1 Change log file for Exim from version 4.21
2 ------------------------------------------
3 This document describes *changes* to previous versions, that might
4 affect Exim's operation, with an unchanged configuration file. For new
5 options, and new features, see the NewStuff file next to this ChangeLog.
11 JH/01 Replace the store_release() internal interface with store_newblock(),
12 which internalises the check required to safely use the old one, plus
13 the allocate and data copy operations duplicated in both (!) of the
16 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
17 modifier. This matches the restriction on the commandline.
19 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
20 Previously only the last row was returned.
22 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
23 we assumed that tags in the header were well-formed, and parsed the
24 element content after inspecting only the first char of the tag.
25 Assumptions at that stage could crash the receive process on malformed
28 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
29 While running the DKIM ACL we operate on the Permanent memory pool so that
30 variables created with "set" persist to the DATA ACL. Also (at any time)
31 DNS lookups that fail create cache records using the Permanent pool. But
32 expansions release any allocations made on the current pool - so a dnsdb
33 lookup expansion done in the DKIM ACL releases the memory used for the
34 DNS negative-cache, and bad things result. Solution is to switch to the
35 Main pool for expansions.
36 While we're in that code, add checks on the DNS cache during store_reset,
37 active in the testsuite.
38 Problem spotted, and debugging aided, by Wolfgang Breyha.
40 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
41 When none of the hosts presented to a transport match an already-open
42 connection, close it and proceed with the list. Previously we would
43 queue the message. Spotted by Lena with Yahoo, probably involving
46 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
47 Previously a spurious "250 OK id=" response was appended to the proper
50 JH/08 The "support for" informational output now, which built with Content
51 Scanning support, has a line for the malware scanner interfaces compiled
52 in. Interface can be individually included or not at build time.
54 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
55 by the template makefile "src/EDITME". The "STREAM" support for an older
56 ClamAV interface method is removed.
58 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
59 rows affected is given instead).
61 JH/11 The runtime Berkeley DB library version is now additionally output by
62 "exim -d -bV". Previously only the compile-time version was shown.
64 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
65 SMTP connection. Previously, when one had more receipients than the
66 first, an abortive onward connection was made. Move to full support for
67 multiple onward connections in sequence, handling cutthrough connection
68 for all multi-message initiating connections.
70 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
71 routers. Previously, a multi-recipient message would fail to match the
72 onward-connection opened for the first recipient, and cause its closure.
74 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
75 a timeout on read on a GnuTLS initiating connection, resulting in the
76 initiating connection being dropped. This mattered most when the callout
77 was marked defer_ok. Fix to keep the two timeout-detection methods
80 JH/15 Relax results from ACL control request to enable cutthrough, in
81 unsupported situations, from error to silently (except under debug)
82 ignoring. This covers use with PRDR, frozen messages, queue-only and
89 JH/01 Rework error string handling in TLS interface so that the caller in
90 more cases is responsible for logging. This permits library-sourced
91 string to be attached to addresses during delivery, and collapses
92 pairs of long lines into single ones.
94 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
95 during configuration. Wildcards are allowed and expanded.
97 JH/02 Rework error string handling in DKIM to pass more info back to callers.
98 This permits better logging.
100 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
101 do not close it down and have the child transport start it up again on
102 the passed-on TCP connection. Instead, proxy the child (and any
103 subsequent ones) for TLS via a unix-domain socket channel. Logging is
104 affected: the continued delivery log lines do not have any DNSSEC, TLS
105 Certificate or OCSP information. TLS cipher information is still logged.
107 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
108 identical IP addresses on different listening ports. Will also affect
111 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
112 add noisy ifdef guards to special-case this sillyness.
113 Patch from Bernd Kuhls.
115 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
116 than 255 are no longer allowed.
118 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
119 Disable the session-cache too, which might reduce our load. Since we
120 currrectly use a new context for every connection, both as server and
121 client, there is no benefit for these.
122 GnuTLS appears to not support tickets server-side by default (we don't
123 call gnutls_session_ticket_enable_server()) but client side is enabled
124 by default on recent versions (3.1.3 +) unless the PFS priority string
127 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
128 <https://reproducible-builds.org/specs/source-date-epoch/>.
130 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
131 the check for any unsuccessful recipients did not notice the limit, and
132 erroneously found still-pending ones.
134 JH/08 Pipeline CHUNKING command and data together, on kernels that support
135 MSG_MORE. Only in-clear (not on TLS connections).
137 JH/09 Avoid using a temporary file during transport using dkim. Unless a
138 transport-filter is involved we can buffer the headers in memory for
139 creating the signature, and read the spool data file once for the
140 signature and again for transmission.
142 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
143 4.77 as the kernel support then wasn't solid, having issues in 64bit
144 mode. Now, it's been long enough. Add support for FreeBSD also.
146 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
147 case where the routing stage had gathered several addresses to send to
148 a host before calling the transport for the first, we previously failed
149 to close down TLS in the old transport process before passing the TCP
150 connection to the new process. The new one sent a STARTTLS command
151 which naturally failed, giving a failed delivery and bloating the retry
152 database. Investigation and fix prototype from Wolfgang Breyha.
154 JH/12 Fix check on SMTP command input synchronisation. Previously there were
155 false-negatives in the check that the sender had not preempted a response
156 or prompt from Exim (running as a server), due to that code's lack of
157 awareness of the SMTP input buffering.
159 PP/04 Add commandline_checks_require_admin option.
160 Exim drops privileges sanely, various checks such as -be aren't a
161 security problem, as long as you trust local users with access to their
162 own account. When invoked by services which pass untrusted data to
163 Exim, this might be an issue. Set this option in main configuration
164 AND make fixes to the calling application, such as using `--` to stop
167 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
168 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
169 responses to those, into a single TLS record each way (this usually means
170 a single packet). As a side issue, smtp_enforce_sync now works on TLS
173 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
174 affects you only if you're dancing at the edge of the param size limits.
175 If you are, and this message makes sense to you, then: raise the
176 configured limit or use OpenSSL 1.1. Nothing we can do for older
179 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
180 cmdline element to get the documented default one. Previously it was
183 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
184 are unsuable for non-message-specific reasons - eg. connection timeout,
185 banner-time rejection.
187 JH/16 Fix logging of delivery remote port, when specified by router, under
190 PP/06 Repair manualroute's ability to take options in any order, even if one
191 is the name of a transport.
194 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
196 JH/17 Change the list-building routines interface to use the expanding-string
197 triplet model, for better allocation and copying behaviour.
199 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
200 Previously it was constructed the first time a possibly-matching string
201 was met in the configuration file input during startup; now it is done
204 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
205 compatible one, to avoid the (poorly documented) possibility of a config
206 file in the working directory redirecting the DB files, possibly correpting
207 some existing file. CVE-2017-10140 assigned for BDB.
209 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
210 cache-hot. Previously, although the result was properly cached, the
211 initial verify call returned a defer.
213 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
214 the main verify for receipient in uncached-mode.
216 JH/22 Retire historical build files to an "unsupported" subdir. These are
217 defined as "ones for which we have no current evidence of testing".
219 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
220 if present. Previously it was ignored.
222 JH/24 Start using specified-initialisers in C structure init coding. This is
223 a C99 feature (it's 2017, so now considered safe).
225 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
226 if was a fixed-sized field and bitmask ops via macros; it is now more
229 PP/07 GitHub PR 56: Apply MariaDB build fix.
230 Patch provided by Jaroslav Škarvada.
232 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
233 during Coverity cleanups [4.87 JH/47]
234 Diagnosis and fix provided by Michael Fischer v. Mollard.
236 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
237 the right size to place the terminating semicolon on its own folded
238 line, the header hash was calculated to an incorrect value thanks to
239 the (relaxed) space the fold became.
241 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
242 and confused the parent.
244 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
245 which could crash as a result. This could lead to undeliverable messages.
247 JH/28 Logging: "next input sent too soon" now shows where input was truncated
250 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
251 matters on fast-turnover and PID-randomising systems, which were getting
252 out-of-order delivery.
254 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
255 a possibly-overlapping copy. The symptom was that "Remote host closed
256 connection in response to HELO" was logged instead of the actual 4xx
259 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
260 Previously only that bufferd was discarded, resulting in SYMTP command
263 JH/32 DKIM: when a message has multiple signatures matching an identity given
264 in dkim_verify_signers, run the dkim acl once for each. Previously only
265 one run was done. Bug 2189.
267 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
268 "panic the current process" to "deliberately defer". The panic log is
269 still written with the problem list name; the mail and reject logs now
270 get a temp-reject line for the message that was being handled, saying
271 something like "domains check lookup or other defer". The SMTP 451
272 message is still "Temporary local problem".
274 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
275 A crafted sequence of BDAT commands could result in in-use memory beeing
276 freed. CVE-2017-16943.
278 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
279 from SMTP input. Previously it was always done; now only done for DATA
280 and not BDAT commands. CVE-2017-16944.
282 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
283 to the message (such as an overlong header line). Previously this was
284 not done and we did not exit BDAT mode. Followon from the previous item
285 though a different problem.
291 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
292 than -2003 did; needs libidn2 in addition to libidn.
294 JH/02 The path option on a pipe transport is now expanded before use.
296 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
297 Patch provided by "Björn", documentation fix added too.
299 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
300 missing a wire-to-host endian conversion.
302 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
303 close after a BDAT command line could be taken as a following command,
304 giving a synch failure. Fix by only checking for synch immediately
305 before acknowledging the chunk.
307 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
308 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
309 macro. Patches provided by Josh Soref.
311 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
312 Previously we did not; the RFC seems ambiguous and VRFY is not listed
313 by IANA as a service extension. However, John Klensin suggests that we
316 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
317 the dkim code may be unix-mode line endings rather than smtp wire-format
318 CRLF, so prepend a CR to any bare LF.
320 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
321 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
323 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
324 the first were themselves being wrongly included in the feed into dkim
325 processing; with most chunk sizes in use this resulted in an incorrect
326 body hash calculated value.
328 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
329 DKIM signature block, for verification. Although advised against by
330 standards it is specifically not ruled illegal.
332 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
334 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
335 missing a body hash (the bh= tag).
337 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
338 It seems that HAProxy sends the Proxy Protocol information in clear and
339 only then does a TLS startup, so do the same.
341 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
342 TCP connections (such as for Spamd) unless the daemon successfully set
343 Fast Open mode on its listening sockets. This fixes breakage seen on
344 too-old kernels or those not configured for Fast Open, at the cost of
345 requiring both directions being enabled for TFO, and TFO never being used
346 by non-daemon-related Exim processes.
348 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
349 endings, at least on the first header line. Try to canonify any that get
350 past that check, despite the cost.
352 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
353 now limited to an arbitrary five deep, while parsing addresses with the
354 strip_excess_angle_brackets option enabled.
356 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
357 instead leave the unprompted TLS handshake in socket buffer for the
358 TLS library to consume.
360 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
362 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
364 JH/16 Drop variables when they go out of scope. Memory management drops a whole
365 region in one operation, for speed, and this leaves assigned pointers
366 dangling. Add checks run only under the testsuite which checks all
367 variables at a store-reset and panics on a dangling pointer; add code
368 explicitly nulling out all the variables discovered. Fixes one known
369 bug: a transport crash, where a dangling pointer for $sending_ip_address
370 originally assigned in a verify callout, is re-used.
372 PP/06 Drop '.' from @INC in various Perl scripts.
374 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
376 PP/08 Reduce a number of compilation warnings under clang; building with
377 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
378 should be warning-free.
380 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
382 HS/01 Fix portability problems introduced by PP/08 for platforms where
383 realloc(NULL) is not equivalent to malloc() [SunOS et al].
385 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
386 chunk. This allows us to accept broken chunked messages. We need a more
387 general solution here.
389 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
390 already-broken messages in the queue.
392 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
394 JH/19 Fix reference counting bug in routing-generated-address tracking.
400 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
401 supports it and a size is available (ie. the sending peer gave us one).
403 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
404 years of being deprecated). The replacements are the ACLs
405 acl_smtp_mime and acl_not_smtp_mime.
407 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
408 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
409 or even an in-clear connection were permitted. Now, if the host lookup
410 was dnssec and dane was requested then the host is only used if the
411 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
412 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
413 if one fails this test.
414 This means that a poorly-configured remote DNS will make it incommunicado;
415 but it protects against a DNS-interception attack on it.
417 JH/04 Bug 1810: make continued-use of an open smtp transport connection
418 non-noisy when a race steals the message being considered.
420 JH/05 If main configuration option tls_certificate is unset, generate a
421 self-signed certificate for inbound TLS connections.
423 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
424 in rewrites and routers.
426 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
427 and logged a warning sing 4.83; now they are a configuration file error.
429 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
430 (lacking @domain). Apply the same qualification processing as RCPT.
432 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
434 JH/10 Support ${sha256:} applied to a string (as well as the previous
437 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
438 a cutthrough deliver is pending, as we always want to make a connection.
439 This also avoids re-routing the message when later placing the cutthrough
440 connection after a verify cache hit.
441 Do not update it with the verify result either.
443 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
444 when routing results in more than one destination address.
446 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
447 signing (which inhibits the cutthrough capability). Previously only
448 the presence of an option was tested; now an expansion evaluating as
449 empty is permissible (obviously it should depend only on data available
450 when the cutthrough connection is made).
452 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
453 the relevant preceding SMTP command did not note the pipelining mode.
455 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
456 Previously they were not counted.
458 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
459 as one having no matching records. Previously we deferred the message
460 that needed the lookup.
462 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
463 distinguished as "(=".
465 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
466 for missing MX records. Previously it only worked for missing A records.
468 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
470 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
471 after the data-go-ahead and data-ack. Patch from Jason Betts.
473 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
474 even for a "none" policy. Patch from Tony Meyer.
476 JH/22 Fix continued use of a connection for further deliveries. If a port was
477 specified by a router, it must also match for the delivery to be
480 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
481 When one of the recipients of a message was unsuitable for the connection
482 (has no matching addresses), we lost track of needing to mark it
483 deferred. As a result mail would be lost.
485 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
487 JH/25 Decoding ACL controls is now done using a binary search; the source code
488 takes up less space and should be simpler to maintain. Merge the ACL
489 condition decode tables also, with similar effect.
491 JH/26 Fix problem with one_time used on a redirect router which returned the
492 parent address unchanged. A retry would see the parent address marked as
493 delivered, so not attempt the (identical) child. As a result mail would
496 JH/27 Fix a possible security hole, wherein a process operating with the Exim
497 UID can gain a root shell. Credit to http://www.halfdog.net/ for
498 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
501 JH/28 Enable {spool,log} filesystem space and inode checks as default.
502 Main config options check_{log,spool}_{inodes,space} are now
503 100 inodes, 10MB unless set otherwise in the configuration.
505 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
506 Previously it only applied to the main-section connection policy
509 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
511 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
512 by me. Added RFC7919 DH primes as an alternative.
514 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
515 are not in the system include path.
517 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
518 GnuTLS, when a session startup failed (eg because the client disconnected)
519 Exim did stdio operations after fclose. This was exposed by a recent
520 change which nulled out the file handle after the fclose.
522 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
523 signed directly by the cert-signing cert, rather than an intermediate
524 OCSP-signing cert. This is the model used by LetsEncrypt.
526 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
528 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
529 an incoming connection.
531 HS/02 Bug 1802: Do not half-close the connection after sending a request
534 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
535 fallback to "prime256v1".
537 JH/34 SECURITY: Use proper copy of DATA command in error message.
538 Could leak key material. Remotely exploitable. CVE-2016-9963.
544 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
545 and 3.4.4 - once the server is enabled to respond to an OCSP request
546 it does even when not requested, resulting in a stapling non-aware
547 client dropping the TLS connection.
549 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
550 support variable-length bit vectors. No functional change.
552 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
553 The I= interface field on outgoing lines is now after the H= remote
554 host field, same as incoming lines. There is a separate
555 outgoing_interface log selector which allows you to disable the
558 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
559 If not running log_selector +smtp_connection the mainlog would be held
560 open indefinitely after a "too many connections" event, including to a
561 deleted file after a log rotate. Leave the per net connection logging
562 leaving it open for efficiency as that will be quickly detected by the
563 check on the next write.
565 HS/01 Bug 1671: Fix post transport crash.
566 Processing the wait-<transport> messages could crash the delivery
567 process if the message IDs didn't exist for some reason. When
568 using 'split_spool_directory=yes' the construction of the spool
569 file name failed already, exposing the same netto behaviour.
571 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
572 mime_regex ACL conditions.
574 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
575 to DSN fail messages (bounces): remote IP, remote greeting, remote response
576 to HELO, local diagnostic string.
578 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
579 log line to debug. Even when configured with a tls authenticator many
580 client connections are expected to not authenticate in this way, so
581 an authenticate fail is not an error.
583 HS/02 Add the Exim version string to the process info. This way exiwhat
584 gives some more detail about the running daemon.
586 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
587 matter for fast-change records such as DNSBLs.
589 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
590 retry record, even if constant. There may be multiple transports with
591 different interface settings and the retry behaviour needs to be kept
594 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
596 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
598 JH/10 Bug 840: fix log_defer_output option of pipe transport
600 JH/11 Bug 830: use same host for all RCPTS of a message, even under
601 hosts_randomize. This matters a lot when combined with mua_wrapper.
603 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
604 ${quote_pgsql:<string>} operator.
606 JH/13 Bug 1708: avoid misaligned access in cached lookup.
608 JH/14 Change header file name for freeradius-client. Relevant if compiling
609 with Radius support; from the Gentoo tree and checked under Fedora.
611 JH/15 Bug 1712: Introduce $prdr_requested flag variable
613 JH/16 Bug 1714: Permit an empty string as expansion result for transport
614 option transport_filter, meaning no filtering.
616 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
618 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
619 defaults to "*" (all hosts). The variable is now available when not built
620 with TLS, default unset, mainly to enable keeping the testsuite sane.
621 If a server certificate is not supplied (via tls_certificate) an error is
622 logged, and clients will find TLS connections fail on startup. Presumably
623 they will retry in-clear.
624 Packagers of Exim are strongly encouraged to create a server certificate
625 at installation time.
627 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
628 with the $config_file variable.
630 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
631 in transport context, after the attempt, and per-recipient. The latter type
632 is per host attempted. The event data is the error message, and the errno
633 information encodes the lookup type (A vs. MX) used for the (first) host,
634 and the trailing two digits of the smtp 4xx response.
636 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
637 to write to mainlog (or rejectlog, paniclog) in the window between file
638 creation and permissions/ownership being changed. Particularly affects
639 installations where exicyclog is run as root, rather than exim user;
640 result is that the running daemon panics and dies.
642 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
644 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
645 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
646 "pri" and "weight". Note that the previous implicit priority given by the
647 list order is no longer honoured.
649 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
652 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
653 by defining SUPPORT_SOCKS.
655 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
656 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
657 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
658 variables are renamed to proxy_{local,external}_{address,port}.
660 JH/27 Move Internationalisation support from Experimental to mainline, enabled
661 for a build by defining SUPPORT_I18N
663 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
664 of the query string, and make ${quote_redis:} do that quoting.
