have caused $sender_host_name to have been used if
4850 matching the list had actually called for a host name (not very likely,
4851 since this list is usually IP addresses). A host name is now passed as
4854 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4855 code that specifies a non-restarting handler (typically sigaction() in
4856 modern systems) in an attempt to fix a rare and obscure crash bug.
4858 15. Narrowed the window for a race in the daemon that could cause it to ignore
4859 SIGCHLD signals. This is not a major problem, because they are used only to
4860 wake it up if nothing else does.
4862 16. A malformed maildirsize file could cause Exim to calculate negative values
4863 for the mailbox size or file count. Odd effects could occur as a result.
4864 The maildirsize information is now recalculated if the size or filecount
4867 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4868 support for a long time. Removed HAVE_SYS_VFS_H.
4870 18. Updated exipick to current release from John Jetmore.
4872 19. Allow an empty sender to be matched against a lookup in an address list.
4873 Previously the only cases considered were a regular expression, or an
4876 20. Exim went into a mad DNS lookup loop when doing a callout where the
4877 host was specified on the transport, if the DNS lookup yielded more than
4880 21. The RFC2047 encoding function was originally intended for short strings
4881 such as real names; it was not keeping to the 75-character limit for
4882 encoded words that the RFC imposes. It now respects the limit, and
4883 generates multiple encoded words if necessary. To be on the safe side, I
4884 have increased the buffer size for the ${rfc2047: expansion operator from
4887 22. Failure to deliver a bounce message always caused it to be frozen, even if
4888 there was an errors_to setting on the router. The errors_to setting is now
4891 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4892 canonical form (fully expanded) before being placed in
4893 $sender_host_address.
4895 24. Updated eximstats to version 1.33
4897 25. Include certificate and key file names in error message when GnuTLS fails
4898 to set them up, because the GnuTLS error message doesn't include the name
4899 of the failing file when there is a problem reading it.
4901 26. Expand error message when OpenSSL has problems setting up cert/key files.
4904 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4905 (this can affect the format of dates).
4907 28. exim_tidydb, when checking for the continued existence of a message for
4908 which it has found a message-specific retry record, was not finding
4909 messages that were in split spool directories. Consequently, it was
4910 deleting retry records that should have stayed in existence.
4912 29. eximstats updated to version 1.35
4913 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4914 1.35 - bugfix such that pie charts by volume are generated correctly
4916 30. The SPA authentication driver was not abandoning authentication and moving
4917 on to the next authenticator when an expansion was forced to fail,
4918 contradicting the general specification for all authenticators. Instead it
4919 was generating a temporary error. It now behaves as specified.
4921 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4922 (the order specifies the preference for clients). The order is now AES256,
4923 AES128, 3DES, ARCFOUR128.
4925 31. Small patch to Sieve code - explicitly set From: when generating an
4928 32. Exim crashed if a remote delivery caused a very long error message to be
4929 recorded - for instance if somebody sent an entire SpamAssassin report back
4930 as a large number of 550 error lines. This bug was coincidentally fixed by
4931 increasing the size of one of Exim's internal buffers (big_buffer) that
4932 happened as part of the Exiscan merge. However, to be on the safe side, I
4933 have made the code more robust (and fixed the comments that describe what
4936 33. Some experimental protocols are using DNS PTR records for new purposes. The
4937 keys for these records are domain names, not reversed IP addresses. The
4938 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4939 leaves it alone. Component reversal etc. now happens only for IP addresses.
4942 34. The host_aton() function is supposed to be passed a string that is known
4943 to be a valid IP address. However, in the case of IPv6 addresses, it was
4944 not checking this. This is a hostage to fortune. Exim now panics and dies
4945 if the condition is not met. A case was found where this could be provoked
4946 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4947 components; fortuitously, this particular loophole had already been fixed
4948 by change 4.50/55 or 4.44/33 above.
4950 If there are any other similar loopholes, the new check in host_aton()
4951 itself should stop them being exploited. The report I received stated that
4952 data on the command line could provoke the exploit when Exim was running as
4953 exim, but did not say which command line option was involved. All I could
4954 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4955 running as the user.
4958 35. There was a buffer overflow vulnerability in the SPA authentication code
4959 (which came originally from the Samba project). I have added a test to the
4960 spa_base64_to_bits() function which I hope fixes it.
4963 36. The daemon start-up calls getloadavg() while still root for those OS that
4964 need the first call to be done as root, but it missed one case: when
4965 deliver_queue_load_max is set with deliver_drop_privilege. This is
4966 necessary for the benefit of the queue runner, because there is no re-exec
4967 when deliver_drop_privilege is set.
4969 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4970 was in use, was not putting the data itself into the right store pool;
4971 consequently, it could be overwritten for a subsequent message in the same
4972 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4975 38. Sometimes the final signoff response after QUIT could fail to get
4976 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4977 < 0 before doing a fflush(). This bug looks as though it goes back to the
4978 introduction of TLS in release 3.20, but "sometimes" must have been rare
4979 because the tests only now provoked it.
4985 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4986 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4987 Now it returns an integer. A number of calls were still expecting a T/F
4988 return. Fortuitously, in all cases, the tests worked in OK situations,
4989 which is the norm. However, things would have gone wrong on any write
4990 failures on the smtp file descriptor. This function is used when sending
4991 messages over SMTP and also when doing verify callouts.
4993 2. When Exim is called to do synchronous delivery of a locally submitted
4994 message (the -odf or -odi options), it no longer closes stderr before doing
4997 3. Implemented the mua_wrapper option.
4999 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5001 5. Implemented the functions header_remove(), header_testname(),
5002 header_add_at_position(), and receive_remove_recipient(), and exported them
5005 6. If an ACL "warn" statement specified the addition of headers, Exim already
5006 inserted X-ACL-Warn: at the start if there was no header name. However, it
5007 was not making this test for the second and subsequent header lines if
5008 there were newlines in the string. This meant that an invalid header could
5009 be inserted if Exim was badly configured.
5011 7. Allow an ACL "warn" statement to add header lines at the start or after all
5012 the Received: headers, as well as at the end.
5014 8. Added the rcpt_4xx retry error code.
5016 9. Added postmaster_mailfrom=xxx to callout verification option.
5018 10. Added mailfrom=xxxx to the callout verification option, for verify=
5021 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5022 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5024 12. Inserted some casts to stop certain compilers warning when using pointer
5025 differences as field lengths or precisions in printf-type calls (mostly
5026 affecting debugging statements).
5028 13. Added optional readline() support for -be (dynamically loaded).
5030 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5031 same clock tick as a message's arrival, so that its received time was the
5032 same as the "first fail" time on the retry record, and that message
5033 remained on the queue past the ultimate address timeout, every queue runner
5034 would try a delivery (because it was past the ultimate address timeout) but
5035 after another failure, the ultimate address timeout, which should have then
5036 bounced the address, did not kick in. This was a "< instead of <=" error;
5037 in most cases the first failure would have been in the next clock tick
5038 after the received time, and all would be well.
5040 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5041 being recognized when the domain list was tested by the match_domain
5042 condition in an expansion string.
5044 16. Added the ${str2b64: operator.
5046 17. Exim was always calling setrlimit() to set a large limit for the number of
5047 processes, without checking whether the existing limit was already
5048 adequate. (It did check for the limit on file descriptors.) Furthermore,
5049 errors from getrlimit() and setrlimit() were being ignored. Now they are
5050 logged to the main and panic logs, but Exim does carry on, to try to do its
5051 job under whatever limits there are.
5053 18. Imported PCRE 5.0.
5055 19. Trivial typo in log message " temporarily refused connection" (the leading
5058 20. If the log selector return_path_on_delivery was set and an address was
5059 redirected to /dev/null, the delivery process crashed because it assumed
5060 that a return path would always be set for a "successful" delivery. In this
5061 case, the whole delivery is bypassed as an optimization, and therefore no
5064 21. Internal re-arrangement: the function for sending a challenge and reading
5065 a response while authentication was assuming a zero-terminated challenge
5066 string. It's now changed to take a pointer and a length, to allow for
5067 binary data in such strings.
5069 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5071 23. Exim was not respecting finduser_retries when seeking the login of the
5072 uid under which it was called; it was always trying 10 times. (The default
5073 setting of finduser_retries is zero.) Also, it was sleeping after the final
5074 failure, which is pointless.
5076 24. Implemented tls_on_connect_ports.
5078 25. Implemented acl_smtp_predata.
5080 26. If the domain in control=submission is set empty, Exim assumes that the
5081 authenticated id is a complete email address when it generates From: or
5082 Sender: header lines.
5084 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5085 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5086 chown and chgrp in /bin and hostname in /usr/bin.
5088 28. Exim was keeping the "process log" file open after each use, just as it
5089 does for the main log. This opens the possibility of it remaining open for
5090 long periods when the USR1 signal hits a daemon. Occasional processlog
5091 errors were reported, that could have been caused by this. Anyway, it seems
5092 much more sensible not to leave this file open at all, so that is what now
5095 29. The long-running daemon process does not normally write to the log once it
5096 has entered its main loop, and it closes the log before doing so. This is
5097 so that log files can straightforwardly be renamed and moved. However,
5098 there are a couple of unusual error situations where the daemon does write
5099 log entries, and I had neglected to close the log afterwards.
5101 30. The text of an SMTP error response that was received during a remote
5102 delivery was being truncated at 512 bytes. This is too short for some of
5103 the long messages that one sometimes sees. I've increased the limit to
5106 31. It is now possible to make retry rules that apply only when a message has a
5107 specific sender, in particular, an empty sender.