666 JH/29 Move Events support from Experimental to mainline, enabled by default
667 and removable for a build by defining DISABLE_EVENT.
669 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
671 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
672 cached by the daemon.
674 JH/32 Move Redis support from Experimental to mainline, enabled for a build
675 by defining LOOKUP_REDIS. The libhiredis library is required.
677 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
678 keys are given for lookup.
680 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
681 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
682 only supported when built with TLS support. The PolarSSL SHA routines
683 are still used when the TLS library is too old for convenient support.
685 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
686 openssl_options), for security. OpenSSL forces this from version 1.1.0
687 server-side so match that on older versions.
689 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
690 allocation for $value could be released as the expansion processing
691 concluded, but leaving the global pointer active for it.
693 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
694 and to use the domains and local_parts ACL conditions.
696 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
697 incorrectly not doubled on cutthrough transmission, hence seen as a
698 body-termination at the receiving system - resulting in truncated mails.
699 Commonly the sender saw a TCP-level error, and retransmitted the message
700 via the normal store-and-forward channel. This could result in duplicates
701 received - but deduplicating mailstores were liable to retain only the
702 initial truncated version.
704 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
706 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
708 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
709 we're in there, support oversigning also; bug 1309.
711 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
713 HS/04 Add support for keep_environment and add_environment options.
715 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
716 either intentional arithmetic overflow during PRNG, or testing config-
719 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
720 delivery resulted in actual delivery. Cancel cutthrough before DATA
723 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
724 rejects a recipient: pass the reject to the originator.
726 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
727 Many were false-positives and ignorable, but it's worth fixing the
730 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
731 for the new environment-manipulation done at startup. Move the routines
732 from being local to tls.c to being global via the os.c file.
734 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
735 an extract embedded as result-arg for a map, the first arg for extract
736 is unavailable so we cannot tell if this is a numbered or keyed
737 extraction. Accept either.
743 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
746 JH/02 The smtp transport option "multi_domain" is now expanded.
748 JH/03 The smtp transport now requests PRDR by default, if the server offers
751 JH/04 Certificate name checking on server certificates, when exim is a client,
752 is now done by default. The transport option tls_verify_cert_hostnames
753 can be used to disable this per-host. The build option
754 EXPERIMENTAL_CERTNAMES is withdrawn.
756 JH/05 The value of the tls_verify_certificates smtp transport and main options
757 default to the word "system" to access the system default CA bundle.
758 For GnuTLS, only version 3.0.20 or later.
760 JH/06 Verification of the server certificate for a TLS connection is now tried
761 (but not required) by default. The verification status is now logged by
762 default, for both outbound TLS and client-certificate supplying inbound
765 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
768 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
769 Status Notification (bounce) messages are now MIME format per RFC 3464.
770 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
771 under the control of the dsn_advertise_hosts option, and routers may
772 have a dsn_lasthop option.
774 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
775 default, modifiable by a malware= option. The list separator for
776 the options can now be changed in the usual way. Bug 68.
778 JH/10 The smtp_receive_timeout main option is now expanded before use.
780 JH/11 The incoming_interface log option now also enables logging of the
781 local interface on delivery outgoing connections.
783 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
784 if the interface and destination host and port all match.
786 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
789 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
790 Patch from Andrew Lewis.
792 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
793 now supports optional time-restrictions, weighting, and priority
794 modifiers per server. Patch originally by <rommer@active.by>.
796 JH/16 The spamd_address main option now supports a mixed list of local
797 and remote servers. Remote servers can be IPv6 addresses, and
798 specify a port-range.
800 JH/17 Bug 68: The spamd_address main option now supports an optional
801 timeout value per server.
803 JH/18 Bug 1581: Router and transport options headers_add/remove can
804 now have the list separator specified.
806 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
809 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
812 JH/21 Support for the A6 type of dns record is withdrawn.
814 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
815 rather than the verbs used.
817 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
818 from 255 to 1024 chars.
820 JH/24 Verification callouts now attempt to use TLS by default.
822 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
823 are generic router options now. The defaults didn't change.
825 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
826 Original patch from Alexander Shikoff, worked over by JH.
828 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
829 files if ZCAT_COMMAND is not executable.
831 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
833 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
835 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
836 Normally benign, it bites when the pair was led to by a CNAME;
837 modern usage is to not canonicalize the domain to a CNAME target
838 (and we were inconsistent anyway for A-only vs AAAA+A).
840 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
842 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
843 when evaluating $sender_host_dnssec.
845 JH/31 Check the HELO verification lookup for DNSSEC, adding new
846 $sender_helo_dnssec variable.
848 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
850 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
852 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
854 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
855 documented as working, but never had. Support all but $spam_report.
857 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
858 added for tls authenticator.
860 HS/03 Add perl_taintmode main config option
866 TL/01 When running the test suite, the README says that variables such as
867 no_msglog_check are global and can be placed anywhere in a specific
868 test's script, however it was observed that placement needed to be near
869 the beginning for it to behave that way. Changed the runtest perl
870 script to read through the entire script once to detect and set these
871 variables, reset to the beginning of the script, and then run through
872 the script parsing/test process like normal.
874 TL/02 The BSD's have an arc4random API. One of the functions to induce
875 adding randomness was arc4random_stir(), but it has been removed in
876 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
877 function when detected.
879 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
880 cause callback expansion.
882 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
883 syntax errors in an expansion can be treated as a string instead of
884 logging or causing an error, due to the internal use of bool_lax
885 instead of bool when processing it.
887 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
888 server certificates when making smtp deliveries.
890 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
892 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
894 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
896 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
897 Merged patch from Sebastian Wiedenroth.
899 JH/05 Fix results-pipe from transport process. Several recipients, combined
900 with certificate use, exposed issues where response data items split
901 over buffer boundaries were not parsed properly. This eventually
902 resulted in duplicates being sent. This issue only became common enough
903 to notice due to the introduction of connection certificate information,
904 the item size being so much larger. Found and fixed by Wolfgang Breyha.
906 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
907 size buffer was used, resulting in syntax errors when an expansion
910 JH/07 Add support for directories of certificates when compiled with a GnuTLS
911 version 3.3.6 or later.
913 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
914 is EXPERIMENTAL_EVENT, the main-configuration and transport options
915 both become "event_action", the variables become $event_name, $event_data
916 and $event_defer_errno. There is a new variable $verify_mode, usable in
917 routers, transports and related events. The tls:cert event is now also
918 raised for inbound connections, if the main configuration event_action
921 TL/06 In test suite, disable OCSP for old versions of openssl which contained
922 early OCSP support, but no stapling (appears to be less than 1.0.0).
924 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
925 server certificate names available under the smtp transport option
926 "tls_verify_cert_hostname" now do not permit multi-component wildcard
929 JH/10 Time-related extraction expansions from certificates now use the main
930 option "timezone" setting for output formatting, and are consistent
931 between OpenSSL and GnuTLS compilations. Bug 1541.
933 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
934 encoded parameter in the incoming message. Bug 1558.
936 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
937 include certificate info, eximon was claiming there were spoolfile
940 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
942 JH/14 Log delivery-related information more consistently, using the sequence
943 "H=<name> [<ip>]" wherever possible.
945 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
946 are problematic for Debian distribution, omit them from the release
949 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
951 JH/16 Fix string representation of time values on 64bit time_t architectures.
954 JH/17 Fix a null-indirection in certextract expansions when a nondefault
955 output list separator was used.
960 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
961 checkers that were complaining about end of non-void function with no
964 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
965 This was a regression introduced in 4.83 by another bugfix.
967 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
969 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
970 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
976 TF/01 Correctly close the server side of TLS when forking for delivery.
978 When a message was received over SMTP with TLS, Exim failed to clear up
979 the incoming connection properly after forking off the child process to
980 deliver the message. In some situations the subsequent outgoing
981 delivery connection happened to have the same fd number as the incoming
982 connection previously had. Exim would try to use TLS and fail, logging
983 a "Bad file descriptor" error.
985 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
986 utilities have not been installed.
988 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
989 temporary space as the ACL may create new global variables.
991 TL/01 LDAP support uses per connection or global context settings, depending
992 upon the detected version of the libraries at build time.
994 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
995 to extract and use the src ip:port in logging and expansions as if it
996 were a direct connection from the outside internet. PPv2 support was
997 updated based on HAProxy spec change in May 2014.
999 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1001 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1002 Properly escape header and check for NULL return.
1004 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1007 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1009 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1010 characters in header names, implemented as a verify condition.
1011 Contributed by Michael Fischer v. Mollard.
1013 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1014 results permerror and temperror. Previous values are deprecated but
1015 still accepted. In a future release, err_perm and err_temp will be
1016 completely removed, which will be a backward incompatibility if the
1017 ACL tests for either of these two old results. Patch contributed by
1018 user bes-internal on the mailing list.
1020 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1022 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1023 selectors, in both main and reject logs.
1025 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1028 JH/07 Add malware type "sock" for talking to simple daemon.
1030 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1032 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1033 routers/transports under cutthrough routing.
1035 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1036 numbers. Touch up "bool" conditional to keep the same definition.
1038 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1040 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1042 JH/12 Expand items in router/transport headers_add or headers_remove lists
1043 individually rather than the list as a whole. Bug 1452.
1045 Required for reasonable handling of multiple headers_ options when
1046 they may be empty; requires that headers_remove items with embedded
1047 colons must have them doubled (or the list-separator changed).
1049 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1050 view the policy declared in the DMARC record. Currently, $dmarc_status
1051 is a combined value of both the record presence and the result of the
1054 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1056 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1057 dnslookup router and the smtp transport (applying to the forward
1060 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1061 of ldap servers used for a specific lookup. Patch provided by Heiko
1064 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1065 New variable $lookup_dnssec_authenticated for observability.
1067 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1068 Patch submitted by Lars Timman.
1070 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1072 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1073 Requires trusted mode and valid format message id, aborts otherwise.
1074 Patch contributed by Heiko Schlichting.
1076 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1077 certextract with support for various fields. Bug 1358.
1079 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1080 is requested by default, modifiable by smtp transport option
1083 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1084 operate on certificate variables to give certificate fingerprints
1085 Also new ${sha256:cert_variable}.
1087 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1089 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1090 Christian Aistleitner.
1092 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1094 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1095 file. Patch from Wolfgang Breyha.
1097 JH/25 Expand the coverage of the delivery $host and $host_address to
1098 client authenticators run in verify callout. Bug 1476.
1100 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1101 align with daemon_smtp_ports. Bug 72.
1103 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1104 support and error reporting did not work properly.
1106 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1107 and is readable. Patch from Andrew Colin Kissa.
1109 TL/14 Enhance documentation of ${run expansion and how it parses the
1110 commandline after expansion, particularly in the case when an
1111 unquoted variable expansion results in an empty value.
1113 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1115 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1116 by renaming away from it. Addresses GH issue 15, reported by
1119 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1121 TL/15 SECURITY: prevent double expansion in math comparison functions
1122 (can expand unsanitized data). Not remotely exploitable.
1129 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1131 PP/02 Make -n do something, by making it not do something.
1132 When combined with -bP, the name of an option is not output.
1134 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1137 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1138 $sender_host_name and config options to manage this, and basic check
1141 PP/05 DSCP support for outbound connections and control modifier for inbound.
1143 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1144 (Only plugin which currently uses this is kerberos4, which nobody should
1145 be using, but we should make it available and other future plugins might
1146 conceivably use it, even though it would break NAT; stuff *should* be
1147 using channel bindings instead).
1149 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1150 name; added for Sendmail compatibility; requires admin caller.
1151 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1152 just ignore it); requires trusted caller.
1153 Also parse but ignore: -Ac -Am -X<logfile>
1156 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1158 TL/02 Add +smtp_confirmation as a default logging option.
1160 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1161 Patch by Magnus Holmgren from 2007-02-20.
1163 TL/04 Bugzilla 1281 - Spec typo.
1164 Bugzilla 1283 - Spec typo.
1165 Bugzilla 1290 - Spec grammar fixes.
1167 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1169 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1171 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1172 mailing list by by Dmitry Isaikin.
1174 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1176 JH/02 Support "G" suffix to numbers in ${if comparisons.
1178 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1180 NM/01 Bugzilla 1197 - Spec typo
1181 Bugzilla 1196 - Spec examples corrections
1183 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1185 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1186 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1189 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1190 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1192 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1193 "acl {{name}{arg}...}", and optional args on acl condition
1196 JH/05 Permit multiple router/transport headers_add/remove lines.
1198 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1200 JH/07 Avoid using a waiting database for a single-message-only transport.
1201 Performance patch from Paul Fisher. Bugzilla 1262.
1203 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1206 JH/09 Add $headers_added variable, with content from use of ACL modifier
1207 add_header (but not yet added to the message). Bugzilla 199.
1209 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1210 Pulled from Bugzilla 817 by Wolfgang Breyha.
1212 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1214 (nb: this is the same fix as in Exim 4.80.1)
1216 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1219 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1222 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1224 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1225 advertise SMTP AUTH mechanism to us, instead of a generic
1226 protocol violation error. Also, make Exim more robust to bad
1227 data from the Dovecot auth socket.
1229 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1231 When a queue runner is handling a message, Exim first routes the
1232 recipient addresses, during which it prunes them based on the retry
1233 hints database. After that it attempts to deliver the message to
1234 any remaining recipients. It then updates the hints database using
1237 So if a recipient address works intermittently, it can get repeatedly
1238 deferred at routing time. The retry hints record remains fresh so the
1239 address never reaches the final cutoff time.
1241 This is a fairly common occurrence when a user is bumping up against
1242 their storage quota. Exim had some logic in its local delivery code
1243 to deal with this. However it did not apply to per-recipient defers
1244 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1246 This change adds a proper retry rule check during routing so that the
1247 final cutoff time is checked against the message's age. We only do
1248 this check if there is an address retry record and there is not a
1249 domain retry record; this implies that previous attempts to handle
1250 the address had the retry_use_local_parts option turned on. We use
1251 this as an approximation for the destination being like a local
1252 delivery, as in LMTP.
1254 I suspect this new check makes the old local delivery cutoff check
1255 redundant, but I have not verified this so I left the code in place.
1257 TF/02 Correct gecos expansion when From: is a prefix of the username.
1259 Test 0254 submits a message to Exim with the header
1263 When I ran the test suite under the user fanf2, Exim expanded
1264 the header to contain my full name, whereas it should have added
1265 a Resent-Sender: header. It erroneously treats any prefix of the
1266 username as equal to the username.
1268 This change corrects that bug.
1270 GF/01 DCC debug and logging tidyup
1271 Error conditions log to paniclog rather than rejectlog.
1272 Debug lines prefixed by "DCC: " to remove any ambiguity.
1274 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1276 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1277 Bug spotted by Jeremy Harris; was flawed since initial commit.
1278 Would have resulted in OCSP responses post-SNI triggering an Exim
1279 NULL dereference and crash.
1281 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1283 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1284 Bug detection, analysis and fix by Samuel Thibault.
1285 Bugzilla 1331, Debian bug #698092.
1287 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1289 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1290 Server implementation by Todd Lyons, client by JH.
1291 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1292 config variable "prdr_enable" controls whether the server
1293 advertises the facility. If the client requests PRDR a new
1294 acl_data_smtp_prdr ACL is called once for each recipient, after
1295 the body content is received and before the acl_smtp_data ACL.
1296 The client is controlled by both of: a hosts_try_prdr option
1297 on the smtp transport, and the server advertisement.
1298 Default client logging of deliveries and rejections involving
1299 PRDR are flagged with the string "PRDR".
1301 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1302 fclose(). Diagnosis by Todd Lyons.
1304 PP/17 Update configure.default to handle IPv6 localhost better.
1305 Patch by Alain Williams (plus minor tweaks).
1308 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1309 This is now consistent with GnuTLS, and is now documented: the
1310 previous undocumented portable approach to treating the option as
1311 unset was to force an expansion failure. That still works, and
1312 an empty string is now equivalent.
1314 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1315 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1316 not performing validation itself.
1318 PP/20 Added force_command boolean option to pipe transport.
1319 Patch from Nick Koston, of cPanel Inc.
1321 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1324 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1326 PP/21 Fix eximon continuous updating with timestamped log-files.
1327 Broken in a format-string cleanup in 4.80, missed when I repaired the
1328 other false fix of the same issue.
1329 Report and fix from Heiko Schlichting.
1332 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1333 Report from Prashanth Katuri.
1335 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1336 It's SecureTransport, so affects any MacOS clients which use the
1337 system-integrated TLS libraries, including email clients.
1339 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1340 using a MIME ACL for non-SMTP local injection.
1341 Report and assistance in diagnosis by Warren Baker.
1343 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1345 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1347 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1348 last id that failed so it may be referenced in subsequent ACL's.
1350 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1353 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1354 ldap library initialization, allowing self-signed CA's to be
1355 used. Also properly sets require_cert option later in code by
1356 using NULL (global ldap config) instead of ldap handle (per
1357 session). Bug diagnosis and testing by alxgomz.
1359 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1360 the src/util/ subdirectory.
1362 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1363 renamed to Transport Post Delivery Action by Jeremy Harris, as
1366 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1367 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1368 redis_servers = needs to be configured which will be used by the redis
1369 lookup. Patch from Warren Baker, of The Packet Hub.
1371 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1373 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1374 hostname or reverse DNS when processing a host list. Used suggestions
1375 from multiple comments on this bug.
1377 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1379 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1380 Missed a few lines, added it to make the runtest require no keyboard
1383 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1384 contains upper case chars. Make router use caseful_local_part.
1386 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1387 support when GnuTLS has been built with p11-kit.
1393 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1395 This, or similar/improved, will also be change PP/11 of 4.82.
1401 PP/01 Handle short writes when writing local log-files.
1402 In practice, only affects FreeBSD (8 onwards).
1403 Bugzilla 1053, with thanks to Dmitry Isaikin.
1405 NM/01 Bugzilla 949 - Documentation tweak
1407 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1410 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1412 PP/02 Implemented gsasl authenticator.
1414 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1416 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1417 `pkg-config foo` for cflags/libs.
1419 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1420 with rest of GSASL and with heimdal_gssapi.
1422 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1423 `pkg-config foo` for cflags/libs for the TLS implementation.
1425 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1426 properties get this fed in as external SSF. A number of robustness
1427 and debugging improvements to the cyrus_sasl authenticator.
1429 PP/08 cyrus_sasl server now expands the server_realm option.
1431 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1432 Patch by Jeremy Harris.
1434 PP/10 Added dbmjz lookup type.
1436 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1438 PP/12 MAIL args handles TAB as well as SP, for better interop with
1439 non-compliant senders.
1440 Analysis and variant patch by Todd Lyons.
1442 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1443 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1444 Patch from Dirk Mueller <dmueller@suse.com>
1446 PP/13 tls_peerdn now print-escaped for spool files.
1447 Observed some $tls_peerdn in wild which contained \n, which resulted
1448 in spool file corruption.