5109 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5110 it possible to be selective about when SMTP synchronization is enforced.
5112 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5114 32. Implemented hosts_connection_nolog.
5116 33. Added an ACL for QUIT.
5118 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5121 35. Added mailbox_size and mailbox_filecount to appendfile.
5123 36. Added control = no_multiline_responses to ACLs.
5125 37. There was a bug in the logic of the code that waits for the clock to tick
5126 in the case where the clock went backwards by a substantial amount such
5127 that the microsecond fraction of "now" was more than the microsecond
5128 fraction of "then" (but the whole seconds number was less).
5130 38. Added support for the libradius Radius client library this is found on
5131 FreeBSD (previously only the radiusclient library was supported).
5137 1. When certain lookups returned multiple values in the form name=value, the
5138 quoting of the values was not always being done properly. Specifically:
5139 (a) If the value started with a double quote, but contained no whitespace,
5141 (b) If the value contained whitespace other than a space character (i.e.
5142 tabs or newlines or carriage returns) it was not quoted.
5143 This fix has been applied to the mysql and pgsql lookups by writing a
5144 separate quoting function and calling it from the lookup code. The fix
5145 should probably also be applied to nisplus, ibase and oracle lookups, but
5146 since I cannot test any of those, I have not disturbed their existing code.
5148 2. A hit in the callout cache for a specific address caused a log line with no
5149 reason for rejecting RCPT. Now it says "Previous (cached) callout
5150 verification failure".
5152 3. There was an off-by-one bug in the queryprogram router. An over-long
5153 return line was truncated at 256 instead of 255 characters, thereby
5154 overflowing its buffer with the terminating zero. As well as fixing this, I
5155 have increased the buffer size to 1024 (and made a note to document this).
5157 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5158 when Exim is waiting for an SMTP response from a remote server, Exim
5159 restarts its select() call on the socket, thereby resetting its timeout.
5160 This is not a problem when such interrupts are rare. Somebody set up a cron
5161 job to run exiwhat every 2 minutes, which is less than the normal select()
5162 timeout (5 or 10 minutes). This meant that the select() timeout never
5163 kicked in because it was always reset. I have fixed this by comparing the
5164 time when an interrupt arrives with the time at the start of the first call
5165 to select(). If more time than the timeout has elapsed, the interrupt is
5166 treated as a timeout.
5168 5. Some internal re-factoring in preparation for the addition of Sieve
5169 extensions (by MH). In particular, the "personal" test is moved to a
5170 separate function, and given an option for scanning Cc: and Bcc: (which is
5171 not set for Exim filters).
5173 6. When Exim created an email address using the login of the caller as the
5174 local part (e.g. when creating a From: or Sender: header line), it was not
5175 quoting the local part when it contained special characters such as @.
5177 7. Installed new OpenBSD configuration files.
5179 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5180 try to make them clearer.
5182 9. Callout options, other than the timeout value, were being ignored when
5183 verifying sender addresses in header lines. For example, when using
5185 verify = header_sender/callout=no_cache
5187 the cache was (incorrectly) being used.
5189 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5190 only the Cygwin environment.
5192 11. When return_path_on_delivery was set as a log selector, if different remote
5193 addresses in the same message used different return paths and parallel
5194 remote delivery occurred, the wrong values would sometimes be logged.
5195 (Whenever a remote delivery process finished, the return path value from
5196 the most recently started remote delivery process was logged.)
5198 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5199 lines when AUTH and/or TLS are in use. This is the "received protocol"
5200 field. Exim used to use "asmtp" for authenticated SMTP, without any
5201 indication (in the protocol name) for TLS use. Now it follows the RFC and
5202 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5203 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5204 names appear in log lines as well as in Received: header lines.
5206 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5207 extensions, and comparison tests, and to fix some bugs.
5209 14. Changes to the "personal" filter test:
5211 (1) The test was buggy in that it was just doing the equivalent of
5212 "contains" tests on header lines. For example, if a user's address was
5213 anne@some.where, the "personal" test would incorrectly be true for
5215 To: susanne@some.where
5217 This test is now done by extracting each address from the header in turn,
5218 and checking the entire address. Other tests that are part of "personal"
5219 are now done using regular expressions (for example, to check local parts
5220 of addresses in From: header lines).
5222 (2) The list of non-personal local parts in From: addresses has been
5223 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5224 taken from the Sieve specification recommendations.
5226 (3) If the message contains any header line starting with "List-" it is
5227 treated as non-personal.
5229 (4) The test for "circular" in the Subject: header line has been removed
5230 because it now seems ill-conceived.
5232 15. Minor typos in src/EDITME comments corrected.
5234 16. Installed latest exipick from John Jetmore.
5236 17. If headers_add on a router specified a text string that was too long for
5237 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5238 of string_sprintf() is now avoided.
5240 18. $message_body_size was not set (it was always zero) when running the DATA
5241 ACL and the local_scan() function.
5243 19. For the "mail" command in an Exim filter, no default was being set for
5244 the once_repeat time, causing a random time value to be used if "once" was
5245 specified. (If the value happened to be <= 0, no repeat happened.) The
5246 default is now 0s, meaning "never repeat". The "vacation" command was OK
5247 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5248 (I found it when inspecting the code).
5250 20. There is now an overall timeout for performing a callout verification. It
5251 defaults to 4 times the callout timeout, which applies to individual SMTP
5252 commands during the callout. The overall timeout applies when there is more
5253 than one host that can be tried. The timeout is checked before trying the
5254 next host. This prevents very long delays if there are a large number of
5255 hosts and all are timing out (e.g. when the network connections are timing
5256 out). The value of the overall timeout can be changed by specifying an
5257 additional sub-option for "callout", called "maxwait". For example:
5259 verify = sender/callout=5s,maxwait=20s
5261 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5262 to the end before writing, but this should make it even safer).
5264 22. Exim was forgetting that it had advertised PIPELINING for the second and
5265 subsequent messages on an SMTP connection. It was also not resetting its
5266 memory on STARTTLS and an internal HELO.
5268 23. When Exim logs an SMTP synchronization error within a session, it now
5269 records whether PIPELINING has been advertised or not.
5271 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5272 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5273 rather than long int.
5275 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5277 26. Added the never_mail option to autoreply.
5283 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5284 crash if the getsockname() call failed; this can happen if a connection is
5285 closed very soon after it is established. The problem was simply in the
5286 order in which certain operations were done, causing Exim to try to write
5287 to the SMTP stream before it had set up the file descriptor. The bug has
5288 been fixed by making things happen in the correct order.
5294 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5295 before the connection was closed, thus losing the rejection response.
5297 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5298 some early Solaris releases, but causes trouble in current releases where
5299 socklen_t is defined.
5301 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5304 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5307 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5308 that releasing the top of it at the end releases what was used for sub-
5309 expansions (unless the block got too big). However, discard this block if
5310 the first thing is a variable or header, so that we can use its block when
5311 it is dynamic (useful for very large $message_headers, for example).
5313 6. Lookups now cache *every* query, not just the most recent. A new, separate
5314 store pool is used for this. It can be recovered when all lookup caches are
5315 flushed. Lookups now release memory at the end of their result strings.
5316 This has involved some general refactoring of the lookup sources.
5318 7. Some code has been added to the store_xxx() functions to reduce the amount
5319 of flapping under certain conditions.
5321 8. log_incoming_interface used to affect only the <= reception log lines. Now
5322 it causes the local interface and port to be added to several more SMTP log
5323 lines, for example "SMTP connection from", and rejection lines.
5325 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5327 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5329 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5330 could be overwritten at the end of the current message (or the start of a
5331 new message if it was set in a HELO ACL). The value is now preserved for
5332 the duration of the SMTP connection.
5334 12. If a transport had a headers_rewrite setting, and a matching header line
5335 contained an unqualified address, that address was qualified, even if it
5336 did not match any rewriting rules. The underlying bug was that the values
5337 of the flags that permit the existence of unqualified sender and recipient
5338 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5339 non-local messages, and by -bnq for local messages) were not being
5340 preserved with the message after it was received.
5342 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5343 "next input=" as part of the text comprising the host identity instead of
5344 the correct text. The code was using the same buffer for two different
5345 strings. However, depending on which order the printing function evaluated
5346 its arguments, the bug did not always show up. Under Linux, for example, my
5347 test suite worked just fine.
5349 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5350 doesn't work with some older versions of Perl. It has been changed to "my",
5351 which in any case is probably the better facility to use.
5353 15. A really picky compiler found some instances of statements for creating
5354 error messages that either had too many or two few arguments for the format
5357 16. The size of the buffer for calls to the DNS resolver has been increased
5358 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5359 for addresses that have a lot of PTR records. This alleviates a problem; it
5360 does not fully solve it.
5362 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5363 buffer now truncates the list and logs the incident, which is the same
5364 action as happens when Exim is looking up a host name and its aliases.
5365 Previously in this situation something unpredictable would happen;
5366 sometimes it was "internal error: store_reset failed".
5368 18. If a server dropped the connection unexpectedly when an Exim client was
5369 using GnuTLS and trying to read a response, the client delivery process
5370 crashed while trying to generate an error log message.
5372 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5373 string, for example:
5375 warn message = H1: something\nH2: something
5377 the text was added as a single header line from Exim's point of view
5378 though it ended up OK in the delivered message. However, searching for the
5379 second and subsequent header lines using $h_h2: did not work. This has been
5380 fixed. Similarly, if a system filter added multiple headers in this way,
5381 the routers could not see them.
5383 20. Expanded the error message when iplsearch is called with an invalid key to
5384 suggest using net-iplsearch in a host list.
5386 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5387 ACLs are no longer actually imposed (and a message to that effect is
5390 22. If a "gecos" field in a passwd entry contained escaped characters, in
5391 particular, if it contained a \" sequence, Exim got it wrong when building
5392 a From: or a Sender: header from that name. A second bug also caused
5393 incorrect handling when an unquoted " was present following a character
5394 that needed quoting.
5396 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5397 was not being matched caselessly.
5399 24. Arranged for all hyphens in the exim.8 source to be escaped with
5402 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5403 properly. Recipient callout cache records were still being keyed to include
5404 the sender, even when use_sender was set false. This led to far more
5405 callouts that were necessary. The sender is no longer included in the key
5406 when use_sender is false.
5408 26. Added "control = submission" modifier to ACLs.
5410 27. Added the ${base62d: operator to decode base 62 numbers.
5412 28. dnsdb lookups can now access SRV records.
5414 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5415 the configuration file.
5417 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5418 (-v) mode. This makes the output for a verbose queue run more intelligible.
5420 31. Added a use_postmaster feature to recipient callouts.
5422 32. Added the $body_zerocount variable, containing the number of binary zero
5423 bytes in the message body.
5425 33. The time of last modification of the "new" subdirectory is now used as the
5426 "mailbox time last read" when there is a quota error for a maildir
5429 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5431 35. Added +ignore_unknown as a special item in host lists.
5433 36. Code for decoding IPv6 addresses in host lists is now included, even if
5434 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5435 address was recognized as an IP address, but was then not correctly decoded
5436 into binary, causing unexpected and incorrect effects when compared with
5443 1. Very minor rewording of debugging text in manualroute to say "list of
5444 hosts" instead of "hostlist".
5446 2. If verify=header_syntax was set, and a header line with an unqualified
5447 address (no domain) and a large number of spaces between the end of the
5448 name and the colon was received, the reception process suffered a buffer
5449 overflow, and (when I tested it) crashed. This was caused by some obsolete
5450 code that should have been removed. The fix is to remove it!
5452 3. When running in the test harness, delay a bit after writing a bounce
5453 message to get a bit more predictability in the log output.
5455 4. Added a call to search_tidyup() just before forking a reception process. In
5456 theory, someone could use a lookup in the expansion of smtp_accept_max_
5457 per_host which, without the tidyup, could leave open a database connection.
5459 5. Added the variables $recipient_data and $sender_data which get set from a
5460 lookup success in an ACL "recipients" or "senders" condition, or a router
5461 "senders" option, similar to $domain_data and $local_part_data.
5463 6. Moved the writing of debug_print from before to after the "senders" test
5466 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5467 problems for message scanning, either using a data ACL, or using
5468 local_scan() because the Received: header was not generated till after they
5469 were called (in order to set the time as the time of reception completion).
5470 I have revised the way this works. The header is now generated after the
5471 body is received, but before the ACL or local_scan() are called. After they
5472 are run, the timestamp in the header is updated.
5478 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5479 before starting a queue runner without re-exec. This happened only when
5480 deliver_drop_privilege was set or when the Exim user was set to root. The
5481 effect of the bug was that timeouts during subsequent deliveries caused
5482 crashes instead of being properly handled. The handler is now left at its
5483 default (and expected) setting.
5485 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5486 message, again when deliver_drop_privilege is set or Exim is run as root.
5487 The bug described in (1) was not present in this case, but the tidying up
5488 of the other signals was missing. I have made the two cases consistent.
5490 3. The ignore_target_hosts setting on a manualroute router was being ignored
5491 for hosts that were looked up using the /MX notation.
5493 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5496 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5497 operated on the sender address. After changing the $sender_address to <>
5498 for the sender address verify, Exim was re-instated it as the original
5499 (before rewriting) address, but remembering that it had rewritten it, so it
5500 wasn't rewriting it again. This bug also had the effect of breaking the
5501 sender address verification caching when the sender address was rewritten.
5503 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5504 This has been changed so that if the ip literal address matches
5505 ignore_target_hosts, the router declines.
5507 7. Added expansion conditions match_domain, match_address, and match_local_
5508 part (NOT match_host).
5510 8. The placeholder for the Received: header didn't have a length field set.
5512 9. Added code to Exim itself and to exim_lock to test for a specific race
5513 condition that could lead to file corruption when using MBX delivery. The
5514 issue is with the lockfile that is created in /tmp. If this file is removed
5515 after a process has opened it but before that process has acquired a lock,
5516 there is the potential for a second process to recreate the file and also
5517 acquire a lock. This could lead to two Exim processes writing to the file
5518 at the same time. The added code performs the same test as UW imapd; it
5519 checks after acquiring the lock that its file descriptor still refers to
5520 the same named file.
5522 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5523 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5526 11. Added the smtp_active_hostname option. If used, this will typically be made
5527 to depend on the incoming interface address. Because $interface_address is
5528 not set up until the daemon has forked a reception process, error responses
5529 that can happen earlier (such as "too many connections") no longer contain
5532 12. If an expansion in a condition on a "warn" statement fails because a lookup
5533 defers, the "warn" statement is abandoned, and the next ACL statement is
5534 processed. Previously this caused the whole ACL to be aborted.
5536 13. Added the iplsearch lookup type.
5538 14. Added ident_timeout as a log selector.
5540 15. Added tls_certificate_verified as a log selector.
5542 16. Added a global option tls_require_ciphers (compare the smtp transport
5543 option of the same name). This controls incoming TLS connections.
5545 17. I finally figured out how to make tls_require_ciphers do a similar thing
5546 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5547 before starting the TLS session.
5549 18. Tabs are now shown as \t in -bP output.
5551 19. If the log selector return_path_on_delivery was set, Exim crashed when
5552 bouncing a message because it had too many Received: header lines.
5554 20. If two routers both had headers_remove settings, and the first one included
5555 a superfluous trailing colon, the final name in the first list and the
5556 first name in the second list were incorrectly joined into one item (with a
5557 colon in the middle).
5563 1. Added -C and -D options to the exinext utility, mainly to make it easier
5564 to include in the automated testing, but these could be helpful when
5565 multiple configurations are in use.
5567 2. The exinext utility was not formatting the output nicely when there was
5568 an alternate port involved in the retry record key, nor when there was a
5569 message id as well (for retries that were specific to a specific message
5570 and a specific host). It was also confused by IPv6 addresses, because of
5571 the additional colons they contain. I have fixed the IPv4 problem, and
5572 patched it up to do a reasonable job for IPv6.
5574 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5575 delivery, the log line now contains "pipelined" if PIPELINING was used.
5577 4. An SMTP transport process used to panic and die if the bind() call to set
5578 an explicit outgoing interface failed. This has been changed; it is now
5579 treated in the same way as a connect() failure.
5581 5. A reference to $sender_host_name in the part of a conditional expansion
5582 that was being skipped was still causing a DNS lookup. This no longer
5585 6. The def: expansion condition was not recognizing references to header lines
5586 that used bh_ and bheader_.
5588 7. Added the _cache feature to named lists.
5590 8. The code for checking quota_filecount in the appendfile transport was
5591 allowing one more file than it should have been.
5593 9. For compatibility with Sendmail, the command line option
5601 and sets the incoming protocol and host name (for trusted callers). The
5602 host name and its colon can be omitted when only the protocol is to be set.
5603 Note the Exim already has two private options, -pd and -ps, that refer to
5604 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5605 "s", but I don't think that's a major issue.
5607 10. A number of refactoring changes to the code, none of which should affect
5610 (a) The number of logging options was getting close to filling up the
5611 32-bit word that was used as a bit map. I have split them into two classes:
5612 those that are passed in the argument to log_write(), and those that are
5613 only ever tested independently outside of that function. These are now in
5614 separate 32-bit words, so there is plenty of room for expansion again.
5615 There is no change in the user interface or the logging behaviour.
5617 (b) When building, for example, log lines, the code previously used a
5618 macro that called string_cat() twice, in order to add two strings. This is
5619 not really sufficiently general. Furthermore, there was one instance where
5620 it was actually wrong because one of the argument was used twice, and in
5621 one call a function was used. (As it happened, calling the function twice
5622 did not affect the overall behaviour.) The macro has been replaced by a
5623 function that can join an arbitrary number of extra strings onto a growing
5626 (c) The code for expansion conditions now uses a table and a binary chop
5627 instead of a serial search (which was left over from when there were very
5628 few conditions). Also, it now recognizes conditions like "pam" even when
5629 the relevant support is not compiled in: a suitably worded error message is
5630 given if an attempt is made to use such a condition.
5632 11. Added ${time_interval:xxxxx}.
5634 12. A bug was causing one of the ddress fields not to be passed back correctly
5635 from remote delivery subprocesses. The field in question was not being
5636 subsequently used, so this caused to problems in practice.
5638 13. Added new log selectors queue_time and deliver_time.
5640 14. Might have fixed a bug in maildirsizefile handling that threw up
5641 "unexpected character" debug warnings, and recalculated the data
5642 unnecessarily. In any case, I expanded the warning message to give more
5645 15. Added the message "Restricted characters in address" to the statements in
5646 the default ACL that block characters like @ and % in local parts.
5648 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5649 Three changes have been made:
5651 (a) There was a serious bug; a negative response to MAIL caused the whole
5652 recipient domain to be cached as invalid, thereby blocking all messages
5653 to all local parts at the same domain, from all senders. This bug has
5654 been fixed. The domain is no longer cached after a negative response to
5655 MAIL if the sender used is not empty.
5657 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5660 (c) A new callout option, "use_sender" has been added for people who want
5661 the modified behaviour.