1450 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1451 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1452 or write after TLS renegotiation, which otherwise led to messages
1455 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1456 as a tracking header (ie: a signed header comes before the signature).
1457 Patch from Wolfgang Breyha.
1459 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1460 comma-sep list; embedded commas doubled.
1462 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1464 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1466 Report and patch from Dmitry Banschikov.
1468 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1469 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1470 failures. We appear to now support TLS1.1+ with Exim.
1472 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1473 lets Exim select keys and certificates based upon TLS SNI from client.
1474 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1475 before an outbound SMTP session. New log_selector, +tls_sni.
1477 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1478 NULL dereference. Report and patch from Alun Jones.
1480 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1481 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1482 Not seeing resolver debug output on NetBSD, but suspect this is a
1483 resolver implementation change.
1485 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1486 Left warnings. Added "eximon gdb" invocation mode.
1488 PP/21 Defaulting "accept_8bitmime" to true, not false.
1490 PP/22 Added -bw for inetd wait mode support.
1492 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1493 locate the relevant includes and libraries. Made this the default.
1495 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1496 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1498 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1499 This may cause build issues on older platforms.
1501 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1502 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1503 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1504 Added SNI support via GnuTLS too.
1505 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1507 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1509 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1510 Applied second patch from Janne, implementing suggestion to default
1511 multiple-strings-in-record handling to match SPF spec.
1513 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1515 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1516 read-only, out of scope).
1517 Patch from Wolfgang Breyha, report from Stuart Northfield.
1519 PP/29 Fix three issues highlighted by clang analyser static analysis.
1520 Only crash-plausible issue would require the Cambridge-specific
1521 iplookup router and a misconfiguration.
1522 Report from Marcin Mirosław.
1524 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1526 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1527 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1528 As part of this, removing so much warning spew let me fix some minor
1529 real issues in debug logging.
1531 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1532 assignment on my part. Fixed.
1534 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1535 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1536 Janne Snabb (who went above and beyond: thank you).
1538 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1539 string otherwise requires a connection and a bunch more work and it's
1540 relatively easy to get wrong. Should also expose TLS library linkage
1543 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1544 64-bit ${eval} (JH/03).
1546 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1547 GNU libc to support some of the 64-bit stuff, should not lead to
1548 conflicts. Defined before os.h is pulled in, so if a given platform
1549 needs to override this, it can.
1551 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1552 protection layer was required, which is not implemented.
1553 Bugzilla 1254, patch from Wolfgang Breyha.
1555 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1556 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1557 tls_dhparam take prime identifiers. Also unbreak combination of
1558 OpenSSL+DH_params+TLSSNI.
1560 PP/39 Disable SSLv2 by default in OpenSSL support.
1566 PP/01 Solaris build fix for Oracle's LDAP libraries.
1567 Bugzilla 1109, patch from Stephen Usher.
1569 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1571 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1574 TF/02 Fix a couple more cases where we did not log the error message
1575 when unlink() failed. See also change 4.74-TF/03.
1577 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1578 lock up or crash if it happened to be inside a call to libc when it
1579 got a SIGUSR1 from exiwhat.
1581 The SIGUSR1 handler appends the current process status to the process
1582 log which is later printed by exiwhat. It used to use the general
1583 purpose logging code to do this, but several functions it calls are
1584 not safe for signals.
1586 The new output code in the SIGUSR1 handler is specific to the process
1587 log, and simple enough that it's easy to inspect for signal safety.
1588 Removing some special cases also simplifies the general logging code.
1589 Removing the spurious timestamps from the process log simplifies
1592 TF/04 Improved ratelimit ACL condition.
1594 The /noupdate option has been deprecated in favour of /readonly which
1595 has clearer semantics. The /leaky, /strict, and /readonly update modes
1596 are mutually exclusive. The update mode is no longer included in the
1597 database key; it just determines when the database is updated. (This
1598 means that when you upgrade Exim will forget old rate measurements.)
1600 Exim now checks that the per_* options are used with an update mode that
1601 makes sense for the current ACL. For example, when Exim is processing a
1602 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1603 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1604 must specify per_mail/readonly. If you omit the update mode it defaults to
1605 /leaky where that makes sense (as before) or /readonly where required.
1607 The /noupdate option is now undocumented but still supported for
1608 backwards compatibility. It is equivalent to /readonly except that in
1609 ACLs where /readonly is required you may specify /leaky/noupdate or
1610 /strict/noupdate which are treated the same as /readonly.
1612 A useful new feature is the /count= option. This is a generalization
1613 of the per_byte option, so that you can measure the throughput of other
1614 aggregate values. For example, the per_byte option is now equivalent
1615 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1617 The per_rcpt option has been generalized using the /count= mechanism
1618 (though it's more complicated than the per_byte equivalence). When it is
1619 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1620 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1621 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1622 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1623 Note that using per_rcpt with a non-readonly update mode in more than
1624 one ACL will cause the recipients to be double-counted. (The per_mail
1625 and per_byte options don't have this problem.)
1627 The handling of very low rates has changed slightly. If the computed rate
1628 is less than the event's count (usually one) then this event is the first
1629 after a long gap. In this case the rate is set to the same as this event's
1630 count, so that the first message of a spam run is counted properly.
1632 The major new feature is a mechanism for counting the rate of unique
1633 events. The new per_addr option counts the number of different
1634 recipients that someone has sent messages to in the last time period. It
1635 behaves like per_rcpt if all the recipient addresses are different, but
1636 duplicate recipient addresses do not increase the measured rate. Like
1637 the /count= option this is a general mechanism, so the per_addr option
1638 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1639 example, measure the rate that a client uses different sender addresses
1640 with the options per_mail/unique=$sender_address. There are further
1641 details in the main documentation.
1643 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1645 TF/06 Removed a few PCRE remnants.
1647 TF/07 Automatically extract Exim's version number from tags in the git
1648 repository when doing development or release builds.
1650 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1651 Bugzilla 879. Patch from Paul Fisher.
1653 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1654 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1657 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1659 PP/05 Variable $av_failed, true if the AV scanner deferred.
1660 Bugzilla 1078. Patch from John Horne.
1662 PP/06 Stop make process more reliably on build failure.
1663 Bugzilla 1087. Patch from Heiko Schlittermann.
1665 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1666 Bugzilla 1089. Patch from Heiko Schlittermann.
1668 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1669 Bugzilla 1131. Patch from Holger Weiß.
1671 PP/09 Handle IPv6 addresses with SPF.
1672 Bugzilla 860. Patch from Wolfgang Breyha.
1674 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1676 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1679 PP/11 match_* no longer expand right-hand-side by default.
1680 New compile-time build option, EXPAND_LISTMATCH_RHS.
1681 New expansion conditions, "inlist", "inlisti".
1683 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1685 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1687 PP/14 fix log_write() format string regression from TF/03.
1688 Bugzilla 1152. Patch from Dmitry Isaikin.
1694 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1696 PP/02 Harmonised TLS library version reporting; only show if debugging.
1697 Layout now matches that introduced for other libraries in 4.74 PP/03.
1699 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1701 PP/04 New "dns_use_edns0" global option.
1703 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1706 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1707 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1709 TK/01 Updated PolarSSL code to 0.14.2.
1710 Bugzilla 1097. Patch from Andreas Metzler.
1712 PP/07 Catch divide-by-zero in ${eval:...}.
1713 Fixes bugzilla 1102.
1715 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1718 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1719 format-string attack -- SECURITY: remote arbitrary code execution.
1721 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1722 time unintentionally subject to list matching rules, letting the header
1723 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1724 arbitrary string expansion). This allowed for information disclosure.
1726 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1727 INT_MIN/-1 -- value coerced to INT_MAX.
1733 NM/01 Workaround for PCRE version dependency in version reporting
1736 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1737 This fixes portability to compilers other than gcc, notably
1738 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1740 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1741 makefiles for portability to HP-UX and POSIX correctness.
1743 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1744 Also via indented variable definition in the Makefile.
1745 (Debugging by Oliver Heesakkers).
1747 PP/02 Restore caching of spamd results with expanded spamd_address.
1748 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1750 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1751 Improves build reliability. Fix from: Frank Elsner
1753 NM/02 Fix wide character breakage in the rfc2047 coding
1754 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1756 NM/03 Allow underscore in dnslist lookups
1757 Fixes bug 1026. Patch from Graeme Fowler
1759 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1760 Code patches from Adam Ciarcinski of NetBSD.
1762 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1765 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1766 is logged, to avoid truncation. Patch from John Horne.
1768 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1769 Patch from Jakob Hirsch.
1771 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1772 SQL string expansion failure details.
1773 Patch from Andrey Oktyabrski.
1775 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1776 Patch from Simon Arlott.
1778 PP/09 New lookups functionality failed to compile on old gcc which rejects
1779 extern declarations in function scope.
1780 Patch from Oliver Fleischmann
1782 PP/10 Use sig_atomic_t for flags set from signal handlers.
1783 Check getgroups() return and improve debugging.
1784 Fixed developed for diagnosis in bug 927 (which turned out to be
1787 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1788 Patch from Mark Zealey.
1790 PP/12 Bugzilla 1056: Improved spamd server selection.
1791 Patch from Mark Zealey.
1793 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1794 Based on patch from Heiko Schlittermann.
1796 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1797 Patch from Uwe Doering, sign-off by Michael Haardt.
1799 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1800 variable declaration deep within a block. Bug and patch from
1803 PP/15 lookups-Makefile IRIX compatibility coercion.
1805 PP/16 Make DISABLE_DKIM build knob functional.
1807 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1808 Patch by Simon Arlott
1810 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1811 variable argument macros. Our copy now differs from upstream.
1817 TF/01 Failure to get a lock on a hints database can have serious
1818 consequences so log it to the panic log.
1820 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1821 controlled using the smtp_confirmation log selector.
1823 TF/03 Include the error message when we fail to unlink a spool file.
1825 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1826 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1827 for maintaining out-of-tree patches for some time.
1829 PP/01 Bugzilla 139: Documentation and portability issues.
1830 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1831 Handle per-OS dynamic-module compilation flags.
1833 PP/02 Let /dev/null have normal permissions.
1834 The 4.73 fixes were a little too stringent and complained about the
1835 permissions on /dev/null. Exempt it from some checks.
1836 Reported by Andreas M. Kirchwitz.
1838 PP/03 Report version information for many libraries, including
1839 Exim version information for dynamically loaded libraries. Created
1840 version.h, now support a version extension string for distributors
1841 who patch heavily. Dynamic module ABI change.
1843 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1844 privilege escalation vulnerability whereby the Exim run-time user
1845 can cause root to append content of the attacker's choosing to
1848 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1851 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1852 If dropping privileges for untrusted macros, we disabled normal logging
1853 on the basis that it would fail; for the Exim run-time user, this is not
1854 the case, and it resulted in successful deliveries going unlogged.
1855 Fixed. Reported by Andreas Metzler.
1861 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1862 only prepend for the Resent-* case. Fixes regression introduced in
1863 Exim 4.70 by NM/22 for Bugzilla 607.
1865 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1866 increasing numbers of administrators be bitten by this.
1868 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1870 PP/03 Bugzilla 994: added openssl_options main configuration option.
1872 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1874 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1876 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1878 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1879 without a peer certificate, leading to a segfault because of an
1880 assumption that peers always have certificates. Be a little more
1881 paranoid. Problem reported by Martin Tscholak.
1883 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1884 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1885 NB: ClamAV planning to remove STREAM in "middle of 2010".
1886 CL also introduces -bmalware, various -d+acl logging additions and
1887 more caution in buffer sizes.
1889 PP/09 Implemented reverse_ip expansion operator.
1891 PP/10 Bugzilla 937: provide a "debug" ACL control.
1893 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1895 PP/12 Bugzilla 973: Implement --version.
1897 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1899 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1901 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1903 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1904 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1905 ignore trailing whitespace.
1907 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1909 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1912 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1913 Notification from Dr Andrew Aitchison.
1915 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1916 ExtendedDetectionInfo response format.
1917 Notification from John Horne.
1919 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1922 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1923 XSL and documented dependency on system catalogs, with examples of how
1926 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1929 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1930 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1931 configuration file which is writeable by the Exim user or group.
1933 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1934 of configuration files to cover files specified with the -C option if
1935 they are going to be used with root privileges, not just the default
1938 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1939 option (effectively making it always true).
1941 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1942 files to be used while preserving root privileges.
1944 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1945 that rogue child processes cannot use them.
1947 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1948 run-time user, instead of root.
1950 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1951 Exim run-time user without dropping privileges.
1953 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1954 result string, instead of calling string_vformat() twice with the same
1957 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1958 for other users. Others should always drop root privileges if they use
1959 -C on the command line, even for a whitelisted configure file.
1961 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1963 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1969 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1970 $data_path, and $header_path variables; fixed documentation bugs and
1973 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1974 exipick to access non-standard spools, including the "frozen" queue
1977 NM/01 Bugzilla 965: Support mysql stored procedures.
1978 Patch from Alain Williams
1980 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1982 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1983 Patch from Andreas Metzler
1985 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1986 Patch from Kirill Miazine
1988 NM/05 Bugzilla 671: Added umask to procmail example.
1990 JJ/03 installed exipick 20100323.0, fixing doc bug
1992 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1993 directory. Notification and patch from Dan Rosenberg.
1995 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1997 TK/02 Improve log output when DKIM signing operation fails.
1999 MH/01 Treat the transport option dkim_domain as a colon separated
2000 list, not as a single string, and sign the message with each element,
2001 omitting multiple occurences of the same signer.
2003 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2004 Bugzilla 985, 986. Patch by Simon Arlott
2006 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2007 Patch by Simon Arlott
2009 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2010 MBX locking. Notification from Dan Rosenberg.
2016 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2018 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2020 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2022 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2024 NM/04 Bugzilla 915: Flag broken perl installation during build.
2030 TK/01 Added patch by Johannes Berg that expands the main option
2031 "spamd_address" if it starts with a dollar sign.
2033 TK/02 Write list of recipients to X-Envelope-Sender header when building
2034 the mbox-format spool file for content scanning (suggested by Jakob
2037 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2038 (http://www.dcc-servers.net/) support via dccifd. Activated by
2039 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2041 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2042 by Mark Daniel Reidel <mr@df.eu>.
2044 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2045 When building exim an external PCRE library is now needed -
2046 PCRE is a system library on the majority of modern systems.
2047 See entry on PCRE_LIBS in EDITME file.
2049 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2050 conversation. Added nologin parameter to request.
2051 Patch contributed by Kirill Miazine.
2053 TF/01 Do not log submission mode rewrites if they do not change the address.
2055 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2057 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2058 log files in place. Contributed by Roberto Lima.
2060 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2062 TF/03 Bugzilla 615: When checking the local_parts router precondition
2063 after a local_part_suffix or local_part_prefix option, Exim now
2064 does not use the address's named list lookup cache, since this
2065 contains cached lookups for the whole local part.
2067 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2068 Robert Millan. Documentation is in experimental-spec.txt.
2070 TF/04 Bugzilla 668: Fix parallel build (make -j).
2072 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2074 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2075 Patch provided by Jan Srzednicki.
2077 TF/05 Leading white space used to be stripped from $spam_report which
2078 wrecked the formatting. Now it is preserved.
2080 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2081 that they are available at delivery time.
2083 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2085 TF/08 TLS error reporting now respects the incoming_interface and
2086 incoming_port log selectors.
2088 TF/09 Produce a more useful error message if an SMTP transport's hosts
2089 setting expands to an empty string.
2091 NM/06 Bugzilla 744: EXPN did not work under TLS.
2092 Patch provided by Phil Pennock.
2094 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2095 Patch provided by Richard Godbee.
2097 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2098 acl_smtp_notquit, added index entry.
2100 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2101 Patch provided by Eugene Bujak.
2103 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2104 accept(). Patch provided by Maxim Dounin.
2106 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2107 Patch provided by Phil Pennock.
2109 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2111 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2112 Patch provided by Brad "anomie" Jorsch.
2114 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2115 Patch provided by Dean Brooks.
2117 TK/05 Add native DKIM support (does not depend on external libraries).
2119 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2120 Patch provided by Graeme Fowler.
2122 NM/16 Bugzilla 851: Documentation example syntax fix.
2124 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2126 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2129 NM/19 Bugzilla 745: TLS version reporting.
2130 Patch provided by Phil Pennock.
2132 NM/20 Bugzilla 167: bool: condition support.
2133 Patch provided by Phil Pennock.
2135 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2136 clients. Patch provided by Phil Pennock.
2138 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2139 Patch provided by Brad "anomie" Jorsch.
2141 NM/23 Bugzilla 687: Fix misparses in eximstats.
2142 Patch provided by Heiko Schlittermann.
2144 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2145 Patch provided by Heiko Schlittermann.
2147 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2148 plus update to original patch.
2150 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2152 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2153 Patch provided by David Brownlee.
2155 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2157 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2159 NM/30 Bugzilla 888: TLS documentation bugfixes.
2161 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2163 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2164 Unlike the original bugzilla I have changed all shell scripts in src tree.
2166 NM/33 Bugzilla 898: Transport filter timeout fix.
2167 Patch by Todd Rinaldo.
2169 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2170 Patch by Serge Demonchaux.
2172 NM/35 Bugzilla 39: Base64 decode bug fixes.
2173 Patch by Jakob Hirsch.
2175 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2177 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2179 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2181 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2187 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2188 ALT-N's libdkim that I have put here:
2189 http://duncanthrax.net/exim-experimental/
2191 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2192 were called 'true' and it seems that C99 defines that as a reserved
2193 keyword to be used with 'bool' variable types. That means you could
2194 not include C99-style headers which use bools without triggering
2195 build errors in sieve.c.
2197 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2198 as mailq or other aliases. Changed the --help handling significantly
2199 to do whats expected. exim_usage() emits usage/help information.
2201 SC/01 Added the -bylocaldomain option to eximstats.
2203 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2205 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2207 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2213 PH/01 Another patch from the Sieve maintainer.
2215 PH/02 When an IPv6 address is converted to a string for single-key lookup
2216 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2217 dots are used instead of colons so that keys in lsearch files need not
2218 contain colons. This was done some time before quoting was made available
2219 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2220 (notated using the quote facility) so as to distinguish them from IPv4
2221 keys. This meant that lookups for IP addresses in host lists did not work
2222 for iplsearch lookups.
2224 This has been fixed by arranging for IPv6 addresses to be expressed with
2225 colons if the lookup type is iplsearch. This is not incompatible, because
2226 previously such lookups could never work.
2228 The situation is now rather anomalous, since one *can* have colons in
2229 ordinary lsearch keys. However, making the change in all cases is
2230 incompatible and would probably break a number of configurations.
2232 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2235 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2236 conversion specification without a maximum field width, thereby enabling
2237 a rogue spamd server to cause a buffer overflow. While nobody in their
2238 right mind would setup Exim to query an untrusted spamd server, an
2239 attacker that gains access to a server running spamd could potentially
2240 exploit this vulnerability to run arbitrary code as the Exim user.
2242 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2243 $primary_hostname instead of what libspf2 thinks the hosts name is.