5667 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5670 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5671 indeed breaks things for older releases.
5673 3. Added additional logging to the case where there is a problem reading data
5674 from a filter that is running in a subprocess using a pipe, in order to
5675 try to track down a specific problem.
5677 4. Testing facility fudge: when running in the test harness and attempting
5678 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5679 getting "No route to host". Convert this to a timeout.
5681 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5684 6. Some OS don't have socklen_t but use size_t instead. This affects the
5685 fifth argument of getsockopt() amongst other things. This is now
5686 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5687 can be set for individual OS. I have set it for SunOS5, OSF1, and
5688 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5689 some earlier ones do not.
5691 7. Change 4.30/15 was not doing the test caselessly.
5693 8. The standard form for an IPv6 address literal was being rejected by address
5694 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5695 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5696 this, as well as the form without the "IPv6" on the front (but only when
5697 address literals are enabled, of course).
5699 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5701 10. Exim crashed if a message with an empty sender address specified by -f
5702 encountered a router with an errors_to setting. This could be provoked only
5703 by a command such as
5707 where an empty string was supplied; "<>" did not hit this bug.
5709 11. Installed PCRE release 4.5.
5711 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5712 remained set. It is now erased.
5714 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5715 times from message ids (which are base 36 rather than the normal 62).
5717 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5718 were being counted as actual protocol errors, and logged if the log
5719 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5720 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5721 and DATA following a set of rejected RCPTs do not count as protocol errors.
5722 In other words, Exim assumes they were pipelined, though this may not
5723 actually be the case. Of course, in all cases the client gets an
5724 appropriate error code.
5726 15. If a lookup fails in an ACL condition, a message about the failure may
5727 be available; it is used if testing the ACL cannot continue, because most
5728 such messages specify what the cause of the deferral is. However, some
5729 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5730 that caused an old message to be retained and used if a later statement
5731 caused a defer, replacing the real cause of the deferral.
5733 16. If an IP address had so many PTR records that the DNS lookup buffer
5734 was not large enough to hold them, Exim could crash while trying to process
5735 the truncated data. It now detects and logs this case.
5737 17. Further to 4.21/58, another change has been made: if (and only if) the
5738 first line of a message (the first header line) ends with CRLF, a bare LF
5739 in a subsequent header line has a space inserted after it, so as not to
5740 terminate the header.
5742 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5743 unnecessarily, used atoi() instead of strtol(), and didn't check the
5744 termination when getting file sizes from file names by regex.
5746 19. Completely re-implemented the support for maildirsize files, in the light
5747 of a number of problems with the previous contributed implementation
5748 (4.30/29). In particular:
5750 . If the quota is zero, the maildirsize file is maintained, but no quota is
5753 . If the maildir directory does not exist, it is created before any attempt
5754 to write a maildirsize file.
5756 . The quota value in the file is just a cache; if the quota is changed in
5757 the transport, the new value overrides.
5759 . A regular expression is available for excluding directories from the
5762 20. The autoreply transport checks the characters in options that define the
5763 message's headers; it allows continued headers, but it was checking with
5764 isspace() after an embedded newline instead of explicitly looking for a
5767 21. If all the "regular" hosts to which an address was routed had passed their
5768 expiry times, and had not reached their retry times, the address was
5769 bounced, even if fallback hosts were defined. Now Exim should go on to try
5772 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5773 equivalent code in the SMTP transport. Some hosts send humungous responses
5774 to HELO/EHLO, more than 1024 it seems.
5776 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5777 gives compiler warnings in some environments. I've now done it "properly",
5780 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5781 (because of problems with the built-in one) was declared to return uschar *
5782 instead of char *, causing compiler failure.
5784 25. Fixed a file descriptor leak when processing alias/forward files.
5786 26. Fixed a minor format string issue in dbfn.c.
5788 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5790 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5791 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5794 29. When a sender address is verified, it is cached, to save repeating the test
5795 when there is more than one recipient in a message. However, when the
5796 verification involves a callout, it is possible for different callout
5797 options to be set for different recipients. It is too complicated to keep
5798 track of this in the cache, so now Exim always runs a verification when a
5799 callout is required, relying on the callout cache for the optimization.
5800 The overhead is duplication of the address routing, but this should not be
5803 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5804 to be verified with callout=postmaster, and the main callout worked but the
5805 postmaster check failed, the verification correctly failed. However, if a
5806 subsequent RCPT command asked for sender verification *without* the
5807 postmaster check, incorrect caching caused this verification also to fail,
5810 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5811 it was not caching the DNS options (qualify_single, search_parents) that
5812 were used when the lookup failed. A subsequent lookup with different
5813 options therefore always gave the same answer, though there were cases
5814 where it should not have. (Example: a "domains = !$mx_any" option on a
5815 dnslookup router: the "domains" option is always processed without any
5816 widening, but the router might have qualify_single set.) Now Exim uses the
5817 cached value only when the same options are set.
5819 32. Added John Jetmore's "exipick" utility to the distribution.
5821 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5822 than a timeout (e.g. a certificate is required, and is not provided), an
5823 Exim server now closes the connection immediately. Previously it waited for
5824 the client to close - but if the client is SSL, it seems that they each
5825 wait for each other, leading to a delay before one of them times out.
5827 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5828 maintained 0.8.x compatibility because I don't think many are using it, and
5829 it is clearly obsolete.
5831 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5834 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5835 client certificate was expired. A simple patch fixes this, though I don't
5836 understand the full logic of why the verify callback is called multiple
5839 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5840 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5841 which causes problems with some clients (such as the Certicom SSL Plus
5842 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5843 disables the coutermeasure allowing Eudora to connect."
5845 38. Exim was not checking that a write() to a log file succeeded. This could
5846 lead to Bad Things if a log got too big, in particular if it hit a file
5847 size limit. Exim now panics and dies if it cannot write to a log file, just
5848 as it does if it cannot open a log file.
5850 39. Modified OS/Makefile-Linux so that it now contains
5852 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5854 The two -D definitions ensure that Exim is compiled with large file
5855 support, which makes it possible to handle log files that are bigger than
5858 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5859 instance) a domain was checked against a named list that involved a lookup,
5860 causing $domain_data to be set, then another domain was checked against the
5861 same list, then the first domain was re-checked, the value of $domain_data
5862 after the final check could be wrong. In particular, if the second check
5863 failed, it could be set empty. This bug probably also applied to
5866 41. The strip_trailing_dot option was not being applied to the address given
5867 with the -f command-line option.
5869 42. The code for reading a message's header from the spool was incrementing
5870 $received_count, but never initializing it. This meant that the value was
5871 incorrect (doubled) while delivering a message in the same process in which
5872 it was received. In the most common configuration of Exim, this never
5873 happens - a fresh exec is done - but it can happen when
5874 deliver_drop_privilege is set.
5876 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5877 it now includes up to 150 characters of the unexpected data in the log
5880 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5881 and building data strings. The size of both of these buffers was 10 000
5882 bytes - far larger than anybody would *ever* want, thought I. Needless to
5883 say, somebody hit the limit. I have increased the maximum line length to
5884 20 000 and the maximum data length of concatenated lines to 100 000. I have
5885 also fixed two bugs, because there was no checking on these buffers. Tsk,
5886 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5887 buffer is too small.
5889 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5890 lsearch lookups. Now it does.
5892 46. When parsing a route_list item in a manualroute router, a fixed-length
5893 buffer was used for the list of hosts. I made this 1024 bytes long,
5894 thinking that nobody would ever have a list of hosts that long. Wrong.
5895 Somebody had a whole pile of complicated expansion conditions, and the
5896 string was silently truncated, leading to an expansion error. It turns out
5897 that it is easier to change to an unlimited length (owing to other changes
5898 that have happened since this code was originally written) than to build
5899 structure for giving a limitation error. The length of the item that
5900 expands into the list of hosts is now unlimited.
5902 47. The lsearch lookup could not handle data where the length of text line was
5903 more than 4095 characters. Such lines were truncated, leading to shortened
5904 data being returned. It should now handle lines of any length.
5906 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5907 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5910 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5911 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5912 no longer happens in comment lines. A list of replacements is now placed
5913 at the head of all of the source files, except those whose only change is
5914 to replace PERL_COMMAND in the very first #! line.
5916 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5917 messages on the queue, with a bottom-up merge sort, using code contributed
5918 by Michael Haardt. This should make operations like -bp somewhat faster on
5919 large queues. It won't affect queue runners, except when queue_run_in_order
5922 51. Installed eximstats 1.31 in the distribution.
5924 52. Added support for SRV lookups to the dnslookup router.
5926 53. If an ACL referred to $message_body or $message_body_end, the value was not
5927 reset for any messages that followed in the same SMTP session.
5929 54. The store-handling optimization for building very long strings was not
5930 differentiating between the different store pools. I don't think this
5931 actually made any difference in practice, but I've tidied it.
5933 55. While running the routers to verify a sender address, $sender_address
5934 was still set to the sender address. This is wrong, because when routing to
5935 send a bounce to the sender, it would be empty. Therefore, I have changed
5936 it so that, while verifying a sender address, $sender_address is set to <>.
5937 (There is no change to what happens when verifying a recipient address.)
5939 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5940 target A or AAAA records (if not already returned) without resetting the
5941 qualify_single or search_parents options of the DNS resolver. These are
5942 inappropriate in this case because the targets of MX and SRV records must
5943 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5944 target that, when qualified, matched something in the local domain. These
5945 two options are now turned off when doing these lookups.
5947 57. It seems that at least some releases of Reiserfs (which does not have the
5948 concept of a fixed number of inodes) returns zero and not -1 for the
5949 number of available inodes. This interacted badly with check_spool_inodes,
5950 which assumed that -1 was the "no such thing" setting. What I have done is
5951 to check that the total number of inodes is greater than zero before doing
5952 the test of how many are available.
5954 58. When a "warn" ACL statement has a log_message modifier, the message is
5955 remembered, and not repeated. This is to avoid a lot of repetition when a
5956 message has many recipients that cause the same warning to be written.
5957 However, Exim was preserving the list of already written lines for an
5958 entire SMTP session, which doesn't seem right. The memory is now reset if a
5959 new message is started.
5961 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5962 debugging output unless log_rewrite was also set.
5964 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5965 of (int)(handle) when we know that handle contains (void *)(-1).
5967 61. The Exim daemon panic-logs an error return when it closes the incoming
5968 connection. However "connection reset by peer" seems to be common, and
5969 isn't really an error worthy of noting specially, so that particular error
5972 62. When Exim is trying to find all the local interfaces, it used to panic and
5973 die if the ioctl to get the interface flags failed. However, it seems that
5974 on at least one OS (Solaris 9) it is possible to have an interface that is
5975 included in the list of interfaces, but for which you get a failure error
5976 for this call. This happens when the interface is not "plumbed" into a
5977 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5978 failure of the "get flags" call assumes that the interface is down.
5980 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5981 makes life easier for people who are doing arithmetic on fields extracted
5982 from dates, where you often get leading zeros that should not be
5983 interpreted as octal.
5985 64. Added qualify_domain to the redirect router, to override the global
5988 65. If a pathologically long header line contained very many addresses (the
5989 report of this problem mentioned 10 000) and each of them was rewritten,
5990 Exim could use up a very large amount of memory. (It kept on making new
5991 copies of the header line as it rewrote, and never released the old ones.)
5992 At the expense of a bit more processing, the header rewriting function has
5993 been changed so that it no longer eats memory in this way.
5995 66. The generation of the Received: header has been moved from the time that a
5996 message starts to be received, to the time that it finishes. The timestamp
5997 in the Received: header should now be very close to that of the <= log
5998 line. There are two side-effects of this change:
6000 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6001 logged header lines no longer include the local Received: line, because
6002 it has not yet been created. The same applies to a copy of the message
6003 that is returned to a non-SMTP sender when a message is rejected.
6005 (b) When a filter file is tested using -bf, no additional Received: header
6006 is added to the test message. After some thought, I decided that this
6009 This change does not affect the value of $received_for. It is still set
6010 after address rewriting, but before local_scan() is called.
6012 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6014 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6015 gave an unhelpful panic error message, and a defer error. I have managed to
6016 change this behaviour so that it now rejects any supplied certificate,
6017 which seems right, as the list of acceptable certificates is empty.
6019 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6020 gave an unhelpful defer error. I have not managed to make this reject any
6021 supplied certificates, but the error message it gives is "no certificate
6022 supplied", which is not helpful.
6024 70. exigrep's output now also includes lines that are not associated with any
6025 message, but which match the given pattern. Implemented by a patch from
6026 Martin Sluka, which also tidied up the Perl a bit.
6028 71. Recipient callout verification, like sender verification, was using <> in
6029 the MAIL FROM command. This isn't really the right thing, since the actual
6030 sender may affect whether the remote host accepts the recipient or not. I
6031 have changed it to use the actual sender in the callout; this means that
6032 the cache record is now keyed on a recipient/sender pair, not just the
6033 recipient address. There doesn't seem to be a real danger of callout loops,
6034 since a callout by the remote host to check the sender would use <>.
6035 [SEE ABOVE: changed after hitting problems.]
6037 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6038 temporary errors. However, in the case of such a code being given after
6039 the end of a data transmission (i.e. after ".") Exim was failing to write
6040 a retry record for the message. (Yes, there was some broken host that was
6041 actually sending 8xx at this point.)
6043 73. An unknown lookup type in a host list could cause Exim to panic-die when
6044 the list was checked. (An example that provoked this was putting <; in the
6045 middle of a list instead of at the start.) If this happened during a DATA
6046 ACL check, a -D file could be left lying around. This kind of configuration
6047 error no longer causes Exim to die; instead it causes a defer error. The
6048 incident is still logged to the main and panic logs.
6050 74. Buglet left over from Exim 3 conversion. The message "too many messages
6051 in one connection" was written to the rejectlog but not the mainlog, except
6052 when address rewriting (yes!) was being logged.
6054 75. Added write_rejectlog option.
6056 76. When a system filter was run not as root (that is, when system_filter_user
6057 was set), the values of the $n variables were not being returned to the
6058 main process; thus, they were not subsequently available in the $sn
6061 77. Added +return_path_on_delivery log selector.
6063 78. A connection timeout was being treated differently from recipients deferred
6064 when testing hosts_max_try with a message that was older than the host's
6065 retry timeout. (The host should not be counted, thus allowing all hosts to
6066 be tried at least once before bouncing.) This may have been the cause of an
6067 occasionally reported bug whereby a message would remain on the queue
6068 longer than the retry timeout, but would be bounced if a delivery was
6069 forced. I say "may" because I never totally pinned down the problem;
6070 setting up timeout/retry tests is difficult. See also the next item.
6072 79. The ultimate address timeout was not being applied to errors that involved
6073 a combination of host plus message (for example, a timeout on a MAIL
6074 command). When an address resolved to a number of possible hosts, and they
6075 were not all tried for each delivery (e.g. because of hosts_max_try), a
6076 message could remain on the queue longer than the retry timeout.
6078 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6081 81. Fixed an obscure SMTP outgoing bug which required at least the following
6082 conditions: (a) there was another message waiting for the same server;
6083 (b) the server returned 5xx to all RCPT commands in the first message so
6084 that the message was not completed; (c) the server dropped the connection
6085 or gave a negative response to the RSET that Exim sends to abort the
6086 transaction. The observed case was a dropped connection after DATA that had
6087 been sent in pipelining mode. That is, the server had advertised PIPELINING
6088 but was not implementing it correctly. The effect of the bug was incorrect
6089 behaviour, such as trying another host, and this could lead to a crash.
6095 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6096 and daemon.c were passed as pointers to ints; they should have been
6097 pointers to socklen_t variables (which are typically unsigned ints).
6099 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6102 3. Fixed a really odd bug that affected only the testing scheme; patching a
6103 certain fixed string in the binary changed the value of another string that
6104 happened to be identical to the end of the original first string.
6106 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6107 name", it returns that address as the IP address. On some operating
6108 systems (e.g. Solaris), it also passes back the IP address string as the
6109 "host name". However, on others (e.g. Linux), it passes back an empty
6110 string. Exim wasn't checking for this, and was changing the host name to an
6111 empty string, assuming it had been canonicalized.
6113 5. Although rare, it is permitted to have more than one PTR record for a given
6114 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6115 all the names associated with an address, because they do in Solaris.
6116 However, it seems that they do not in Linux for data that comes from the
6117 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6118 I found this out when I moved to a new Linux workstation and tried to run
6119 the Exim test suite.
6121 To get round this problem I have changed the code so that it now does its
6122 own call to the DNS to look up PTR records when searching for a host name.
6123 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6124 addresses that are only in /etc/hosts are still found.
6126 This behaviour is, however, controlled by an option called host_lookup_
6127 order, which defaults to "bydns:byaddr". If people want to use the other
6128 order, or indeed, just use one or the other means of lookup, they can
6129 specify it in this variable.
6131 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6132 some operating systems, this comes back from gethostbyaddr() as an empty
6133 string, and this is what Exim used to test for. However, it seems that in
6134 other systems, "." is yielded. Exim now tests for this case too.
6136 7. The values of check_spool_space and check_log_space are now held internally
6137 as a number of kilobytes instead of an absolute number of bytes. If a
6138 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6139 kilobyte. This means that much larger values can be stored.
6141 8. Exim monitor: an attempt to get the action menu when not actually pointing
6142 at a message produces an empty menu entitled "No message selected". This
6143 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6144 no entries in it ("Shell widget menu has zero width and/or height"). So I
6145 have added a single, blank menu entry in this case.
6147 9. Added ${quote_local_part.
6149 10. MIME decoding is now applied to the contents of Subject: header lines when
6152 11. Now that a reference to $sender_host_address automatically causes a reverse
6153 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6154 host lookup before query-style lookups in lists that might use this
6155 variable. This has therefore been abolished, and the "net-" prefix is no
6156 longer necessary for query-style lookups.
6158 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6159 have been a typo for LFLAGS, so it has been changed.
6161 13. The install script calls Exim with "-C /dev/null" in order to find the
6162 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6163 to be output. However, since Exim outputs its version number before the
6164 error, it didn't break the script. It just looked ugly. I fixed this by
6165 always allowing "-C /dev/null" if the caller is root.
6167 14. Ignore overlarge ACL variable number when reading spool file - insurance
6168 against a later release with more variables having written the file.
6170 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6171 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6172 this, as well as the form without the "IPv6" on the front.
6174 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6175 OS/Makefile-Darwin file.
6177 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6178 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6180 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6181 variable. However, some OS use ints for inodes. I've added cast to long int
6182 to get rid of the compiler warning.
6184 19. I had forgotten to lock out "/../" in configuration file names when
6185 ALT_CONFIG_PREFIX was set.
6187 20. Routers used for verification do not need to specify transports. However,
6188 if such a router generated a host list, and callout was configured, Exim
6189 crashed, because it could not find a port number from the (non-existent)
6190 transport. It now assumes port 25 in this circumstance.