2245 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2246 a directory entry by the name of the lookup key. Previously, if a
2247 symlink pointed to a non-existing file or a file in a directory that
2248 Exim lacked permissions to read, a lookup for a key matching that
2249 symlink would fail. Now it is enough that a matching directory entry
2250 exists, symlink or not. (Bugzilla 503.)
2252 PH/03 The body_linecount and body_zerocount variables are now exported in the
2255 PH/04 Added the $dnslist_matched variable.
2257 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2258 This means they are set thereafter only if the connection becomes
2261 PH/06 Added the client_condition to authenticators so that some can be skipped
2262 by clients under certain conditions.
2264 PH/07 The error message for a badly-placed control=no_multiline_responses left
2265 "_responses" off the end of the name.
2267 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2269 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2270 (without spaces) instead of just copying the configuration text.
2272 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2274 PH/11 Added $max_received_linelength.
2276 PH/12 Added +ignore_defer and +include_defer to host lists.
2278 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2279 way in which PCRE > 7.0 is built.
2281 PH/14 Implemented queue_only_load_latch.
2283 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2284 MAIL command. The effect was to mangle the value on 64-bit systems.
2286 PH/16 Another patch from the Sieve maintainer.
2288 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2290 PH/18 If a system quota error occurred while trying to create the file for
2291 a maildir delivery, the message "Mailbox is full" was not appended to the
2292 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2293 only to a quota excession during the actual writing of the file.
2295 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2296 characters?) which causes problems in log lines. The DN values are now
2297 passed through string_printing() before being added to log lines.
2299 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2300 and InterBase are left for another time.)
2302 PH/21 Added message_body_newlines option.
2304 PH/22 Guard against possible overflow in moan_check_errorcopy().
2306 PH/23 POSIX allows open() to be a macro; guard against that.
2308 PH/24 If the recipient of an error message contained an @ in the local part
2309 (suitably quoted, of course), incorrect values were put in $domain and
2310 $local_part during the evaluation of errors_copy.
2316 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2317 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2320 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2321 issue a MAIL command.
2323 PH/02 In an ACL statement such as
2325 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2327 if a client was not listed at all, or was listed with a value other than
2328 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2329 the condition was not true (as it should be), so access was not denied.
2330 The bug was that the ! inversion was incorrectly passed on to the second
2331 item. This has been fixed.
2333 PH/03 Added additional dnslists conditions == and =& which are different from
2334 = and & when the dns lookup returns more than one IP address.
2336 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2337 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2339 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2340 FSYNC, which compiles an option called disable_fsync that allows for
2341 bypassing fsync(). The documentation is heavily laced with warnings.
2343 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2345 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2346 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2347 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2348 including adding "make clean"; (3) Added -fPIC when compiling the test
2349 dynamically loaded module, to get rid of a warning.
2351 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2352 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2353 = 0s. The bug is otherwise harmless.
2355 PH/07 There was a bug in the dovecot authenticator such that the value of
2356 $auth1 could be overwritten, and so not correctly preserved, after a
2357 successful authentication. This usually meant that the value preserved by
2358 the server_setid option was incorrect.
2360 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2362 PH/09 Installed PCRE release 7.0.
2364 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2365 run for batched SMTP input. It is now run at the start of every message
2366 in the batch. While fixing this I discovered that the process information
2367 (output by running exiwhat) was not always getting set for -bs and -bS
2368 input. This is fixed, and it now also says "batched" for BSMTP.
2370 PH/11 Added control=no_pipelining.
2372 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2373 patch, slightly modified), and move the expansion of helo_data till after
2374 the connection is made in the smtp transport (so it can use these
2377 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2379 PH/14 Added log_selector = +pid.
2381 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2383 PH/16 Add ${if forany and ${if forall.
2385 PH/17 Added dsn_from option to vary the From: line in DSNs.
2387 PH/18 Flush SMTP output before performing a callout, unless control =
2388 no_callout_flush is set.
2390 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2391 was true (the default) a successful delivery failed to delete the retry
2392 item, thus causing premature timeout of the address. The bug is now
2395 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2397 PH/21 Long custom messages for fakedefer and fakereject are now split up
2398 into multiline responses in the same way that messages for "deny" and
2399 other ACL rejections are.
2401 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2402 with slight modification.
2404 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2405 draft, changing the syntax and factoring some duplicate code".
2407 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2408 for deliveries of the second and subsequent messages over the same SMTP
2411 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2412 ${reduce, with only minor "tidies".
2414 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2416 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2417 expansion side effects.
2419 PH/27 When a message times out after an over-quota error from an Exim-imposed
2420 quota, the bounce message says "mailbox is full". This message was not
2421 being given when it was a system quota that was exceeded. It now should
2424 MH/03 Made $recipients available in local_scan(). local_scan() already has
2425 better access to the recipient list through recipients_list[], but
2426 $recipients can be useful in postmaster-provided expansion strings.
2428 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2429 in the case of a MAIL command with additional options following the
2430 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2431 were accidentally chopped off.
2433 PH/29 SMTP synchronization checks are implemented when a command is read -
2434 there is a check that no more input is waiting when there shouldn't be
2435 any. However, for some commands, a delay in an ACL can mean that it is
2436 some time before the response is written. In this time, more input might
2437 arrive, invalidly. So now there are extra checks after an ACL has run for
2438 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2439 pipelining has not been advertised.
2441 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2443 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2444 returned in the SMTP response when smtp_return_error_details was false.
2445 This has been fixed.
2447 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2448 instead of the C I/O that was originally supplied, because problems were
2449 reported on Solaris.
2451 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2452 Exim which did not show up earlier: it was assuming that a call to
2453 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2454 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2455 was a macro that became an assignment, so it seemed to work. This has
2456 changed to a proper function call with a void return, hence the compile
2457 error. Exim's code has been fixed.
2459 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2462 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2464 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2466 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2467 because it turns out to be too big or there is a timeout) there is no
2468 "Completed" line in the log. When some messages of this type were
2469 selected by exigrep, they were listed as "not completed". Others were
2470 picked up by some special patterns. I have improved the selection
2471 criteria to be more general.
2473 PH/38 The host_find_failed option in the manualroute router can now be set
2474 to "ignore", to completely ignore a host whose IP address cannot be
2475 found. If all hosts are ignored, the behaviour is controlled by the new
2476 host_all_ignored option.
2478 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2479 homing or because of multiple MX records with /mx) generated more than
2480 one IP address, and the following item turned out to be the local host,
2481 all the secondary addresses of the first item were incorrectly removed
2482 from the list, along with the local host and any following hosts (which
2483 is what is supposed to happen).
2485 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2486 whoever called Exim into the -H file. In the case of the daemon it was
2487 behaving confusingly. When first started, it used values for whoever
2488 started the daemon, but after a SIGHUP it used the Exim user (because it
2489 calls itself on a restart). I have changed the code so that it now always
2492 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2493 message are rejected with the same error (e.g. no authentication or bad
2494 sender address), and a DATA command is nevertheless sent (as can happen
2495 with PIPELINING or a stupid MUA), the error message that was given to the
2496 RCPT commands is included in the rejection of the DATA command. This is
2497 intended to be helpful for MUAs that show only the final error to their
2500 PH/42 Another patch from the Sieve maintainer.
2502 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2503 Eximstats - Fixed some broken HTML links and added missing column headers
2505 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2506 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2508 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2510 PH/43 Yet another patch from the Sieve maintainer.
2512 PH/44 I found a way to check for a TCP/IP connection going away before sending
2513 the response to the final '.' that terminates a message, but only in the
2514 case where the client has not sent further data following the '.'
2515 (unfortunately, this is allowed). However, in many cases there won't be
2516 any further data because there won't be any more messages to send. A call
2517 to select() can be used: if it shows that the input is "ready", there is
2518 either input waiting, or the socket has been closed. An attempt to read
2519 the next input character can distinguish the two cases. Previously, Exim
2520 would have sent an OK response which the client would never have see.
2521 This could lead to message repetition. This fix should cure that, at
2522 least in a lot of common cases.
2524 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2525 advertised in response to EHLO.
2531 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2532 fixed by 4.65/MH/01 (is this a record?) are fixed:
2534 (i) An empty string was always treated as zero by the numeric comparison
2535 operators. This behaviour has been restored.
2537 (ii) It is documented that the numeric comparison operators always treat
2538 their arguments as decimal numbers. This was broken in that numbers
2539 starting with 0 were being interpreted as octal.
2541 While fixing these problems I realized that there was another issue that
2542 hadn't been noticed. Values of message_size_limit (both the global option
2543 and the transport option) were treated as octal if they started with 0.
2544 The documentation was vague. These values are now always treated as
2545 decimal, and I will make that clear in the documentation.
2551 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2552 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2555 MH/01 Don't check that the operands of numeric comparison operators are
2556 integers when their expansion is in "skipping" mode (fixes bug
2557 introduced by 4.64-PH/07).
2559 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2560 child addresses, Exim now panics and dies. Previously, because the count
2561 is held in a short int, deliveries were likely to be lost. As such a
2562 large number of recipients for a single message is ridiculous
2563 (performance will be very, very poor), I have chosen to impose a limit
2564 rather than extend the field.
2570 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2571 leftover -K file (the existence of which was triggered by #402).
2572 While we were at it, introduced process PID as part of the -K
2573 filename. This should rule out race conditions when creating
2576 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2577 processing considerably. Previous code took too long for large mails,
2578 triggering a timeout which in turn triggers #401.
2580 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2581 in the DK code in transports.c. sendfile() is not really portable,
2582 hence the _LINUX specificness.
2584 TF/01 In the add_headers option to the mail command in an Exim filter,
2585 there was a bug that Exim would claim a syntax error in any
2586 header after the first one which had an odd number of characters
2589 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2590 callout verification, Exim cached a "reject" for the entire domain. This
2591 is correct for most verifications, but it is not correct for a recipient
2592 verification with use_sender or use_postmaster set, because in that case
2593 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2594 case of MAIL FROM:<> rejection from other early rejections (e.g.
2595 rejection of HELO). When verifying a recipient using a non-null MAIL
2596 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2597 Whatever the result of the callout, the value of the domain cache is
2598 left unchanged (for any other kind of callout, getting as far as trying
2599 RCPT means that the domain itself is ok).
2601 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2604 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2605 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2606 session, the daemon ignores ECONNECT errors and logs others; it now
2607 ignores EPIPE as well.
2609 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2610 (quoted-printable decoding).
2612 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2613 later the small subsequent patch to fix an introduced bug.
2615 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2617 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2619 PH/08 An error is now given if message_size_limit is specified negative.
2621 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2622 to be given (somewhat) arbitrary names.
2624 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2627 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2628 miscellaneous code fixes
2630 PH/10 Added the log_reject_target ACL modifier to specify where to log
2633 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2634 hostname. This is wrong, because it relates to the incoming message (and
2635 probably the interface on which it is arriving) and not to the outgoing
2636 callout (which could be using a different interface). This has been
2637 changed to use the value of the helo_data option from the smtp transport
2638 instead - this is what is used when a message is actually being sent. If
2639 there is no remote transport (possible with a router that sets up host
2640 addresses), $smtp_active_hostname is used.
2642 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2643 tweaks were necessary in order to get it to work (see also 21 below):
2644 (a) The code assumed that strncpy() returns a negative number on buffer
2645 overflow, which isn't the case. Replaced with Exim's string_format()
2647 (b) There were several signed/unsigned issues. I just did the minimum
2648 hacking in of casts. There is scope for a larger refactoring.
2649 (c) The code used strcasecmp() which is not a standard C function.
2650 Replaced with Exim's strcmpic() function.
2651 (d) The code set only $1; it now sets $auth1 as well.
2652 (e) A simple test gave the error "authentication client didn't specify
2653 service in request". It would seem that Dovecot has changed its
2654 interface. Fortunately there's a specification; I followed it and
2655 changed what the client sends and it appears to be working now.
2657 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2660 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2661 address A is aliased to B and C, where B exists and C does not. Without
2662 -v the output is "A verified" because verification stops after a
2663 successful redirection if more than one address is generated. However,
2664 with -v the child addresses are also verified. Exim was outputting "A
2665 failed to verify" and then showing the successful verification for C,
2666 with its parentage. It now outputs "B failed to verify", showing B's
2667 parentage before showing the successful verification of C.
2669 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2670 look up a TXT record in a specific list after matching in a combined
2673 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2674 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2675 they consult the DNS. I had assumed they would set it the way they
2676 wanted; and indeed my experiments on Linux seem to show that in some
2677 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2678 To be on the safe side, however, I have now made the interface to
2679 host_find_byname() similar to host_find_bydns(), with an argument
2680 containing the DNS resolver options. The host_find_byname() function now
2681 sets these options at its start, just as host_find_bydns() does. The smtp
2682 transport options dns_qualify_single and dns_search_parents are passed to
2683 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2684 of host_find_byname() use the default settings of RES_DEFNAMES
2685 (qualify_single) but not RES_DNSRCH (search_parents).
2687 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2688 spool_read_header() do less string testing, by means of a preliminary
2689 switch on the second character of optional "-foo" lines. (This is
2690 overdue, caused by the large number of possibilities that now exist.
2691 Originally there were few.) While I was there, I also converted the
2692 str(n)cmp tests so they don't re-test the leading "-" and the first
2693 character, in the hope this might squeeze out yet more improvement.
2695 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2696 flag allowing group syntax was set by the header_syntax check but not
2697 turned off, possible causing trouble later; (2) The flag was not being
2698 set at all for the header_verify test, causing "group"-style headers to
2699 be rejected. I have now set it in this case, and also caused header_
2700 verify to ignore an empty address taken from a group. While doing this, I
2701 came across some other cases where the code for allowing group syntax
2702 while scanning a header line wasn't quite right (mostly, not resetting
2703 the flag correctly in the right place). These bugs could have caused
2704 trouble for malformed header lines. I hope it is now all correct.
2706 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2707 with the "reply" argument non-NULL. The code, however (which originally
2708 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2709 but it didn't always do it. This confused somebody who was copying the
2710 code for some other use. I have removed all the tests.
2712 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2713 feature that was used to support insecure browsers during the U.S. crypto
2714 embargo. It requires special client support, and Exim is probably the
2715 only MTA that supported it -- and would never use it because real RSA is
2716 always available. This code has been removed, because it had the bad
2717 effect of slowing Exim down by computing (never used) parameters for the
2718 RSA_EXPORT functionality.
2720 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2721 authenticator to fail if there's a tab character in the incoming data
2722 (there should never be unless someone is messing about, as it's supposed
2723 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2724 if the connection is using TLS or if the remote IP is the same as the
2725 local IP, and the "valid-client-cert option" if a client certificate has
2728 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2729 authenticators. This can be used for authorization after authentication
2730 succeeds. (In the case of plaintext, it servers for both authentication
2733 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2734 if any retry times were supplied.
2736 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2737 connection, where there is no client IP address to check. In this
2738 situation, the verify now always succeeds.
2740 PH/25 Applied John Jetmore's -Mset patch.
2742 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2744 PH/27 In a string expansion for a processed (not raw) header when multiple
2745 headers of the same name were present, leading whitespace was being
2746 removed from all of them, but trailing whitespace was being removed only
2747 from the last one. Now trailing whitespace is removed from each header
2748 before concatenation. Completely empty headers in a concatenation (as
2749 before) are ignored.
2751 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2752 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2754 PH/29 [Removed. This was a change that I later backed out, and forgot to
2755 correct the ChangeLog entry (that I had efficiently created) before
2756 committing the later change.]
2758 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2759 address errors (4xx response to RCPT) over the same connection as other
2760 messages routed to the same hosts. Such deliveries are always "forced",
2761 so retry times are not inspected. This resulted in far too many retries
2762 for the affected addresses. The effect occurred only when there were more
2763 hosts than the hosts_max_try setting in the smtp transport when it had
2764 the 4xx errors. Those hosts that it had tried were not added to the list
2765 of hosts for which the message was waiting, so if all were tried, there
2766 was no problem. Two fixes have been applied:
2768 (i) If there are any address or message errors in an SMTP delivery, none
2769 of the hosts (tried or untried) are now added to the list of hosts
2770 for which the message is waiting, so the message should not be a
2771 candidate for sending over the same connection that was used for a
2772 successful delivery of some other message. This seems entirely
2773 reasonable: after all the message is NOT "waiting for some host".
2774 This is so "obvious" that I'm not sure why it wasn't done
2775 previously. Hope I haven't missed anything, but it can't do any
2776 harm, as the worst effect is to miss an optimization.
2778 (ii) If, despite (i), such a delivery is accidentally attempted, the
2779 routing retry time is respected, so at least it doesn't keep
2780 hammering the server.
2782 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2783 in ${readsocket because some servers need this prod.
2785 PH/32 Added some extra debug output when updating a wait-xxx database.
2787 PH/33 The hint "could be header name not terminated by colon", which has been
2788 given for certain expansion errors for a long time, was not being given
2789 for the ${if def:h_colon_omitted{... case.
2791 PH/34 The spec says: "With one important exception, whenever a domain list is
2792 being scanned, $domain contains the subject domain." There was at least
2793 one case where this was not true.
2795 PH/35 The error "getsockname() failed: connection reset by peer" was being
2796 written to the panic log as well as the main log, but it isn't really
2797 panic-worthy as it just means the connection died rather early on. I have
2798 removed the panic log writing for the ECONNRESET error when getsockname()
2801 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2802 runs only) independently of the message's sender address. This meant
2803 that, if the 4xx error was in fact related to the sender, a different
2804 message to the same recipient with a different sender could confuse
2805 things. In particular, this can happen when sending to a greylisting
2806 server, but other circumstances could also provoke similar problems.
2807 I have changed the default so that the retry time for these errors is now
2808 based a combination of the sender and recipient addresses. This change
2809 can be overridden by setting address_retry_include_sender=false in the
2812 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2813 remote server are returned as part of bounce messages. This was not
2814 happening for LMTP over a pipe (the lmtp transport), but now it is the
2815 same for both kinds of LMTP.
2817 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2818 in header lines that were in fact CNAMEs. This is no longer the case.
2820 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2821 and queue runs started by the daemon processed all messages. This has
2822 been fixed so that -R and -S can now usefully be given with -q<time>.
2824 PH/40 Import PCRE release 6.7 (fixes some bugs).
2826 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2828 PH/42 Give an error if -q is specified more than once.
2830 PH/43 Renamed the variables $interface_address and $interface_port as
2831 $received_ip_address and $received_port, to make it clear that these
2832 values apply to message reception, and not to the outgoing interface when
2833 a message is delivered. (The old names remain recognized, of course.)
2835 PH/44 There was no timeout on the connect() call when using a Unix domain
2836 socket in the ${readsocket expansion. There now is.
2838 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2839 be meaningful with "accept".
2841 SC/01 Eximstats V1.43
2842 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2844 SC/02 Eximstats V1.44
2845 Use a glob alias rather than an array ref in the generated
2846 parser. This improves both readability and performance.
2848 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2849 Collect SpamAssassin and rejection statistics.
2850 Don't display local sender or destination tables unless
2851 there is data to show.
2852 Added average volumes into the top table text output.
2854 SC/04 Eximstats V1.46
2855 Collect data on the number of addresses (recipients)
2856 as well as the number of messages.
2858 SC/05 Eximstats V1.47
2859 Added 'Message too big' to the list of mail rejection
2860 reasons (thanks to Marco Gaiarin).
2862 SC/06 Eximstats V1.48
2863 Mainlog lines which have GMT offsets and are too short to
2864 have a flag are now skipped.
2866 SC/07 Eximstats V1.49 (Alain Williams)
2867 Added the -emptyok flag.
2869 SC/08 Eximstats V1.50
2870 Fixes for obtaining the IP address from reject messages.
2872 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2873 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2874 whitespace changes from 4.64-PH/27
2876 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2879 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2880 are found, allow negative numbers in numeric criteria)
2882 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2884 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2885 to match changes made in 4.64-PH/43
2887 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2889 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2890 "domain" is actually an IP address, and force a failure. This locks out
2891 those revolvers/nameservers that support "A-for-A" lookups, in
2892 contravention of the specifications.
2894 PH/48 When a host name was looked up from an IP address, and the subsequent
2895 forward lookup of the name timed out, the host name was left in
2896 $sender_host_name, contrary to the specification.
2898 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2899 restricted to single-key lookups, Exim was not diagnosing an error if
2900 * or *@ was used with a query-style lookup.
2902 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2904 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2905 long ago, but noone interested enough thought of it. Let's just say that
2906 the "1.1" means that there are some new functions that weren't there at
2907 some point in the past.
2909 PH/51 Error processing for expansion failure of helo_data from an smtp
2910 transport during callout processing was broken.
2912 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2913 tested/used via the -bh/-bhc/-bs options.
2915 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2916 bug, fixed in subsequent PCRE releases).
2918 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2919 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2921 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2927 SC/01 Use a glob alias rather than an array ref in eximstats generated
2928 parser. This improves both readability and performance.
2930 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2931 Don't display local sender or destination tables in eximstats unless
2932 there is data to show.
2933 Added average volumes into the eximstats top table text output.
2935 SC/03 Collect data on the number of addresses (recipients) as well
2936 as the number of messages in eximstats.
2938 TF/01 Correct an error in the documentation for the redirect router. Exim
2939 does (usually) call initgroups() when daemonizing.
2941 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2942 with consistent privilege compared to when running as a daemon.
2944 TF/03 Note in the spec that $authenticated_id is not set for local
2945 submissions from trusted users.
2947 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2948 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2950 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2951 by adding some example configuration directives to the default
2952 configuration file. A little bit of work is required to uncomment the
2953 directives and define how usernames and passwords are checked, but
2954 there is now a framework to start from.
2956 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2957 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2958 without this. I don't know how relevant this is to other LDAP libraries.
2960 PH/02 Add the verb name to the "unknown ACL verb" error.
2962 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2964 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2966 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2967 directory not expanded when it should be if an expanded home directory
2968 was set for the address (which is overridden by the transport).
2970 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2973 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2974 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2975 because it is too late at that time, and has no effect.
2977 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2978 security issue with \' (bugzilla #107). I could not use the
2979 PQescapeStringConn() function, because it needs a PGconn value as one of
2982 PH/08 When testing addresses using -bt, indicate those final addresses that
2983 are duplicates that would not cause an additional delivery. At least one
2984 person was confused, thinking that -bt output corresponded to deliveries.
2985 (Suppressing duplicates isn't a good idea as you lose the information
2986 about possibly different redirections that led to the duplicates.)
2988 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2989 systems where poll() doesn't work, in particular OS X.
2991 PH/10 Added more information to debugging output for retry time not reached.
2993 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2994 operations in malware.c.
2996 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2999 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3000 syslog_duplication set false, log lines that would normally be written
3001 both the the main log and to the reject log were not written to syslog at
3004 PH/14 In the default configuration, change the use of "message" in ACL warn
3005 statements to "add_header".
3007 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3008 not followed by a command (e.g. "seen endif").
3010 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3011 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3014 PH/17 Added extra conditions to the default value of delay_warning_condition
3018 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3019 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3020 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3023 The Auto-Submitted: and various List- headers are standardised, whereas I
3024 don't think Precedence: ever was.
3026 PH/18 Refactored debugging code in route_finduser() to show more information,
3027 in particular, the error code if getpwnam() issues one.
3029 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3030 This is apparently needed in addition to the PH/07 change above to avoid
3031 any possible encoding problems.
3033 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3034 but not after initializing Perl.
3036 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3037 output them only if debugging. By default they are written stderr,
3038 apparently, which is not desirable.
3040 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3043 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3046 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3048 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3049 authenticated or an ident call has been made. Suppress the default
3050 values for $authenticated_id and $authenticated_sender (but permit -oMai
3051 and -oMas) when testing with -bh.
3053 PH/24 Re-jigged the order of the tests in the default configuration so that the
3054 tests for valid domains and recipients precede the DNS black list and CSA
3055 tests, on the grounds that those ones are more expensive.
3057 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3058 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3059 This bug exists in every version of Exim that I still have, right back to
3062 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3063 However, an attempt to turn on case-sensitivity in a regex key by
3064 including (?-i) didn't work because the subject string was already
3065 lowercased, and the effects were non-intuitive. It turns out that a
3066 one-line patch can be used to allow (?-i) to work as expected.
3072 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3073 other effects) broke the use of negated acl sub-conditions.
3075 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3078 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3079 "Deny" causes Exim to reject the incoming connection with a 554 error.
3080 Unfortunately, if there is a major crisis, such as a disk failure,
3081 tcp-wrappers gives "deny", whereas what one would like would be some
3082 kind of temporary error. A kludge has been added to help with this.
3083 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3084 554 error is used if errno is still zero or contains ENOENT (which occurs
3085 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3088 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3090 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3091 errors. Otherwise a message that provokes a temporary error (when other
3092 messages do not) can cause a whole host to time out.
3094 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3095 addresses were routed directly to files or pipes from a redirect router.
3096 File deliveries just didn't batch; pipe deliveries might have suffered
3099 PH/06 A failure to get a lock for a hints database would erroneously always say
3100 "Failed to get write lock", even when it was really a read lock.
3102 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3103 of 0600. This has been changed to use the value of the lockfile_mode
3104 option (which defaults to 0600).
3106 PH/08 Applied small patch from the Sieve maintainer.
3108 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3109 folder from quota calculations, a direct delivery into this folder messed
3110 up the contents of the maildirsize file. This was because the regex was
3111 used only to exclude .Trash (or whatever) when the size of the mailbox
3112 was calculated. There was no check that a delivery was happening into an
3113 excluded directory. This bug has been fixed by ignoring all quota
3114 processing for deliveries into excluded directories.
3116 PH/10 Added the maildirfolder_create_regex option to appendfile.
3122 PH/01 The code for finding all the local interface addresses on a FreeBSD
3123 system running IPv6 was broken. This may well have applied to all BSD
3124 systems, as well as to others that have similar system calls. The broken
3125 code found IPv4 interfaces correctly, but gave incorrect values for the
3126 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3127 that it would not match correctly against @[] and not recognize the IPv6
3130 PH/02 The ipliteral router was not recognizing addresses of the form user@
3131 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3133 PH/03 Added disable_ipv6.
3135 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3136 decoded form, because it is most often used to construct To: headers
3137 lines in autoreplies, and the decoded form may well be syntactically
3138 invalid. However, $reply_address has leading white space removed, and all
3139 newlines turned into spaces so that the autoreply transport does not
3142 PH/05 If group was specified without a user on a router, and no group or user
3143 was specified on a transport, the group from the router was ignored.
3145 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3146 for visible compile-time settings that can be used to change these
3147 numbers, for those that want even more. Backwards compatibility with old
3148 spool files has been maintained. However, going back to a previous Exim
3149 release will lost any variables that are in spool files.
3151 PH/07 Two small changes when running in the test harness: increase delay when
3152 passing a TCP/IP connection to a new process, in case the original
3153 process has to generate a bounce, and remove special handling of
3154 127.0.0.2 (sic), which is no longer necessary.
3156 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3157 be the same on different OS.
3159 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3162 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3163 whether --show-vars was specified or not
3165 JJ/02 exipick: Added support for new ACL variable spool format introduced
3168 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3169 syntactically invalid From: or Reply-to: line, and a filter used this to
3170 generate an autoreply, and therefore failed to obtain an address for the
3171 autoreply, Exim could try to deliver to a non-existent relative file
3172 name, causing unrelated and misleading errors. What now happens is that
3173 it logs this as a hard delivery error, but does not attempt to create a
3176 PH/11 The exinext utility has a -C option for testing purposes, but although
3177 the given file was scanned by exinext itself; it wasn't being passed on
3178 when Exim was called.
3180 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3181 an end-of-file indication when reading a command response.
3183 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3184 compiled. In many other places in Exim, IPv6 addresses are always
3185 recognized, so I have changed this. It also means that IPv4 domain
3186 literals of the form [IPV4:n.n.n.n] are now always recognized.
3188 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3189 used if the router is not running as root, for example, when verifying at
3190 ACL time, or when using -bh. The debugging output from this situation was
3191 non-existent - all you got was a failure to exec. I have made two
3194 (a) Failures to set uid/gid, the current directory, or a process leader
3195 in a subprocess such as that created by queryprogram now generate
3196 suitable debugging output when -d is set.
3198 (b) The queryprogram router detects when it is not running as root,
3199 outputs suitable debugging information if -d is set, and then runs
3200 the subprocess without attempting to change uid/gid.
3202 PH/15 Minor change to Makefile for building test_host (undocumented testing
3205 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3206 additional section of a DNS packet that returns MX or SRV records.
3207 Instead, it always explicitly searches for A/AAAA records. This avoids
3208 major problems that occur when a DNS server includes only records of one
3209 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3210 fixed another bug: if SRV records were looked up and the corresponding
3211 address records were *not* found in the additional section, the port
3212 values from the SRV records were lost.
3214 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3215 using the correct key (the original address) when searching the retry
3216 rules in order to find which one to use for generating the retry hint.
3218 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3219 adding the default one. Similarly, if it contains a Reply-To: header, the
3220 errors_reply_to option, if set, is not used.
3222 PH/19 When calculating a retry time, Exim used to measure the "time since
3223 failure" by looking at the "first failed" field in the retry record. Now
3224 it does not use this if it is later than than the arrival time of the
3225 message. Instead it uses the arrival time. This makes for better
3226 behaviour in cases where some deliveries succeed, thus re-setting the
3227 "first failed" field. An example is a quota failure for a huge message
3228 when small messages continue to be delivered. Without this change, the
3229 "time since failure" will always be short, possible causing more frequent
3230 delivery attempts for the huge message than are intended.
3231 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3233 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3234 $1, $2, $3) because the numerical variables can be reset during some
3235 expansion items (e.g. "match"), thereby losing the authentication data.
3237 PH/21 Make -bV show the size of off_t variables so that the test suite can
3238 decide whether to run tests for quotas > 2G.
3240 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3241 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3242 filecount value is greater than 2G or if a quota value is greater than 2G
3243 on a system where the size of off_t is not greater than 4, a panic error
3246 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3247 never match. The debug and -bh output now contains an explicit error
3248 message indicating a malformed IPv4 address or mask.
3250 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3251 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3252 PH/23 above applies.
3254 PH/25 Do not write to syslog when running in the test harness. The only
3255 occasion when this arises is a failure to open the main or panic logs
3256 (for which there is an explicit test).
3258 PH/26 Added the /no_tell option to "control=freeze".
3260 PH/27 If a host name lookup failed very early in a connection, for example, if
3261 the IP address matched host_lookup and the reverse lookup yielded a name
3262 that did not have a forward lookup, an error message of the form "no IP
3263 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3264 could be logged. Now it outputs the IP address instead of "NULL".
3266 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3267 allows the sender and the authenticated sender to be set when
3268 submitting a message from within Exim. Since child_open_exim() is
3269 documented for local_scan(), the new function should be too.
3271 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3272 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3273 results in an empty string is now treated as unset.
3275 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3277 PH/31 Added sender_verify_fail logging option.
3279 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3280 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3281 tidied the source and removed it altogether.
3283 PH/33 When a queue run was abandoned because the load average was too high, a
3284 log line was always written; now it is written only if the queue_run log
3285 selector is set. In addition, the log line for abandonment now contains
3286 information about the queue run such as the pid. This is always present
3287 in "start" and "stop" lines but was omitted from the "abandon" line.
3289 PH/34 Omit spaces between a header name and the colon in the error message that
3290 is given when verify = headers_syntax fails (if there are lots of them,
3291 the message gets confusing).
3293 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3294 names, as there are now some PTR records that contain slashes. This check
3295 is only to protect against broken name servers that fall over on strange
3296 characters, so the fact that it applies to all lookups doesn't matter.
3298 PH/36 Now that the new test suite is complete, we can remove some of the
3299 special code in Exim that was needed for the old test suite. For example,
3300 sorting DNS records because real resolvers return them in an arbitrary
3301 order. The new test suite's fake resolver always returns records in the
3304 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3305 bounces) except when queue_only is set, to avoid logging races between
3306 the different processes.
3308 PH/38 Panic-die if .include specifies a non-absolute path.
3310 PH/39 A tweak to the "H" retry rule from its user.
3312 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3313 a label. They prevented compilation on older perls.
3315 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3316 a warning to be raised on newish perls.
3318 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3319 on queue. Changes to match documented behaviour of showing count of
3320 messages matching specified criteria.
3322 PH/40 Changed the default ident timeout from 30s to 5s.
3324 PH/41 Added support for the use of login_cap features, on those BSD systems
3325 that have them, for controlling the resources used by pipe deliveries.
3327 PH/42 The content-scanning code uses fopen() to create files in which to put
3328 message data. Previously it was not paying any attention to the mode of
3329 the files. Exim runs with umask(0) because the rest of the code creates
3330 files with open(), and sets the required mode explicitly. Thus, these
3331 files were ending up world-writeable. This was not a big issue, because,
3332 being within the spool directory, they were not world-accessible. I have
3333 created a function called modefopen, which takes an additional mode
3334 argument. It sets umask(777), creates the file, chmods it to the required
3335 mode, then resets the umask. All the relevant calls to fopen() in the
3336 content scanning code have been changed to use this function.
3338 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3339 to 24 hours. This avoids potential overflow problems when processing G
3340 and H retry rules. I suspect nobody ever tinkers with this value.
3342 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3344 PH/45 When the plaintext authenticator is running as a client, the server's
3345 challenges are checked to ensure they are valid base64 strings. By
3346 default, the authentication attempt is cancelled if an invalid string is
3347 received. Setting client_ignore_invalid_base64 true ignores these errors.
3348 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3349 they are received. Thus, the responses can be made to depend on the
3350 challenges. If an invalid string is ignored, an empty string is placed in
3353 PH/46 Messages that are created by the autoreply transport now contains a
3354 References: header, in accordance with RFCs 2822 and 3834.
3356 PH/47 Added authenticated_sender_force to the smtp transport.
3358 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3360 PH/49 Installed latest patch from the Sieve maintainer.
3362 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3363 was also set, the appendfile transport was unnecessarily scanning a
3364 directory of message files (e.g. for maildir delivery) to find the count
3365 of files (along with the size), even though it did not need this
3366 information. It now does the scan only if it needs to find either the
3367 size of the count of files.
3369 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3371 PH/52 Two bugs concerned with error handling when the smtp transport is
3374 (i) Exim was not creating retry information for temporary errors given
3375 for individual recipients after the DATA command when the smtp transport
3376 was used in LMTP mode. This meant that they could be retried too
3377 frequently, and not timed out correctly.
3379 (ii) Exim was setting the flag that allows error details to be returned
3380 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3381 recipients that were returned after the DATA command.
3383 PH/53 This is related to PH/52, but is more general: for any failing address,
3384 when detailed error information was permitted to be returned to the
3385 sender, but the error was temporary, then after the final timeout, only
3386 "retry timeout exceeded" was returned. Now it returns the full error as
3387 well as "retry timeout exceeded".
3389 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3390 do this, and (what is worse) MTAs that accept it.
3392 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3393 will now be deprecated.
3395 PH/56 New os.c-cygwin from the Cygwin maintainer.
3397 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3398 formats (previously only available in exim formats via -bpr, -bpru,
3399 and -bpra. Now also available in native and exiqgrep formats)
3401 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3402 with very large, slow to parse queues
3404 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3406 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3408 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3409 responses to authentication challenges, though it was showing the
3410 challenges; (ii) I've removed the CR characters from the debug output for
3413 PH/58 Allow for the insertion of a newline as well as a space when a string
3414 is turned into more than one encoded-word during RFC 2047 encoding. The
3415 Sieve code now uses this.
3417 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3418 data_4xx, lost_connection, tls_required.
3420 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3421 message was being sent as an SMTP response.
3423 PH/61 Add -l and -k options to exicyclog.
3425 PH/62 When verifying, if an address was redirected to one new address, so that
3426 verification continued, and the new address failed or deferred after
3427 having set something in $address_data, the value of $address_data was not
3428 passed back to the ACL. This was different to the case when no
3429 redirection occurred. The value is now passed back in both cases.
3431 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3432 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3433 use login_cap.h, so on its own it isn't the distinguishing feature. The
3434 new name refers directly to the setclassresources() function.
3436 PH/65 Added configuration files for NetBSD3.
3438 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3440 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3441 is preferred over IPv4.
3443 PH/68 The bounce_return_message and bounce_return_body options were not being
3444 honoured for bounces generated during the reception of non-SMTP messages.
3445 In particular, this applied to messages rejected by the ACL. This bug has
3446 been fixed. However, if bounce_return_message is true and bounce_return_
3447 body is false, the headers that are returned for a non-SMTP message
3448 include only those that have been read before the error was detected.
3449 (In the case of an ACL rejection, they have all been read.)
3451 PH/69 The HTML version of the specification is now built in a directory called
3452 spec_html instead of spec.html, because the latter looks like a path with
3453 a MIME-type, and this confuses some software.
3455 PH/70 Catch two compiler warnings in sieve.c.
3457 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3458 function verify_get_ident() calls ip_connect() to connect a socket, but
3459 if the "connect()" function timed out, ip_connect() used to close the
3460 socket. However, verify_get_ident() also closes the socket later, and in
3461 between Exim writes to the log, which may get opened at this point. When
3462 the socket was closed in ip_connect(), the log could get the same file
3463 descriptor number as the socket. This naturally causes chaos. The fix is
3464 not to close the socket in ip_connect(); the socket should be closed by
3465 the function that creates it. There was only one place in the code where
3466 this was missing, in the iplookup router, which I don't think anybody now
3467 uses, but I've fixed it anyway.
3469 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3470 well as to direct DNS lookups. Otherwise the handling of names in host
3471 lists is inconsistent and therefore confusing.
3477 PH/01 Two changes to the default runtime configuration:
3479 (1) Move the checks for relay_from_hosts and authenticated clients from
3480 after to before the (commented out) DNS black list checks.