6192 21. Added the -t option to exigrep.
6194 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6195 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6196 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6197 them is set, LOOKUP_LSEARCH is forced.
6199 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6201 24. Added sender and host information to the "rejected by local_scan()" log
6202 line; previously there was no indication of these.
6204 25. Added .include_if_exists.
6206 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6207 Linux. It turns out that not all file systems support this. Apparently some
6208 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6209 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6210 not supported on the file descriptor, is now ignored when Exim is trying to
6211 sync a directory. This applies only to Linux.
6213 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6215 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6216 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6219 29. Added support for maildirsize files from supplied patch (modified a bit).
6221 30. The use of :fail: followed by an empty string could lead Exim to respond to
6222 sender verification failures with (e.g.):
6224 550 Verification failed for <xxx>
6225 550 Sender verify failed
6227 where the first response line was missing the '-' that indicates it is not
6228 the final line of the response.
6230 31. The loop for finding the name of the user that called Exim had a hardwired
6231 limit of 10; it now uses the value of finduser_retries, which is used for
6232 all other user lookups.
6234 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6237 33. Exim was neglecting to zero errno before one call of strtol() when
6238 expanding a string and expecting an integer value. On some systems this
6239 resulted in spurious "integer overflow" errors. Also, it was casting the
6240 result into an int without checking.
6242 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6243 did not work. The code looks as if it has *never* worked, though it appears
6244 to have been documented since at least release 1.62. I have made it work.
6246 35. The "timeout_DNS" error in retry rules, also documented since at least
6247 1.62, also never worked. As it isn't clear exactly what this means, and
6248 clearly it isn't a major issue, I have abolished the feature by treating it
6249 as "timeout", and writing a warning to the main and panic logs.
6251 36. The display of retry rules for -brt wasn't always showing the error code
6254 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6255 timeout_connect_A, timeout_connect_MX.
6257 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6258 to the empty sender.
6260 39. The daemon was not analysing the content of -oX till after it had closed
6261 stderr and disconnected from the controlling terminal. This meant that any
6262 syntax errors were only noted on the panic log, and the return code from
6263 the command was 0. By re-arranging the code a little, I've made the
6264 decoding happen first, so such errors now appear on stderr, and the return
6265 code is 1. However, the actual setting up of the sockets still happens in
6266 the disconnected process, so errors there are still only recorded on the
6269 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6270 connections (as happens on some IP stacks) was logged at start up time as
6271 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6272 it from "IPv6 and IPv4", which means that two separate sockets are being
6275 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6276 says whether AF_INET or AF_INET6 was passed as an argument.
6278 42. Exiwhat output was messed up when time zones were included in log
6281 43. Exiwhat now gives more information about the daemon's listening ports,
6282 and whether -tls-on-connect was used.
6284 44. The "port" option of the smtp transport is now expanded.
6286 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6287 silently ignored. Now an error message is written to the main and panic
6290 46. There's a new ACL modifier called "logwrite" which writes to a log file
6291 as soon as it is encountered.
6293 47. Added $local_user_uid and $local_user_gid at routing time.
6295 48. Exim crashed when trying to verify a sender address that was being
6298 49. Exim was recognizing only a space character after ".include". It now also
6299 recognizes a tab character.
6301 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6302 extracting the relevant information from the specification. The man page no
6303 longer contains scrambled data for the -d option, and I've added a section
6304 at the front about calling Exim under different names.
6306 51. Added "extra_headers" argument to the "mail" command in filter files.
6308 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6311 53. Installed eximstats 1.29.
6313 54. Added transport_filter_timeout as a generic transport option.
6315 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6316 Cc: header lines. This was required by RFC 822, but it not required by RFC
6319 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6320 incoming messages that did not have them. Now it does so only if the
6321 message originates locally, that is, if there is no associated remote host
6322 address. When Resent- header lines are present, this applies to the Resent-
6323 lines rather than the non-Resent- lines.
6325 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6326 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6328 58. Messages for configuration errors now include the name of the main
6329 configuration file - useful now that there may be more than one file in a
6330 list (.included file names were always shown).
6332 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6333 for those rare installations that do not start the daemon as root or run it
6334 setuid root. I've cut out the call to initgroups() if the daemon is not
6337 60. The Exim user and group can now be bound into the binary as text strings
6338 that are looked up at the start of Exim's processing.
6340 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6342 62. Added $mailstore_basename variable.
6344 63. Installed patch to sieve.c from Michael Haardt.
6346 64. When Exim failed to open the panic log after failing to open the main log,
6347 the original message it was trying to log was written to stderr and debug
6348 output, but if they were not available (the usual case in production), it
6349 was lost. Now it is written to syslog before the two lines that record the
6350 failures to open the logs.
6352 65. Users' Exim filters run in subprocesses under the user's uid. It is
6353 possible for a "deliver" command or an alias in a "personal" command to
6354 provoke an address rewrite. If logging of address rewriting is configured,
6355 this fails because the process is not running as root or exim. There may be
6356 a better way of dealing with this, but for the moment (because 4.30 needs
6357 to be released), I have disabled address rewrite logging when running a
6358 filter in a non-root, non-exim process.
6364 1. The buildconfig auxiliary program wasn't quoting the value set for
6365 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6366 not defined. This bug was masked in 4.22 by the effect that was fixed in
6369 2. Some messages that were rejected after a message id was allocated were
6370 shown as "incomplete" by exigrep. It no longer does this for messages that
6371 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6373 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6374 have allow_domain_literals set, the ID did not get logged in the <= line.
6375 Domain literals are now always recognized in Message-ID: header lines.
6377 4. The first argument for a ${extract expansion item is the key name or field
6378 number. Leading and trailing spaces in this item were not being ignored,
6379 causing some misleading effects.
6381 5. When deliver_drop_privilege was set, single queue runner processes started
6382 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6383 same command in the process it spins off) were not dropping privilege.
6385 6. When the daemon running as "exim" started a queue runner, it always
6386 re-executed Exim in the spun-off process. This is a waste of effort when
6387 deliver_drop_privilege is set. The new process now just calls the
6388 queue-runner function directly.
6394 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6397 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6398 ignored. Though the use of -oP was forcing the writing of a pid file, it
6399 was always written to the default place.
6401 3. If the message "no IP address found for host xxxx" is generated during
6402 incoming verification, it is now followed by identification of the incoming
6403 connection (so you can more easily find what provoked it).
6405 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6407 5. Added some features to "harden" Exim a bit more against certain attacks:
6409 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6410 be put in Local/Makefile. This is like the never_users runtime option,
6411 but it cannot be overridden. The default setting is "root".
6413 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6414 prefix string with which any file named in a -C command line option
6417 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6418 is retained for -C and -D only if the caller of Exim is root. Without
6419 it, the exim user may also use -C and -D and retain privilege.
6421 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6422 command line option is disabled.
6424 6. Macro names set by the -D option must start with an upper case letter, just
6425 like macro names defined in the configuration file.
6427 7. Added "dereference=" facility to LDAP.
6429 8. Two instances of the typo "uknown" in the source files are fixed.
6431 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6432 the Configure-Makefile script screwed up while processing it.
6434 10. Incorporated PCRE 4.4.
6436 11. The SMTP synchronization check was not operating right at the start of an
6437 SMTP session. For example, it could not catch a HELO sent before the client
6438 waited for the greeting. There is now a check for outstanding input at the
6439 point when the greeting is written. Because of the duplex, asynchronous
6440 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6441 way, but not yet received, when the check is performed.
6443 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6444 on TCP/IP sockets, because this apparently causes some broken clients to
6447 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6448 unchanged) from the Cygwin maintainer.
6450 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6451 was supported instead of testing for mbx. Effectively a typo.
6453 15. The spa authenticator server code was not checking that the input it
6454 received was valid base64.
6456 16. The debug output line for the "set" modifier in ACLs was not showing the
6457 name of the variable that was being set.
6459 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6461 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6462 lookup on its own. Something else was needed to trigger the lookup. For
6463 example, a match in host_lookup or the need for a host name in a host list.
6464 Now, if $sender_host_name is referenced and the host name has not yet been
6465 looked up, a lookup is performed. If the lookup fails, the variable remains
6466 empty, and $host_lookup_failed is set to "1".
6468 19. Added "eqi" as a case-independent comparison operator.
6470 20. The saslauthd authentication condition could segfault if neither service
6471 nor realm was specified.
6473 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6474 error message that was logged was misleading, and incoming SMTP
6475 connections were dropped. The message is now more accurate, and temporary
6476 errors are given to SMTP connections.
6478 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6479 (except RSET) with a 421 error, until QUIT is received. However, it was
6480 failing to send a response to QUIT.
6482 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6483 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6484 ACL is now run after the helo_try_verify_hosts code.
6486 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6487 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6488 case-independent, so other case variants are also recognized). Apparently
6489 some systems use these upper case variants.
6491 25. If more than two messages were waiting for the same host, and a transport
6492 filter was specified for the transport, Exim sent two messages over the
6493 same TCP/IP connection, and then failed with "socket operation on non-
6494 socket" when it tried to send the third.
6496 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6498 27. The extern definition of crypt16() in expand.c was not being excluded when
6499 the OS had its own crypt16() function.
6501 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6502 as a preferred synonym for return_size_limit, both as an option and as an
6505 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6507 30. Changed the default configuration ACL to relax the local part checking rule
6508 for addresses that are not in any local domains. For these addresses,
6509 slashes and pipe symbols are allowed within local parts, but the sequence
6510 /../ is explicitly forbidden.
6512 31. SPA server authentication was not clearing the challenge buffer before
6515 32. log_message in a "warn" ACL statement was writing to the reject log as
6516 well as to the main log, which contradicts the documentation and doesn't
6517 seem right (because no rejection is happening). So I have stopped it.
6519 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6520 However, I am unable to do any testing of this.
6522 34. Fixed an infelicity in the appendfile transport. When checking directories
6523 for a mailbox, to see if any needed to be created, it was accidentally
6524 using path names with one or more superfluous leading slashes; tracing
6525 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6527 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6528 contents are added to the log line that is written for every discarded
6529 recipient. (Previously a log_message setting was ignored.)
6531 36. The ${quote: operator now quotes the string if it is empty.
6533 37. The install script runs exim in order to find its version number. If for
6534 some reason other than non-existence or emptiness, which it checks, it
6535 could not run './exim', it was installing it with an empty version number,
6536 i.e. as "exim-". This error state is now caught, and the installation is
6539 38. An argument was missing from the function that creates an error message
6540 when Exim fails to connect to the socket for saslauthd authentication.
6541 This could cause Exim to crash, or give a corrupted message.
6543 39. Added isip, isip4, and isip6 to ${if conditions.
6545 40. The ACL variables $acl_xx are now saved with the message, and can be
6546 accessed later in routers, transports, and filters.
6548 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6549 strings in the file are not string-expanded.
6551 42. If a MAIL command specified a SIZE value that was too large to fit into an
6552 int variable, the check against message_size_limit failed. Such values are
6553 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6554 day this will have to be increased, but I don't think I want to be around
6555 when emails are that large.
6562 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6563 iconv() is not standard in FreeBSD.
6565 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6566 IPv6 enabled. The observed symptom was a segmentation fault on return from
6567 the function os_common_find_running_interfaces() in src/os.c.
6569 3. In the check_special_case() function in daemon.c I had used "errno" as an
6570 argument name, which causes warnings on some systems. This was basically a
6571 typo, since it was named "eno" in the comments!
6573 4. The code that waits for the clock to tick (at a resolution of some fraction
6574 of a second) so as to ensure message-id uniqueness was always waiting for
6575 at least one whole tick, when it could have waited for less. [This is
6576 almost certainly not relevant at current processor speeds, where it is
6577 unlikely to ever wait at all. But we try to future-proof.]
6579 5. The function that sleeps for a time interval that includes fractions of a
6580 second contained a race. It did not block SIGALRM between setting the
6581 timer, and suspending (a couple of lines later). If the interval was short
6582 and the sigsuspend() was delayed until after it had expired, the suspension
6583 never ended. On busy systems this could lead to processes getting stuck for
6586 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6587 process, before it forks any delivery processes. The open lookup caching
6588 mechanism meant that the open file or database connection was passed into
6589 the delivery process. The problem was that delivery processes always tidy
6590 up cached lookup data. This could cause a problem for the next delivery
6591 process started by the queue runner, because the external queue runner
6592 process does not know about the closure. So the next delivery process
6593 still has data in the lookup cache. In the case of a file lookup, there was
6594 no problem because closing a file descriptor in a subprocess doesn't affect
6595 the parent. However, if the lookup was caching a connection to a database,
6596 the connection was closed, and the second delivery process was likely to
6597 see errors such as "PGSQL: query failed: server closed the connection
6598 unexpectedly". The problem has been fixed by closing all cached lookups
6599 in a queue runner before running a delivery process.
6601 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6602 seem to have the "const" qualifier which it has on other OS. I've
6605 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6606 *of the same type* (client or server) with the same public name that an
6607 error should be diagnosed.
6609 9. When Exim looked up a host name for an IP address, but failed to find the
6610 original IP address when looking up the host name (a safety check), it
6611 output the message "<ip address> does not match any IP for NULL", which was
6612 confusing, to say the least. The bug was that the host name should have
6613 appeared instead of "NULL".
6615 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6616 user that is built into the binary, and the -C or -D options is used, root
6617 privilege is dropped before the configuration file is read. In addition,
6618 logging is switched to stderr instead of the normal log files. If the
6619 configuration then re-defines the Exim user, the unprivileged environment
6620 is probably not what is expected, so Exim logs a panic warning message (but
6623 However, if deliver_drop_privilege is set, the unprivileged state may well
6624 be exactly what is intended, so the warning has been cut out in that case,
6625 and Exim is allowed to try to write to its normal log files.
6631 1. smtp_return_error_details was not giving details for temporary sender
6632 or receiver verification errors.
6634 2. Diagnose a configuration error if two authenticators have the same public
6637 3. Exim used not to create the message log file for a message until the first
6638 delivery attempt. This could be confusing when incoming messages were held
6639 for policy or load reasons. The message log file is now created at the time
6640 the message is received, and an initial "Received" line is written to it.
6642 4. The automatically generated man page for command line options had a minor
6643 bug that caused no ill effects; however, a more serious problem was that
6644 the procedure for building the man page automatically didn't always
6645 operate. Consequently, release 4.20 contains an out-of-date version. This
6646 shouldn't happen again.
6648 5. When building Exim with embedded Perl support, the script that builds the
6649 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6650 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6652 6. The freeze_tell option was not being used for messages that were frozen on
6653 arrival, either by an ACL or by local_scan().
6655 7. Added the smtp_incomplete_transaction log selector.
6657 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6658 was accepting AUTH without a new EHLO.
6660 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6661 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6664 10. Logging of TCP/IP connections (when configured) now happens in the main
6665 daemon process instead of the child process, so that the TCP/IP connection
6666 count is more accurate (but it can never be perfect).
6668 11. The use of "drop" in a nested ACL was not being handled correctly in the
6669 outer ACL. Now, if condition failure induced by the nested "drop" causes
6670 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6671 or "require"), the connection is dropped.
6673 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6674 that yield "discard" can now be used with an "accept" or a "discard" verb,
6675 but an error is generated for any others (because I can't see a useful way
6676 to define what should happen).
6678 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6679 are now processed in the same way as lines in the Exim configuration file.
6680 In particular, continuation lines are supported.
6682 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6684 15. Added -ti meaning -t -i.
6686 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6687 domains, and warn by logging if others are found.
6689 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6690 returned from GIFCONF when Exim is trying to find the list of interfaces on
6691 a host. The code in os.c has been modified to copy each ifreq to an aligned
6692 structure in all cases.
6694 Also, in some cases, the returned ifreq's were being copied to a 'struct
6695 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6696 means the last couple of bytes of an IPv6 address could be chopped if the
6697 ifreq contained only a normal sockaddr (14 bytes storage).
6699 18. Named domain lists were not supported in the hosts_treat_as_local option.
6700 An entry such as +xxxx was not recognized, and was treated as a literal
6703 19. Ensure that header lines added by a DATA ACL are included in the reject log
6704 if the ACL subsequently rejects the message.
6706 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6707 MD5 (which is deprecated).
6709 21. When testing a filter file using -bf, Exim was writing a message when it
6710 took the sender from a "From " line in the message, but it was not doing so
6711 when it took $return_path from a Return-Path: header line. It now does.
6713 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6714 with a valid header line field name (a series of printing characters
6715 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6717 23. Changed "disc" in the source to "disk" to conform to the documentation and
6718 the book and for uniformity.
6720 24. Ignore Sendmail's -Ooption=value command line item.
6722 25. When execve() failed while trying to run a command in a pipe transport,
6723 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6724 could be confused with a return value of 69 from the command itself. This
6725 has been changed to 127, the value the shell returns if it is asked to run
6726 a non-existent command. The wording for the related log line suggests a
6727 non-existent command as the problem.
6729 26. If received_header_text expands to an empty string, do not add a Received:
6730 header line to the message. (Well, it adds a token one on the spool, but
6731 marks it "old" so that it doesn't get used or transmitted.)
6733 27. Installed eximstats 1.28 (addition of -nt option).
6735 28. There was no check for failure on the call to getsockname() in the daemon
6736 code. This can fail if there is a shortage of resources on the system, with
6737 ENOMEM, for example. A temporary error is now given on failure.
6739 29. Contrary to the C standard, it seems that in some environments, the
6740 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6741 program. Exim now does this explicitly; it affects the formatting of
6742 timestamps using strftime().
6744 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6745 complaints. I've now initialized all the variables, to avoid this.
6747 32. Header lines added by a system filter were not being "seen" during
6748 transport-time rewrites.
6750 33. The info_callback() function passed to OpenSSL is set up with type void
6751 (*)(SSL *, int, int), as described somewhere. However, when calling the
6752 function (actually a macro) that sets it up, the type void(*)() is
6753 expected. I've put in a cast to prevent warnings from picky compilers.
6755 34. If a DNS black list lookup found a CNAME record, but there were no A
6756 records associated with the domain it pointed at, Exim crashed.
6758 35. If a DNS black list lookup returned more than one A record, Exim ignored
6759 all but the first. It now scans all returned addresses if a particular IP
6760 value is being sought. In this situation, the contents of the
6761 $dnslist_value variable are a list of all the addresses, separated by a
6764 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6765 to accept a host name and all its aliases if the forward lookup for any of
6766 them yielded the IP address of the incoming connection. Now it accepts only
6767 those names whose forward lookup yields the correct IP address. Any other
6768 names are discarded. This closes a loophole whereby a rogue DNS
6769 administrator could create reverse DNS records to break through a
6770 wildcarded host restriction in an ACL.