3482 (2) Add control=submission to the relay_from_hosts and authenticated
3483 clients checks, on the grounds that messages accepted by these
3484 statements are most likely to be submissions.
3486 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3488 (1) Generate an error if the third argument for the ${prvs expansion is
3491 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3494 (3) Reset the variables that are obtained from the first argument of
3495 ${prvscheck and used in the second argument before leaving the code,
3496 because their memory is reclaimed, so using them afterwards may do
3499 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3500 one (it's much easier than Tom thought :-).
3502 (5) Because of (4), we can now allow for the use of $prvscheck_result
3503 inside the third argument.
3505 PH/03 For some reason, the default setting of PATH when running a command from
3506 a pipe transport was just "/usr/bin". I have changed it to
3509 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3510 anything to be listed in the output from -bV.
3512 PH/05 When a filter generated an autoreply, the entire To: header line was
3513 quoted in the delivery log line, like this:
3515 => >A.N.Other <ano@some.domain> <original@ddress> ...
3517 This has been changed so that it extracts the operative address. There
3518 may be more than one such address. If so, they are comma-separated, like
3521 => >ano@some.domain,ona@other.domain <original@ddress> ...
3523 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3524 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3525 not being looked up in the rDNS to get a host name, Exim was showing the
3526 IP address twice in Received: lines, even though the IP addresses were
3527 identical. For example:
3529 Received: from [1.2.3.4] (helo=[1.2.3.4])
3531 However, if the real host name was known, it was omitting the HELO data
3532 if it matched the actual IP address. This has been tidied up so that it
3533 doesn't show the same IP address twice.
3535 PH/07 When both +timestamp and +memory debugging was on, the value given by
3536 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3537 called by the debug printing, thereby overwriting the timestamp buffer.
3538 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3540 PH/08 When the original message was included in an autoreply transport, it
3541 always said "this is a copy of the message, including all the headers",
3542 even if body_only or headers_only was set. It now gives an appropriate
3545 PH/09 Applied a patch from the Sieve maintainer which:
3547 o fixes some comments
3548 o adds the (disabled) notify extension core
3549 o adds some debug output for the result of if/elsif tests
3550 o points to the current vacation draft in the documentation
3551 and documents the missing references header update
3555 o fixes a bug in processing the envelope test (when testing
3556 multiple envelope elements, the last element determined the
3559 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3560 Electronic Mail") by including:
3562 Auto-submitted: auto-generated
3564 in the messages that it generates (bounce messages and others, such as
3565 warnings). In the case of bounce messages for non-SMTP messages, there was
3566 also a typo: it was using "Auto_submitted" (underscore instead of
3567 hyphen). Since every message generated by Exim is necessarily in response
3568 to another message, thes have all been changed to:
3570 Auto-Submitted: auto-replied
3572 in accordance with these statements in the RFC:
3574 The auto-replied keyword:
3576 - SHOULD be used on messages sent in direct response to another
3577 message by an automatic process,
3579 - MUST NOT be used on manually-generated messages,
3581 - MAY be used on Delivery Status Notifications (DSNs) and Message
3582 Disposition Notifications (MDNs),
3584 - MUST NOT be used on messages generated by automatic or periodic
3585 processes, except for messages which are automatic responses to
3588 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3589 to the default Received: header definition.
3591 PH/12 Added log selector acl_warn_skipped (default on).
3593 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3594 variables because (a) they are in the wrong storage pool and (b) even if
3595 they were copied, it wouldn't work properly because of the caching.
3597 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3598 checking when decoding. Apparently there are clients that generate
3599 overlong encoded strings. Why am I not surprised?
3601 PH/15 If the first argument of "${if match_address" was not empty, but did not
3602 contain an "@" character, Exim crashed. Now it writes a panic log message
3603 and treats the condition as false.
3605 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3607 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3608 extension "envelope-auth". The code is finished and in agreement with
3609 other implementations, but there is no documentation so far and in fact,
3610 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3611 not changing the active code.
3613 Print executed "if" and "elsif" statements when debugging is used. This
3614 helps a great deal to understand what a filter does.
3616 Document more things not specified clearly in RFC3028. I had all this
3617 sorted out, when out of a sudden new issues came to my mind. Oops."
3619 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3622 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3623 canonical form (as documented). However, after a host name lookup from
3624 the IP address, check_host() was doing a simple string comparison with
3625 addresses acquired from the DNS when checking that the found name did
3626 have the original IP as one of its addresses. Since any found IPv6
3627 addresses are likely to be in abbreviated form, the comparison could
3628 fail. Luckily, there already exists a function for doing the comparison
3629 by converting both addresses to binary, so now that is used instead of
3630 the text comparison.
3632 PH/20 There was another similar case to PH/19, when a complete host name was
3633 given in a host list; looking up its IP address could give an abbreviated
3634 form, whereas the current host's name might or might not be abbreviated.
3635 The same fix has been applied.
3641 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3642 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3645 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3646 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3648 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3650 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3651 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3652 Also, for "Auto-Submitted", treat anything other than "no" as
3653 non-personal, in accordance with RFC 3834. (Previously it treated
3654 anything starting "auto-" as non-personal.)
3656 TF/01 The control=submission/name=... option had a problem with syntax
3657 errors if the name included a slash character. The /name= option
3658 now slurps the rest of the string, so it can include any characters
3659 but it must come last in the list of options (after /sender_retain
3662 PH/05 Some modifications to the interface to the fake nameserver for the new
3670 TK/01 Added the "success_on_redirect" address verification option. See
3671 NewStuff for rationale and an example.
3673 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3675 PH/02 Patch to exigrep to allow it to work on syslog lines.
3677 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3678 fread() to skip over the body file's header line, because in Cygwin the
3679 header line is locked and is inaccessible.
3681 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3682 co-exist for some time) to make it clear that it is the Exim ID that is
3683 referenced, not the Message-ID: header line.
3685 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3686 string_format() function, because snprintf() does not exist on all
3689 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3690 ${stat: expansion item.
3692 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3693 protocol synchronization error", to keep the pedants happy.
3695 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3696 well as for IRIX systems, when gcc is being used. See the host.c source
3699 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3701 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3704 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3705 transport and to the smtp transport in LMTP mode.
3707 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3709 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3710 run a filter in a subprocess. This could lead to confusion in subsequent
3711 lookups in the parent process. There should also be a search_tidyup() at
3712 the end of the subprocess.
3714 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3715 only if the host matched helo_try_verify_hosts, which caused the
3716 verification to occur when the EHLO/HELO command was issued. The ACL just
3717 tested the remembered result. Now, if a previous verification attempt has
3718 not happened, "verify = helo" does it there and then.
3720 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3722 TK/03 Fix log output including CR from clamd.
3724 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3725 exist provoked a memory error which could cause a segfault.
3727 PH/15 Installed PCRE 6.2
3729 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3731 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3732 of the problem. Specifically, suggested +O2 rather than +O1 for the
3735 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3737 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3738 record, the port was not correctly logged when the outgoing_port log
3739 selector was set (it logged the transort's default port).
3741 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3742 fallback_hosts, and "hosts" in the smtp transport.
3744 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3745 host errors such as "Connection refused".
3747 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3748 authentication with radiusclient 0.4.9:
3750 - Error returned from rc_read_config was caught wrongly
3751 - Username/password not passed on to radius server due to wrong length.
3753 The presumption is that some radiusclient API changes for 4.51/PH/17
3754 were not taken care of correctly. The code is still untested by me (my
3755 Linux distribution still has 0.3.2 of radiusclient), but it was
3756 contributed by a Radius user.
3758 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3759 expanding the "port" option of the smtp transport.
3761 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3762 while reading a MIME header. Thanks to Tom Hughes for a patch.
3764 PH/24 Include config.h inside local_scan.h so that configuration settings are
3767 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3768 that in an ACL for RCPT (for example), you can examine exactly what was
3771 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3772 commands, but it was not correctly comparing the address with the actual
3773 client host address. Thus, it would show the EHLO address in Received:
3774 header lines when this was not necessary.
3776 PH/27 Added the % operator to ${eval:}.
3778 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3779 it should be ignoring failures (because with -C, for example, it has lost
3780 privilege). It wasn't ignoring creation failures other than "already
3783 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3786 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3787 because an input line was too long, either on its own, or by virtue of
3788 too many continuations, the temporary file was not being removed, and the
3789 return code was incorrect.
3791 PH/31 Missing "BOOL" in function definition in filtertest.c.
3793 PH/32 Applied Sieve patches from the maintainer.
3795 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3797 PH/33 Added "verify = not_blind".
3799 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3800 Local/Makefile (with some defaults set). These are used in built scripts
3801 such as exicyclog, but they have never been used in the exim_install
3802 script (though there are many overriding facilities there). I have
3803 arranged that the exim_install script now takes note of these two
3806 PH/35 Installed configuration files for Dragonfly.
3808 PH/36 When a locally submitted message by a trusted user did not contain a
3809 From: header, and the sender address was obtained from -f or from an SMTP
3810 MAIL command, and the trusted user did not use -F to supply a sender
3811 name, $originator_name was incorrectly used when constructing a From:
3812 header. Furthermore, $originator_name was used for submission mode
3813 messages from external hosts without From: headers in a similar way,
3814 which is clearly wrong.
3816 PH/37 Added control=suppress_local_fixups.
3818 PH/38 When log_selector = +received_sender was set, and the addition of the
3819 sender made the log line's construction buffer exactly full, or one byte
3820 less than full, an overflow happened when the terminating "\n" was
3823 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3824 when the result of a list match is failure because a DNS lookup failed.
3826 PH/40 RM_COMMAND is now used in the building process.
3828 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3829 the "build-* directories that it finds.
3831 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3832 address in a domain literal was a prefix of an interface address.
3834 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3835 when verifying a sender address, unless rewrite_headers is false.
3837 PH/44 Wrote a long comment about why errors_to addresses are verified as
3838 recipients, not senders.
3840 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3841 the ratelimit ACL was added.
3843 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3845 PH/46 Added extra information about PostgreSQL errors to the error string.
3847 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3848 avoiding the need to install special zones in a real server. This is
3849 backwards compatible; if it can't find the fake resolver, it drops back.
3850 Thus, both old and new test suites can be run.
3852 TF/02 Added util/ratelimit.pl
3854 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3855 clock is set back in time.
3857 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3858 Candler <B.Candler@pobox.com>.
3860 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3861 OK for addresses that are the result of redirections.
3863 PH/48 A number of further additions for the benefit of the new test suite,
3864 including a fake gethostbyname() that interfaces to the fake DNS resolver
3867 TF/06 The fix for widen_domains has also been applied to qualify_single and
3868 search_parents which are the other dnslookup options that can cause
3871 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3874 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3876 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3877 service" if it encountered an MX record with an empty target hostname.
3878 The message is now "an MX or SRV record indicated no SMTP service".
3880 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3881 if the DNS of the sending site is misconfigured. This is quite a
3882 common situation. This change restores the behaviour of treating a
3883 helo verification defer as a failure.
3885 PH/51 If self=fail was set on a router, the bounce message did not include the
3886 actual error message.
3892 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3894 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3895 command itself ended in error, the underlying message about the transport
3896 filter timeout was being overwritten with the pipe command error. Now the
3897 underlying error message should be appended to the second error message.
3899 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3901 PH/02 Reduce the amount of output that "make" produces by default. Full output
3902 can still be requested.
3904 PH/03 The warning log line about a condition test deferring for a "warn" verb
3905 was being output only once per connection, rather than after each
3906 occurrence (because it was using the same function as for successful
3907 "warn" verbs). This seems wrong, so I have changed it.
3909 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3910 it should not have, which might have caused a crash in the right
3911 circumstances, but probably never did.
3913 PH/04 Installed a modified version of Tony Finch's patch to make submission
3914 mode fix the return path as well as the Sender: header line, and to
3915 add a /name= option so that you can make the user's friendly name appear
3918 TF/03 Added the control = fakedefer ACL modifier.
3920 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3921 Mark Lowes for thorough testing.
3923 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3925 TK/03 Merged latest SRS patch from Miles Wilton.
3927 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3928 with the definition in sysexits.h (which is #included earlier).
3929 Fortunately, Exim does not actually use EX_OK. The code used to try to
3930 preserve the sysexits.h value, by assuming that macro definitions were
3931 scanned for macro replacements. I have been disabused of this notion,
3932 so now the code just undefines EX_OK before #including unistd.h.
3934 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3935 in the smtp transport. When a block could not be written in a single
3936 write() function, the timeout was being re-applied to each part-write.
3937 This seems wrong - if the receiver was accepting one byte at a time it
3938 would take for ever. The timeout is now adjusted when this happens. It
3939 doesn't have to be particularly precise.
3941 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3942 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3944 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3945 without a domain if the check to <postmaster@domain> fails.
3947 SC/01 Eximstats: added -xls and the ability to specify output files
3948 (patch written by Frank Heydlauf).
3950 SC/02 Eximstats: use FileHandles for outputting results.
3952 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3954 SC/04 Eximstats: fixed display of large numbers with -nvr option
3956 SC/05 Eximstats: fixed merging of reports with empty tables.
3958 SC/06 Eximstats: added the -include_original_destination flag
3960 SC/07 Eximstats: removed tabs and trailing whitespace.
3962 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3964 TK/06 MBOX spool code: Add real "From " MBOX separator line
3965 so the .eml file is really in mbox format (even though
3966 most programs do not really care). Patch from Alex Miller.
3968 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3969 The latter is generated from $received_to and is only set if the
3970 message has one envelope recipient. SA can use these headers,
3971 obviously out-of-the-box. Patch from Alex Miller.
3973 PH/08 The ${def test on a variable was returning false if the variable's
3974 value was "0", contrary to what the specification has always said!
3975 The result should be true unless the variable is empty.
3977 PH/09 The syntax error of a character other than { following "${if
3978 def:variable_name" (after optional whitespace) was not being diagnosed.
3979 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3980 accidental colon was present, for example, could give incorrect results.
3982 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3983 result is used (not including appendfile, where other changes are about
3986 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3987 This involved changing a lot of size variables from int to off_t. It
3988 should work with maildirs and everything.
3990 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3991 spamd dying while we are connected to it.
3993 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3996 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3997 function for BDB 4.3.
3999 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4001 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4002 the sources for PCRE have been re-arranged and more files are now
4005 PH/15 The code I had for printing potentially long long variables in PH/11
4006 above was not the best (it lost precision). The length of off_t variables
4007 is now inspected at build time, and an appropriate printing format (%ld
4008 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4009 to be "long long int" or "long int". This is needed for the internal
4010 formatting function string_vformat().
4012 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4013 the configuration file to be ":syslog", then the script "guesses" where
4014 the logs files are, rather than using the compiled in default. In our
4015 case the guess is not the same as the compiled default, so the script
4016 suddenly stopped working when I started to use syslog. The patch checks
4017 to see if log_file_path is "". If so, it attempts to read it from exim
4018 with no configuration file to get the compiled in version, before it
4019 falls back to the previous guessing code."
4021 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4022 implementing BATV in an Exim configuration. See NewStuff for the gory
4025 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4026 Makefile that are specific to HP-UX.
4028 PH/18 If the "use_postmaster" option was set for a recipient callout together
4029 with the "random" option, the postmaster address was used as the MAIL
4030 FROM address for the random test, but not for the subsequent recipient
4031 test. It is now used for both.
4033 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4034 patch removes a few documentation additions to RFC 3028, because the
4035 latest draft now contains them. It adds the new en;ascii-case comparator
4036 and a new error check for 8bit text in MIME parts. Comparator and
4037 require names are now matched exactly. I enabled the subaddress
4038 extension, but it is not well tested yet (read: it works for me)."
4040 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4041 rework some of the code of TK/09 above to avoid the hardwired use of
4042 "%lld" and "long long". Replaced the call to snprintf() with a call to
4045 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4046 records point to non-existent hosts", "retry timeout exceeded", and
4047 "retry time not reached for any host after a long failure period".
4049 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4050 experimental DomainKeys support:
4052 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4053 (2) On an error such as an illegally used "control", the wrong name for
4054 the control was given.
4056 These problems did NOT occur unless DomainKeys support was compiled.
4058 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4060 PH/24 Added ${if match_ip condition.
4062 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4063 databases so that it will be absolutely obvious if a crash occurs in the
4064 DB library. This is a regular occurrence (often caused by mis-matched
4067 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4068 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4069 picked up on a user's system that detects such things. There doesn't seem
4070 to be a gcc warning option for this - only an attribute that has to be
4071 put on the function's prototype. It seems that in Fedora Core 4 they have
4072 set this on a number of new functions. No doubt there will be more in due
4075 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4076 specify a transport. However, if an address that was verified by such a
4077 router was the subject of a callout, Exim crashed because it tried to
4078 read the rcpt_include_affixes from the non-existent transport. Now it
4079 just assumes that the setting of that option is false. This bug was
4080 introduced by 4.51/PH/31.
4082 PH/28 Changed -d+all to exclude +memory, because that information is very
4083 rarely of interest, but it makes the output a lot bigger. People tend to
4084 do -d+all out of habit.
4086 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4087 code in os-type was giving problems when libc.so lives in lib64, like on
4090 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4091 aren't the modern standard, and it seems that some systems' include files
4092 don't always have them. Exim was already checking for some of the newer
4093 ones like T_AAAA, and defining it itself. I've added checks for all the
4094 record types that Exim uses.
4096 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4097 not automatically generating a new one, as it is supposed to. This
4098 prevented TLS from working. If the file did exist, but contained invalid
4099 data, a new version was generated, as expected. It was only the case of a
4100 non-existent file that was broken.
4102 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4103 with a change in libdomainkeys > 0.64.
4105 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4106 from DNS. If the selector record carries the flag, it now has
4107 precedence over the domain-wide flag.
4109 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4111 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4112 the use of a port name that isn't defined in /etc/services) occurred, the
4113 message was deferred as in a normal delivery, and thus remained on the
4114 spool, instead of being failed because of the mua_wrapper setting. This
4115 is now fixed, and I tidied up some of the mua_wrapper messages at the
4118 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4119 the messages in a hash of arrays rather than using individual hashes.
4120 This is a bit cleaner and results in dramatic memory savings, albeit
4121 at a slight CPU cost.
4123 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4124 as requested by Marc Sherman.
4126 SC/10 Eximstats: added histograms for user specified patterns as requested
4129 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4131 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4132 fopen() in the content-scanning modules that did not already have it.
4138 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4139 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4141 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4143 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4145 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4146 to test Sieve filters that use "vacation".
4148 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4149 that changes the way the GnuTLS parameters are stored in the cache file.
4150 The new format can be generated externally. For backward compatibility,
4151 if the data in the cache doesn't make sense, Exim assumes it has read an
4152 old-format file, and it generates new data and writes a new file. This
4153 means that you can't go back to an older release without removing the
4156 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4157 work if there are any delivery delays because "one_time" forces the
4158 parent to be marked "delivered", so its unseen clone is never tried
4159 again. For this reason, Exim now forbids the simultaneous setting of
4162 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4163 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4164 entry if you want to know the details. The fix, however, neglected to
4165 consider the case when local delivery batching is involved. The test for
4166 "previously delivered" was not happening when checking to see if an
4167 address could be batched with a previous (undelivered) one; under
4168 certain circumstances this could lead to multiple deliveries to the same
4171 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4172 in its include files, and this causes problems building Exim.
4174 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4175 header_syntax) but Exim was just ignoring anything given after a slash.
4176 In particular, this caused confusion with an attempt to use "verify =
4177 reverse_host_lookup/defer_ok". An error is now given when options are
4178 supplied for verify items that do not have them. (Maybe reverse_host_
4179 lookup should have a defer_ok option, but that's a different point.)
4181 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4182 defined by RFC 821) to 2048, because there were problems with some AUTH
4183 commands, and RFC 1869 says the size should be increased for extended
4184 SMTP commands that take arguments.
4186 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4189 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4190 "unknown" error; now it says that the functionality isn't in the binary.
4192 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4193 an address' error message when a string expansion fails (syntax or
4194 whatever). Otherwise the password may appear in the log. Following change
4195 PH/42 below, there is no longer a chance of it appearing in a bounce
4198 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4200 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4201 of its addresses was ever tried. (Bugzilla bug #2.)
4203 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4204 the result incorrectly in the debug output. (It correctly added a newline
4205 to what was transported.)
4207 TF/01 Added $received_time.
4209 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4210 commented out examples of how to interface to a virus scanner and to
4211 SpamAssassin. Also added commented examples of av_scanner and
4212 spamd_address settings.
4214 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4215 and controls are allowed in which ACLs. There were a couple of minor
4216 errors. Some of the entries in the conditions table (which is a table of
4217 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4218 negation of where the condition IS allowed.
4220 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4222 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4223 header file does not have a version number, so I've had to invent a new
4224 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4225 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4226 radiusclient), but it was contributed by a Radius user.
4228 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4229 files or directories, for OpenSSL.
4231 PH/19 When an Exim process that is running as root has to create an Exim log
4232 file, it does so in a subprocess that runs as exim:exim so as to get the
4233 ownership right at creation (otherwise, other Exim processes might see
4234 the file with the wrong ownership). There was no test for failure of this
4235 fork() call, which would lead to the process getting stuck as it waited
4236 for a non-existent subprocess. Forks do occasionally fail when resources
4237 run out. I reviewed all the other calls to fork(); they all seem to check
4240 PH/20 When checking for unexpected SMTP input at connect time (before writing
4241 the banner), Exim was not dealing correctly with a non-positive return
4242 from the read() function. If the client had disconnected by this time,
4243 the result was a log entry for a synchronization error with an empty
4244 string after "input=" when read() returned zero. If read() returned -1
4245 (an event I could not check), uninitialized data bytes were printed.
4246 There were reports of junk text (parts of files, etc) appearing after
4249 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4251 PH/22 Added support for macro redefinition, and (re)definition in between
4252 driver and ACL definitions.
4254 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4255 forgetting to use the resulting value; it was using the unexpanded value.
4257 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4258 hadn't been configured. The fix is from Juergen Kreileder, who
4259 understands it better than I do:
4261 "Here's what I see happening with three configured cyrus_sasl
4262 authenticators configured (plain, login, cram-md5):
4264 On startup auth_cyrus_sasl_init() gets called for each of these.
4265 This means three calls to sasl_listmech() without a specified mech_list.
4266 => SASL tests which mechs of all available mechs actually work
4267 => three warnings about OTP not working
4268 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4270 With the patch, sasl_listmech() also gets called three times. But now
4271 SASL's mech_list option is set to the server_mech specified in the the
4272 authenticator. Or in other words, the answer from sasl_listmech()
4273 gets limited to just the mech you're testing for (which is different
4275 => the return list contains just 'plain' or 'login', 'cram-md5' or
4276 nothing depending on the value of ob->server_mech.
4278 I've just tested the patch: Authentication still works fine,
4279 unavailable mechs specified in the exim configuration are still
4280 caught, and the auth.log warnings about OTP are gone."
4282 PH/25 When debugging is enabled, the contents of the command line are added
4283 to the debugging output, even when log_selector=+arguments is not
4286 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4287 answer is "GNU", and only if the return is "GNU/something" is the answer
4290 PH/27 $acl_verify_message is now set immediately after the failure of a
4291 verification in an ACL, and so is available in subsequent modifiers. In
4292 particular, the message can be preserved by coding like this:
4294 warn !verify = sender
4295 set acl_m0 = $acl_verify_message
4297 Previously, $acl_verify_message was set only while expanding "message"
4298 and "log_message" when a very denied access.
4300 PH/28 Modified OS/os.c-Linux with
4302 -#ifndef OS_LOAD_AVERAGE
4303 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4305 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4306 nomenclature these days.)
4308 PH/29 Installed patch from the Sieve maintainer that adds the options
4309 sieve_useraddress and sieve_subaddress to the redirect router.
4311 PH/30 In these circumstances:
4312 . Two addresses routed to the same list of hosts;
4313 . First host does not offer TLS;
4314 . First host accepts first address;
4315 . First host gives temporary error to second address;
4316 . Second host offers TLS and a TLS session is established;
4317 . Second host accepts second address.
4318 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4319 and peerdn, if requested) that were in fact used only for the second
4322 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4323 attention to any local part prefix or suffix that was matched by the
4324 router that accepted the address. It now behaves in the same way as it
4325 does for delivery: the affixes are removed from the local part unless
4326 rcpt_include_affixes is set on the transport.
4328 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4329 timeout during the DATA phase of an incoming message.
4331 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4332 applied a patch sanctioned by the Sieve maintainer.
4334 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4335 the uid or gid is negative. A case of a negative gid caused this to be
4336 noticed. The fix allows for either to be negative.
4338 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4339 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4342 PH/36 The change PH/12 above was broken. Fixed it.
4344 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4345 the grounds that routing the same address twice would always produce the
4346 same answer. This might have been true once, but it is certainly no
4347 longer true now. Routing a child address may depend on the previous
4348 routing that produced that child. Some complicated redirection strategies
4349 went wrong when messages had multiple recipients, and made Exim's
4350 behaviour dependent on the order in which the addresses were given.
4352 I have moved the duplicate checking until after the routing is complete.
4353 Exim scans the addresses that are assigned to local and remote
4354 transports, and removes any duplicates. This means that more work will be
4355 done, as duplicates will always all be routed, but duplicates are
4356 presumably rare, so I don't expect this is of any significance.
4358 For deliveries to pipes, files, and autoreplies, the duplicate checking
4359 still happens during the routing process, since they are not going to be
4362 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4363 It corrects a timeout issue with spamd. This is Ian's comment: "The
4364 background is that sometimes spamd either never reads data from a
4365 connection it has accepted, or it never writes response data. The exiscan
4366 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4367 blindly assumes that writes won't block so it may never time out."
4369 PH/39 Allow G after quota size as well as K and M.
4371 PH/40 The value set for $authenticated_id in an authenticator may not contain
4372 binary zeroes or newlines because the value is written to log lines and
4373 to spool files. There was no check on this. Now the value is run through
4374 the string_printing() function so that such characters are converted to
4375 printable escape sequences.
4377 PH/41 $message_linecount is a new variable that contains the total number of
4378 lines in the message. Compare $body_linecount, which is the count for the
4381 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4382 bounce and delay warning messages, except in certain special cases, which
4385 (a) An SMTP error message from a remote host;
4386 (b) A message specified in a :fail: redirection;
4387 (c) A message specified in a "fail" command in a system filter;
4388 (d) A message specified in a FAIL return from the queryprogram router;
4389 (e) A message specified by the cannot_route_message router option.
4391 In these cases only, Exim does include the error details in bounce and
4392 warning messages. There are also a few cases where bland messages such
4393 as "unrouteable address" or "local delivery error" are given.
4395 PH/43 $value is now also set for the "else" part of a ${run expansion.
4397 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4398 being worked on, but at least Exim now implements the latest version to
4401 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4402 process to complete was treated as a delivery failure, a timeout while
4403 writing the message to the pipe was logged, but erroneously treated as a
4404 successful delivery. Such timeouts include transport filter timeouts. For
4405 consistency with the overall process timeout, these timeouts are now
4406 treated as errors, giving rise to delivery failures by default. However,
4407 there is now a new Boolean option for the pipe transport called
4408 timeout_defer, which, if set TRUE, converts the failures into defers for
4409 both kinds of timeout. A transport filter timeout is now identified in
4412 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4413 systems where "make" and "gmake" are different, calling "gmake" at top
4414 level broke things. I've arranged for the value of $(MAKE) to be passed
4415 from the Makefile to this script so that it can call the same version of
4419 A note about Exim versions 4.44 and 4.50
4420 ----------------------------------------
4422 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4423 changes of various kinds. As a consequence, a big documentation update was
4424 needed. This delayed the release for rather longer than seemed good, especially
4425 in the light of a couple of (minor) security issues. Therefore, the changes
4426 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4427 release. So 4.44 and 4.50 are in effect two different branches that both start
4430 I have left the 4.50 change log unchanged; it contains all the changes since
4431 4.43. The change log for 4.44 is below; many of its items are identical to
4432 those for 4.50. This seems to be the most sensible way to preserve the
4433 historical information.
4439 1. Minor wording change to the doc/README.SIEVE file.
4441 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4442 computation of the current number of files was incorrect.
4444 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4445 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4448 4. Give more explanation in the error message when the command for a transport
4449 filter fails to execute.
4451 5. There are several places where Exim runs a non-Exim command in a
4452 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4453 was being done only for the command run by the queryprogram router. It is
4454 now done for all such subprocesses. The other cases are: ${run, transport
4455 filters, and the commands run by the lmtp and pipe transports.
4457 6. Added CONFIGURE_GROUP build-time option.
4459 7. Some older OS have a limit of 256 on the maximum number of file
4460 descriptors. Exim was using setrlimit() to set 1000 as a large value
4461 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4462 systems. I've change it so that if it can't get 1000, it tries for 256.
4464 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4465 was an oversight, and furthermore, ever since the addition of extra
4466 controls (e.g. 4.43/32), the checks on when to allow different forms of
4467 "control" were broken. There should now be diagnostics for all cases when a
4468 control that does not make sense is encountered.
4470 9. Added the /retain_sender option to "control=submission".
4472 10. $recipients is now available in the predata ACL (oversight).
4474 11. Tidy the search cache before the fork to do a delivery from a message
4475 received from the command line. Otherwise the child will trigger a lookup
4476 failure and thereby defer the delivery if it tries to use (for example) a
4477 cached ldap connection that the parent has called unbind on.
4479 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4480 of $address_data from the recipient verification was clobbered by the
4481 sender verification.
4483 13. The value of address_data from a sender verification is now available in
4484 $sender_address_data in subsequent conditions in the ACL statement.
4486 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4488 15. Added a new option "connect=<time>" to callout options, to set a different
4491 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4492 was its contents. (It was OK if the option was not defined at all.)
4494 17. A "Completed" log line is now written for messages that are removed from
4495 the spool by the -Mrm option.
4497 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4498 information about exactly what failed.
4500 19. Added -dd to debug only the daemon process.
4502 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4503 handles timeouts, both on the server side and network timeouts. Renamed the
4504 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4506 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4507 It is now set to "smtps".
4509 22. $host_address is now set to the target address during the checking of
4510 ignore_target_hosts.
4512 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4513 being passed; this would have caused $sender_host_name to have been used if
4514 matching the list had actually called for a host name (not very likely,
4515 since this list is usually IP addresses). A host name is now passed as
4518 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4519 code that specifies a non-restarting handler (typically sigaction() in
4520 modern systems) in an attempt to fix a rare and obscure crash bug.
4522 25. Narrowed the window for a race in the daemon that could cause it to ignore
4523 SIGCHLD signals. This is not a major problem, because they are used only to
4524 wake it up if nothing else does.
4526 26. A malformed maildirsize file could cause Exim to calculate negative values
4527 for the mailbox size or file count. Odd effects could occur as a result.
4528 The maildirsize information is now recalculated if the size or filecount
4531 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4532 support for a long time. Removed HAVE_SYS_VFS_H.
4534 28. Installed the latest version of exipick from John Jetmore.
4536 29. In an address list, if the pattern was not a regular expression, an empty
4537 subject address (from a bounce message) matched only if the pattern was an
4538 empty string. Non-empty patterns were not even tested. This was the wrong
4539 because it is perfectly reasonable to use an empty address as part of a
4540 database query. An empty address is now tested by patterns that are
4541 lookups. However, all the other forms of pattern expect the subject to
4542 contain a local part and a domain, and therefore, for them, an empty
4543 address still always fails if the pattern is not itself empty.
4545 30. Exim went into a mad DNS loop when attempting to do a callout where the
4546 host was specified on an smtp transport, and looking it up yielded more
4547 than one IP address.
4549 31. Re-factored the code for checking spool and log partition space into a
4550 function that finds that data and another that does the check. The former
4551 is then used to implement four new variables: $spool_space, $log_space,
4552 $spool_inodes, and $log_inodes.
4554 32. The RFC2047 encoding function was originally intended for short strings
4555 such as real names; it was not keeping to the 75-character limit for
4556 encoded words that the RFC imposes. It now respects the limit, and
4557 generates multiple encoded words if necessary. To be on the safe side, I
4558 have increased the buffer size for the ${rfc2047: expansion operator from
4561 33. It is now permitted to omit both strings after an "if" condition; if the
4562 condition is true, the result is "true". As before, when the second string
4563 is omitted, a false condition yields an empty string. This makes it less
4564 cumbersome to write custom ACL and router conditions.
4566 34. Failure to deliver a bounce message always caused it to be frozen, even if
4567 there was an errors_to setting on the router. The errors_to setting is now
4570 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4571 canonical form (fully expanded) before being placed in
4572 $sender_host_address.
4574 36. The table in the code that translates DNS record types into text (T_A to
4575 "A" for instance) was missing entries for NS and CNAME. It is just possible
4576 that this could have caused confusion if both these types were looked up
4577 for the same domain, because the text type is used as part of Exim's
4578 per-process caching. But the chance of anyone hitting this buglet seems
4581 37. The dnsdb lookup has been extended in a number of ways.
4583 (1) There is a new type, "zns", which walks up the domain tree until it
4584 finds some nameserver records. It should be used with care.
4586 (2) There is a new type, "mxh", which is like "mx" except that it returns
4587 just the host names, not the priorities.
4589 (3) It is now possible to give a list of domains (or IP addresses) to be
4590 looked up. The behaviour when one of the lookups defers can be
4591 controlled by a keyword.
4593 (4) It is now possible to specify the separator character for use when
4594 multiple records are returned.
4596 38. The dnslists ACL condition has been extended: it is now possible to supply
4597 a list of IP addresses and/or domains to be looked up in a particular DNS
4600 39. Added log_selector=+queue_time_overall.
4602 40. When running the queue in the test harness, wait just a tad after forking a
4603 delivery process, to get repeatability of debugging output.
4605 41. Include certificate and key file names in error message when GnuTLS fails
4606 to set them up, because the GnuTLS error message doesn't include the name
4607 of the failing file when there is a problem reading it.
4609 42. Allow both -bf and -bF in the same test run.
4611 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4613 44. The "Exiscan patch" is now merged into the mainline Exim source.
4615 45. Sometimes the final signoff response after QUIT could fail to get
4616 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4617 < 0 before doing a fflush(). This bug looks as though it goes back to the
4618 introduction of TLS in release 3.20, but "sometimes" must have been rare
4619 because the tests only now provoked it.
4621 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4622 (this can affect the format of dates).
4624 47. exim_tidydb, when checking for the continued existence of a message for
4625 which it has found a message-specific retry record, was not finding
4626 messages that were in split spool directories. Consequently, it was
4627 deleting retry records that should have stayed in existence.
4629 48. Steve fixed some bugs in eximstats.
4631 49. The SPA authentication driver was not abandoning authentication and moving
4632 on to the next authenticator when an expansion was forced to fail,
4633 contradicting the general specification for all authenticators. Instead it
4634 was generating a temporary error. It now behaves as specified.
4636 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4637 (the order specifies the preference for clients). The order is now AES256,
4638 AES128, 3DES, ARCFOUR128.
4640 51. Small patch to Sieve code - explicitly set From: when generating an
4643 52. Exim crashed if a remote delivery caused a very long error message to be
4644 recorded - for instance if somebody sent an entire SpamAssassin report back
4645 as a large number of 550 error lines. This bug was coincidentally fixed by
4646 increasing the size of one of Exim's internal buffers (big_buffer) that
4647 happened as part of the Exiscan merge. However, to be on the safe side, I
4648 have made the code more robust (and fixed the comments that describe what
4651 53. Now that there can be additional text after "Completed" in log lines (if
4652 the queue_time_overall log selector is set), a one-byte patch to exigrep
4653 was needed to allow it to recognize "Completed" as not the last thing in
4656 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4657 patch that reportedly fixes this has been added. I am not expert enough to
4658 create a test for it. This is what the patch creator wrote:
4660 "I found a little strange behaviour of ldap code when working with
4661 Windows 2003 AD Domain, where users was placed in more than one
4662 Organization Units. When I tried to give exim partial DN, the exit code
4663 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4664 But simultaneously result of request was absolutely normal ldap result,
4665 so I produce this patch..."
4667 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4668 REFERENCE, so I have modified the code to exclude the patch when that macro
4671 55. Some experimental protocols are using DNS PTR records for new purposes. The
4672 keys for these records are domain names, not reversed IP addresses. The
4673 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4674 leaves it alone. Component reversal etc. now happens only for IP addresses.
4677 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4679 57. Double the size of the debug message buffer (to 2048) so that more of very
4680 long debug lines gets shown.
4682 58. The exicyclog utility now does better if the number of log files to keep
4683 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4685 59. Two changes related to the smtp_active_hostname option:
4687 (1) $smtp_active_hostname is now available as a variable.
4688 (2) The default for smtp_banner uses $smtp_active_hostname instead
4689 of $primary_hostname.
4691 60. The host_aton() function is supposed to be passed a string that is known
4692 to be a valid IP address. However, in the case of IPv6 addresses, it was
4693 not checking this. This is a hostage to fortune. Exim now panics and dies
4694 if the condition is not met. A case was found where this could be provoked
4695 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4696 components; fortuitously, this particular loophole had already been fixed
4697 by change 4.50/55 above.
4699 If there are any other similar loopholes, the new check in host_aton()
4700 itself should stop them being exploited. The report I received stated that
4701 data on the command line could provoke the exploit when Exim was running as
4702 exim, but did not say which command line option was involved. All I could
4703 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4704 running as the user.
4707 61. There was a buffer overflow vulnerability in the SPA authentication code
4708 (which came originally from the Samba project). I have added a test to the
4709 spa_base64_to_bits() function which I hope fixes it.
4712 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4713 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4715 63. The daemon start-up calls getloadavg() while still root for those OS that
4716 need the first call to be done as root, but it missed one case: when
4717 deliver_queue_load_max is set with deliver_drop_privilege. This is
4718 necessary for the benefit of the queue runner, because there is no re-exec
4719 when deliver_drop_privilege is set.
4721 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4722 This has been fixed.
4724 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4725 was in use, was not putting the data itself into the right store pool;
4726 consequently, it could be overwritten for a subsequent message in the same
4727 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4730 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4732 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4733 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4734 the function were treating the return as a boolean value, which happened to
4735 work because 0=false and not-0=true, but is not correct code.
4737 68. The host_aton() function was not handling scoped IPv6 addresses (those
4738 with, for example, "%eth0" on the end) correctly.
4740 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4741 negated items (that is, ~something) in unsigned ints. Some compilers
4742 apparently mutter when there is no cast.
4744 70. If an address verification called from an ACL failed, and did not produce a
4745 user-specific message (i.e. there was only a "system" message), nothing was
4746 put in $acl_verify_message. In this situation, it now puts the system
4749 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4750 session; change 4.31/43 added the unwanted input to the log line - except
4751 that it did not do this in the start of session case. It now does.
4753 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4754 This is wrong and can cause the other end to generate a synchronization
4755 error if it is another Exim or anything else that does the synchronization
4756 check. A QUIT command is no longer sent after a timeout.
4758 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4759 during host lookups.
4761 74. The defer_ok option of callout verification was not working if it was used
4762 when verifying addresses in header lines, that is, for this case:
4764 verify = header_sender/callout=defer_ok
4766 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4767 those file descriptors could be used for SMTP connections. If anything
4768 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4769 could be sent to the SMTP client, causing chaos. The daemon now opens
4770 stdin, stdout, and stderr to /dev/null when it puts itself into the
4773 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4774 log by default. The user can override this with suitable Perl magic.
4776 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4777 the log message when discard triggers, was not working for the DATA ACL or
4778 for the non-SMTP ACL.
4780 78. Error message wording change in sieve.c.
4782 79. If smtp_accept_max_per_host was set, the number of connections could be
4783 restricted to fewer than expected, because the daemon was trying to set up
4784 a new connection before checking whether the processes handling previous
4785 connections had finished. The check for completed processes is now done
4786 earlier. On busy systems, this bug wouldn't be noticed because something
4787 else would have woken the daemon, and it would have reaped the completed
4790 80. If a message was submitted locally by a user whose login name contained one
4791 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4792 It caused a spool format error. I have fixed the spool reading code. A
4793 related problem was that the "from" clause in the Received: line became
4794 illegal because of the space(s). It is now covered by ${quote_local_part.
4796 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4799 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4801 83. Patch from Sieve maintainer to fix unterminated string problem in
4802 "vacation" handling.
4804 84. Some minor changes to the Linux configuration files to help with other
4805 OS variants using glibc.
4807 85. One more patch for Sieve to update vacation handling to latest spec.
4810 ----------------------------------------------------
4811 See the note above about the 4.44 and 4.50 releases.
4812 ----------------------------------------------------
4818 1. Change 4.43/35 introduced a bug that caused file counts to be
4819 incorrectly computed when quota_filecount was set in an appendfile
4822 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4823 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4826 3. Give more explanation in the error message when the command for a transport
4827 filter fails to execute.
4829 4. There are several places where Exim runs a non-Exim command in a
4830 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4831 was being done only for the command run by the queryprogram router. It is
4832 now done for all such subprocesses. The other cases are: ${run, transport
4833 filters, and the commands run by the lmtp and pipe transports.
4835 5. Some older OS have a limit of 256 on the maximum number of file
4836 descriptors. Exim was using setrlimit() to set 1000 as a large value
4837 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4838 systems. I've change it so that if it can't get 1000, it tries for 256.
4840 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4841 was an oversight, and furthermore, ever since the addition of extra
4842 controls (e.g. 4.43/32), the checks on when to allow different forms of
4843 "control" were broken. There should now be diagnostics for all cases when a
4844 control that does not make sense is encountered.
4846 7. $recipients is now available in the predata ACL (oversight).
4848 8. Tidy the search cache before the fork to do a delivery from a message
4849 received from the command line. Otherwise the child will trigger a lookup
4850 failure and thereby defer the delivery if it tries to use (for example) a
4851 cached ldap connection that the parent has called unbind on.
4853 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4854 of $address_data from the recipient verification was clobbered by the
4855 sender verification.
4857 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4858 was its contents. (It was OK if the option was not defined at all.)
4860 11. A "Completed" log line is now written for messages that are removed from
4861 the spool by the -Mrm option.
4863 12. $host_address is now set to the target address during the checking of
4864 ignore_target_hosts.
4866 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4867 being passed; this would have caused $sender_host_name to have been used if
4868 matching the list had actually called for a host name (not very likely,
4869 since this list is usually IP addresses). A host name is now passed as
4872 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4873 code that specifies a non-restarting handler (typically sigaction() in
4874 modern systems) in an attempt to fix a rare and obscure crash bug.
4876 15. Narrowed the window for a race in the daemon that could cause it to ignore
4877 SIGCHLD signals. This is not a major problem, because they are used only to
4878 wake it up if nothing else does.
4880 16. A malformed maildirsize file could cause Exim to calculate negative values
4881 for the mailbox size or file count. Odd effects could occur as a result.
4882 The maildirsize information is now recalculated if the size or filecount
4885 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4886 support for a long time. Removed HAVE_SYS_VFS_H.
4888 18. Updated exipick to current release from John Jetmore.
4890 19. Allow an empty sender to be matched against a lookup in an address list.
4891 Previously the only cases considered were a regular expression, or an
4894 20. Exim went into a mad DNS lookup loop when doing a callout where the
4895 host was specified on the transport, if the DNS lookup yielded more than
4898 21. The RFC2047 encoding function was originally intended for short strings
4899 such as real names; it was not keeping to the 75-character limit for
4900 encoded words that the RFC imposes. It now respects the limit, and
4901 generates multiple encoded words if necessary. To be on the safe side, I
4902 have increased the buffer size for the ${rfc2047: expansion operator from
4905 22. Failure to deliver a bounce message always caused it to be frozen, even if
4906 there was an errors_to setting on the router. The errors_to setting is now
4909 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4910 canonical form (fully expanded) before being placed in
4911 $sender_host_address.
4913 24. Updated eximstats to version 1.33
4915 25. Include certificate and key file names in error message when GnuTLS fails
4916 to set them up, because the GnuTLS error message doesn't include the name
4917 of the failing file when there is a problem reading it.
4919 26. Expand error message when OpenSSL has problems setting up cert/key files.
4922 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4923 (this can affect the format of dates).
4925 28. exim_tidydb, when checking for the continued existence of a message for
4926 which it has found a message-specific retry record, was not finding
4927 messages that were in split spool directories. Consequently, it was
4928 deleting retry records that should have stayed in existence.
4930 29. eximstats updated to version 1.35
4931 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4932 1.35 - bugfix such that pie charts by volume are generated correctly
4934 30. The SPA authentication driver was not abandoning authentication and moving
4935 on to the next authenticator when an expansion was forced to fail,
4936 contradicting the general specification for all authenticators. Instead it
4937 was generating a temporary error. It now behaves as specified.
4939 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4940 (the order specifies the preference for clients). The order is now AES256,
4941 AES128, 3DES, ARCFOUR128.
4943 31. Small patch to Sieve code - explicitly set From: when generating an
4946 32. Exim crashed if a remote delivery caused a very long error message to be
4947 recorded - for instance if somebody sent an entire SpamAssassin report back
4948 as a large number of 550 error lines. This bug was coincidentally fixed by
4949 increasing the size of one of Exim's internal buffers (big_buffer) that
4950 happened as part of the Exiscan merge. However, to be on the safe side, I
4951 have made the code more robust (and fixed the comments that describe what
4954 33. Some experimental protocols are using DNS PTR records for new purposes. The
4955 keys for these records are domain names, not reversed IP addresses. The
4956 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4957 leaves it alone. Component reversal etc. now happens only for IP addresses.
4960 34. The host_aton() function is supposed to be passed a string that is known
4961 to be a valid IP address. However, in the case of IPv6 addresses, it was
4962 not checking this. This is a hostage to fortune. Exim now panics and dies
4963 if the condition is not met. A case was found where this could be provoked
4964 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4965 components; fortuitously, this particular loophole had already been fixed
4966 by change 4.50/55 or 4.44/33 above.
4968 If there are any other similar loopholes, the new check in host_aton()
4969 itself should stop them being exploited. The report I received stated that
4970 data on the command line could provoke the exploit when Exim was running as
4971 exim, but did not say which command line option was involved. All I could
4972 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4973 running as the user.
4976 35. There was a buffer overflow vulnerability in the SPA authentication code
4977 (which came originally from the Samba project). I have added a test to the
4978 spa_base64_to_bits() function which I hope fixes it.
4981 36. The daemon start-up calls getloadavg() while still root for those OS that
4982 need the first call to be done as root, but it missed one case: when
4983 deliver_queue_load_max is set with deliver_drop_privilege. This is
4984 necessary for the benefit of the queue runner, because there is no re-exec
4985 when deliver_drop_privilege is set.
4987 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4988 was in use, was not putting the data itself into the right store pool;
4989 consequently, it could be overwritten for a subsequent message in the same
4990 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4993 38. Sometimes the final signoff response after QUIT could fail to get
4994 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4995 < 0 before doing a fflush(). This bug looks as though it goes back to the
4996 introduction of TLS in release 3.20, but "sometimes" must have been rare
4997 because the tests only now provoked it.
5003 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5004 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5005 Now it returns an integer. A number of calls were still expecting a T/F
5006 return. Fortuitously, in all cases, the tests worked in OK situations,
5007 which is the norm. However, things would have gone wrong on any write
5008 failures on the smtp file descriptor. This function is used when sending
5009 messages over SMTP and also when doing verify callouts.
5011 2. When Exim is called to do synchronous delivery of a locally submitted
5012 message (the -odf or -odi options), it no longer closes stderr before doing
5015 3. Implemented the mua_wrapper option.
5017 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5019 5. Implemented the functions header_remove(), header_testname(),
5020 header_add_at_position(), and receive_remove_recipient(), and exported them
5023 6. If an ACL "warn" statement specified the addition of headers, Exim already
5024 inserted X-ACL-Warn: at the start if there was no header name. However, it
5025 was not making this test for the second and subsequent header lines if
5026 there were newlines in the string. This meant that an invalid header could
5027 be inserted if Exim was badly configured.
5029 7. Allow an ACL "warn" statement to add header lines at the start or after all
5030 the Received: headers, as well as at the end.
5032 8. Added the rcpt_4xx retry error code.
5034 9. Added postmaster_mailfrom=xxx to callout verification option.
5036 10. Added mailfrom=xxxx to the callout verification option, for verify=
5039 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5040 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5042 12. Inserted some casts to stop certain compilers warning when using pointer
5043 differences as field lengths or precisions in printf-type calls (mostly
5044 affecting debugging statements).
5046 13. Added optional readline() support for -be (dynamically loaded).
5048 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5049 same clock tick as a message's arrival, so that its received time was the
5050 same as the "first fail" time on the retry record, and that message
5051 remained on the queue past the ultimate address timeout, every queue runner
5052 would try a delivery (because it was past the ultimate address timeout) but
5053 after another failure, the ultimate address timeout, which should have then
5054 bounced the address, did not kick in. This was a "< instead of <=" error;
5055 in most cases the first failure would have been in the next clock tick
5056 after the received time, and all would be well.
5058 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5059 being recognized when the domain list was tested by the match_domain
5060 condition in an expansion string.
5062 16. Added the ${str2b64: operator.
5064 17. Exim was always calling setrlimit() to set a large limit for the number of
5065 processes, without checking whether the existing limit was already
5066 adequate. (It did check for the limit on file descriptors.) Furthermore,
5067 errors from getrlimit() and setrlimit() were being ignored. Now they are
5068 logged to the main and panic logs, but Exim does carry on, to try to do its
5069 job under whatever limits there are.
5071 18. Imported PCRE 5.0.
5073 19. Trivial typo in log message " temporarily refused connection" (the leading
5076 20. If the log selector return_path_on_delivery was set and an address was
5077 redirected to /dev/null, the delivery process crashed because it assumed
5078 that a return path would always be set for a "successful" delivery. In this
5079 case, the whole delivery is bypassed as an optimization, and therefore no
5082 21. Internal re-arrangement: the function for sending a challenge and reading
5083 a response while authentication was assuming a zero-terminated challenge
5084 string. It's now changed to take a pointer and a length, to allow for
5085 binary data in such strings.
5087 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5089 23. Exim was not respecting finduser_retries when seeking the login of the
5090 uid under which it was called; it was always trying 10 times. (The default
5091 setting of finduser_retries is zero.) Also, it was sleeping after the final
5092 failure, which is pointless.
5094 24. Implemented tls_on_connect_ports.
5096 25. Implemented acl_smtp_predata.
5098 26. If the domain in control=submission is set empty, Exim assumes that the
5099 authenticated id is a complete email address when it generates From: or
5100 Sender: header lines.
5102 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5103 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5104 chown and chgrp in /bin and hostname in /usr/bin.
5106 28. Exim was keeping the "process log" file open after each use, just as it
5107 does for the main log. This opens the possibility of it remaining open for
5108 long periods when the USR1 signal hits a daemon. Occasional processlog
5109 errors were reported, that could have been caused by this. Anyway, it seems
5110 much more sensible not to leave this file open at all, so that is what now
5113 29. The long-running daemon process does not normally write to the log once it
5114 has entered its main loop, and it closes the log before doing so. This is
5115 so that log files can straightforwardly be renamed and moved. However,
5116 there are a couple of unusual error situations where the daemon does write
5117 log entries, and I had neglected to close the log afterwards.
5119 30. The text of an SMTP error response that was received during a remote
5120 delivery was being truncated at 512 bytes. This is too short for some of
5121 the long messages that one sometimes sees. I've increased the limit to
5124 31. It is now possible to make retry rules that apply only when a message has a
5125 specific sender, in particular, an empty sender.
5127 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5128 it possible to be selective about when SMTP synchronization is enforced.
5130 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5132 32. Implemented hosts_connection_nolog.
5134 33. Added an ACL for QUIT.
5136 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5139 35. Added mailbox_size and mailbox_filecount to appendfile.
5141 36. Added control = no_multiline_responses to ACLs.
5143 37. There was a bug in the logic of the code that waits for the clock to tick
5144 in the case where the clock went backwards by a substantial amount such
5145 that the microsecond fraction of "now" was more than the microsecond
5146 fraction of "then" (but the whole seconds number was less).
5148 38. Added support for the libradius Radius client library this is found on
5149 FreeBSD (previously only the radiusclient library was supported).
5155 1. When certain lookups returned multiple values in the form name=value, the
5156 quoting of the values was not always being done properly. Specifically:
5157 (a) If the value started with a double quote, but contained no whitespace,
5159 (b) If the value contained whitespace other than a space character (i.e.
5160 tabs or newlines or carriage returns) it was not quoted.
5161 This fix has been applied to the mysql and pgsql lookups by writing a
5162 separate quoting function and calling it from the lookup code. The fix
5163 should probably also be applied to nisplus, ibase and oracle lookups, but
5164 since I cannot test any of those, I have not disturbed their existing code.
5166 2. A hit in the callout cache for a specific address caused a log line with no
5167 reason for rejecting RCPT. Now it says "Previous (cached) callout
5168 verification failure".
5170 3. There was an off-by-one bug in the queryprogram router. An over-long
5171 return line was truncated at 256 instead of 255 characters, thereby
5172 overflowing its buffer with the terminating zero. As well as fixing this, I
5173 have increased the buffer size to 1024 (and made a note to document this).
5175 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5176 when Exim is waiting for an SMTP response from a remote server, Exim
5177 restarts its select() call on the socket, thereby resetting its timeout.
5178 This is not a problem when such interrupts are rare. Somebody set up a cron
5179 job to run exiwhat every 2 minutes, which is less than the normal select()
5180 timeout (5 or 10 minutes). This meant that the select() timeout never
5181 kicked in because it was always reset. I have fixed this by comparing the
5182 time when an interrupt arrives with the time at the start of the first call
5183 to select(). If more time than the timeout has elapsed, the interrupt is
5184 treated as a timeout.
5186 5. Some internal re-factoring in preparation for the addition of Sieve
5187 extensions (by MH). In particular, the "personal" test is moved to a
5188 separate function, and given an option for scanning Cc: and Bcc: (which is
5189 not set for Exim filters).
5191 6. When Exim created an email address using the login of the caller as the
5192 local part (e.g. when creating a From: or Sender: header line), it was not
5193 quoting the local part when it contained special characters such as @.
5195 7. Installed new OpenBSD configuration files.
5197 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5198 try to make them clearer.
5200 9. Callout options, other than the timeout value, were being ignored when
5201 verifying sender addresses in header lines. For example, when using
5203 verify = header_sender/callout=no_cache
5205 the cache was (incorrectly) being used.
5207 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5208 only the Cygwin environment.
5210 11. When return_path_on_delivery was set as a log selector, if different remote
5211 addresses in the same message used different return paths and parallel
5212 remote delivery occurred, the wrong values would sometimes be logged.
5213 (Whenever a remote delivery process finished, the return path value from
5214 the most recently started remote delivery process was logged.)
5216 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5217 lines when AUTH and/or TLS are in use. This is the "received protocol"
5218 field. Exim used to use "asmtp" for authenticated SMTP, without any
5219 indication (in the protocol name) for TLS use. Now it follows the RFC and
5220 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5221 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5222 names appear in log lines as well as in Received: header lines.
5224 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5225 extensions, and comparison tests, and to fix some bugs.
5227 14. Changes to the "personal" filter test:
5229 (1) The test was buggy in that it was just doing the equivalent of
5230 "contains" tests on header lines. For example, if a user's address was
5231 anne@some.where, the "personal" test would incorrectly be true for
5233 To: susanne@some.where
5235 This test is now done by extracting each address from the header in turn,
5236 and checking the entire address. Other tests that are part of "personal"
5237 are now done using regular expressions (for example, to check local parts
5238 of addresses in From: header lines).
5240 (2) The list of non-personal local parts in From: addresses has been
5241 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5242 taken from the Sieve specification recommendations.
5244 (3) If the message contains any header line starting with "List-" it is
5245 treated as non-personal.
5247 (4) The test for "circular" in the Subject: header line has been removed
5248 because it now seems ill-conceived.
5250 15. Minor typos in src/EDITME comments corrected.
5252 16. Installed latest exipick from John Jetmore.
5254 17. If headers_add on a router specified a text string that was too long for
5255 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5256 of string_sprintf() is now avoided.
5258 18. $message_body_size was not set (it was always zero) when running the DATA
5259 ACL and the local_scan() function.
5261 19. For the "mail" command in an Exim filter, no default was being set for
5262 the once_repeat time, causing a random time value to be used if "once" was
5263 specified. (If the value happened to be <= 0, no repeat happened.) The
5264 default is now 0s, meaning "never repeat". The "vacation" command was OK
5265 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5266 (I found it when inspecting the code).
5