6772 37. If a user filter or a system filter that ran in a subprocess used any of
6773 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6774 the wrong values were passed to the pipe command ($thisaddress had the
6775 value of $0, $0 had the value of $1, etc). This bug was introduced by
6776 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6778 38. Improved the line breaking for long SMTP error messages from ACLs.
6779 Previously, if there was no break point between 40 and 75 characters, Exim
6780 left the rest of the message alone. Two changes have been made: (a) I've
6781 reduced the minimum length to 35 characters; (b) if it can't find a break
6782 point between 35 and 75 characters, it looks ahead and uses the first one
6783 that it finds. This may give the occasional overlong line, but at least the
6784 remaining text gets split now.
6786 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6787 file descriptors might be low, and that setting 1000 would always raise it.
6788 It turns out that in some environments, the limit is already over 1000 and
6789 that lowering it causes trouble. So now Exim takes care not to decrease it.
6791 40. When delivering a message, the value of $return_path is set to $sender_
6792 address at the start of routing (routers may change the value). By an
6793 oversight, this default was not being set up when an address was tested by
6794 -bt or -bv, which affected the outcome if any router or filter referred to
6797 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6798 log only when "message" or "log_message" are set. However, if one of the
6799 conditions was an address verification, or a call to a nested ACL, the
6800 messages generated by the underlying test were being passed through. This
6801 no longer happens. The underlying message is available in $acl_verify_
6802 message for both "message" and "log_message" expansions, so it can be
6803 passed through if needed.
6805 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6806 new expansion $bh_ to give the encoded byte string without charset
6807 translation. Translation happens only if iconv() is available; HAVE_ICONV
6808 indicates this at build time. HEADERS_CHARSET gives the charset to
6809 translate to; headers_charset can change it in the configuration, and
6810 "headers charset" can change it in an individual filter file.
6812 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6813 that creates RFC 2047 encoded "words" labels them as that charset instead
6814 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6815 expansion operator; (ii) when Exim creates a From: line for a local
6816 message; (iii) when a header line is rewritten to include a "phrase" part.
6818 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6819 buggy, causing it to skip the first lines of messages whose message ID
6820 ended in 'D'. This would not have bitten before Exim release 4.14, because
6821 message IDs were unlikely to end in 'D' before then. The effect was to have
6822 incorrect size information for certain domains.
6824 45. #include "config.h" was missing at the start of the crypt16.c module. This
6825 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6828 46. If there was a timeout during a "random" callout check, Exim treated it as
6829 a failure of the random address, and carried on sending RSET and the real
6830 address. If the delay was just some slowness somewhere, the response to the
6831 original RCPT would be taken as a response to RSET and so on, causing
6832 mayhem of various kinds.
6834 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6835 when I implemented it. It didn't allow for the fact that some option values
6836 may legitimately be negative (e.g. size_addition), and it didn't even do
6837 the right test for positive values.
6839 48. Domain names in DNS records are case-independent. Exim always looks them up
6840 in lower case. Some resolvers return domain names in exactly the case they
6841 appear in the zone file, that is, they may contain uppercase letters. Not
6842 all resolvers do this - some return always lower case. Exim was treating a
6843 change of case by a resolver as a change of domain, similar to a widening
6844 of a domain abbreviation. This triggered its re-routing code and so it was
6845 trying to route what was effectively the same domain again. This normally
6846 caused routing to fail (because the router wouldn't handle the domain
6847 twice). Now Exim checks for this case specially, and just changes the
6848 casing of the domain that it ultimately uses when it transmits the message
6851 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6854 50. If a filter generated a file delivery with a non-absolute name (possible if
6855 no home directory exists for the router), the forbid_file option was not
6858 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6859 the existing equality matching.
6861 52. Exim was using ints instead of ino_t variables in some places where it was
6862 dealing with inode numbers.
6864 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6865 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6866 called TMPDIR, and if it finds it is different, it changes its value.
6868 54. The smtp_printf() function is now made available to local_scan() so
6869 additional output lines can be written before returning. There is also an
6870 smtp_fflush() function to enable the detection of a dropped connection.
6871 The variables smtp_input and smtp_batched_input are exported to
6874 55. Changed the default runtime configuration: the message "Unknown user"
6875 has been removed from the ACL, and instead placed on the localuser router,
6876 using the cannot_route_message feature. This means that any verification
6877 failures that generate their own messages won't get overridden. Similarly,
6878 the "Unrouteable address" message that was in the ACL for unverifiable
6879 relay addresses has also been removed.
6881 56. Added hosts_avoid_esmtp to the smtp transport.
6883 57. The exicyclog script was not checking for the esoteric option
6884 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6885 will work only if exicyclog is run under the appropriate euid.
6887 58. Following a discussion on the list, the rules by which Exim recognises line
6888 endings on incoming messages have been changed. The -dropcr and drop_cr
6889 options are now no-ops, retained only for backwards compatibility. The
6890 following line terminators are recognized: LF CRLF CR. However, special
6891 processing applies to CR:
6893 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6894 nor a local message in the state where . is a terminator.
6896 (ii) If a bare CR is encountered in a header line, an extra space is added
6897 after the line terminator so as not to end the header. The reasoning
6898 behind this is that bare CRs in header lines are most likely either
6899 to be mistakes, or people trying to play silly games.
6901 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6902 was being incorrectly given as 18 bytes larger than it should have been.
6903 This is a VOB (very old bug).
6905 60. This may never have affected anything current, but just in case it has:
6906 When the local host is found other than at the start of a list of hosts,
6907 the local host, those with the same MX, and any that follow, are discarded.
6908 When the list in question was part of a longer list of hosts, the following
6909 hosts (not currently being processed) were also being discarded. This no
6910 longer happens. I'm not sure if this situation could ever has previously
6913 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6916 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6917 header. This is something that is recommended in a new Internet Draft, and
6918 is something that is documented as being done by Sendmail. There are two
6919 possible values. For messages generated by the autoreply transport, Exim
6922 Auto-Submitted: auto-replied
6924 whereas for all other generated messages (e.g. bounces) it adds
6926 Auto-Submitted: auto-generated
6928 63. The "personal" condition in filters now includes a test for the
6929 Auto-Submitted: header. If it contains the string "auto-" the message it
6930 not considered personal.
6932 64. Added rcpt_include_affixes as a generic transport option.
6934 65. Added queue_only_override (default true).
6936 66. Added the syslog_duplication option.
6938 67. If what should have been the first header line of a message consisted of
6939 a space followed by a colon, Exim was mis-interpreting it as a header line.
6940 It isn't of course - it is syntactically invalid and should therefore be
6941 treated as the start of the message body. The misbehaviour could have
6942 caused a number of strange effects, including loss of data in subsequent
6943 header lines, and spool format errors.
6945 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6946 client host had authenticated. This control can now be exercised by an ACL
6947 for more flexibility.
6949 69. By default, callouts do not happen when testing with -bh. There is now a
6950 variant, -bhc, which does actually run the callout code, including
6951 consulting and updating the callout cache.
6953 70. Added support for saslauthd authentication, courtesy of Alexander
6956 71. If statvfs() failed on the spool or log directories while checking their
6957 size for availability, Exim confusingly gave the error "space shortage".
6958 Furthermore, in debugging mode it crashed with a floating point exception.
6959 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6960 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6961 serious problem, Exim now writes to the main and panic logs when this
6962 happens, with details of the failure. It then refuses to accept the
6963 incoming message, giving the message "spool directory problem" or "log
6964 directory problem" with a 421 code for SMTP messages.
6966 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6967 0, 1, and 2 exist, because some OS complain for execs without them (see
6968 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6969 descriptors. However, the code omitted to check that the open succeeded,
6970 causing mysterious errors if for some reason the permissions on /dev/null
6971 got screwed. Now Exim writes a message to the main and panic logs, and
6972 bombs out if it can't open /dev/null.
6974 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6975 interact so that it is all more flexible. It is supposed to remain
6976 backwards compatible. Also added extra_local_interfaces.
6978 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6979 to bomb out with an assertion failure - to the client this appears as a
6980 connection drop. This problem occurs in the part of the code that was taken
6981 from the Samba project. Fortunately, the assertion is in a very simple
6982 function, so I have fixed this by reproducing the function inline in the
6983 one place where it is called, and arranging for authentication to fail
6984 instead of killing the process with assert().
6986 75. The SPA client code was not working when the server requested OEM rather
6987 than Unicode encoding.
6989 76. Added code to make require_files with a specific uid setting more usable in
6990 the case where statting the file as root fails - usually a non-root-mounted
6991 NFS file system. When this happens and the failure is EACCES, Exim now
6992 forks a subprocess and does the per-uid checking as the relevant uid.
6994 77. Added process_log_path.
6996 78. If log_file_path was not explicitly set, a setting of check_log_space or
6997 check_log_inodes was ignored.
6999 79. If a space check for the spool or log partitions fails, the incident is now
7000 logged. Of course, in the latter case the data may get lost...
7002 80. Added the %p formatting code to string_format() so that it can be used to
7003 print addresses in debug_print(). Adjusted all the address printing in the
7004 debugging in store.c to use %p rather than %d.
7006 81. There was a concern that a line of code in smtp_in.c could overflow a
7007 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7008 initially expressed, the concern was not well-founded, because trailing
7009 spaces are removed early. However, if the trailing spaces were followed by
7010 a NULL, they did not get removed, so the overflow was possible. Two fixes
7013 (a) I re-wrote the offending code in a cleaner fashion.
7014 (b) If an incoming SMTP command contains a NULL character, it is rejected
7017 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7018 runs initgroups(), so that if the Exim user is in any additional groups,
7019 they will be used during message reception.
7025 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob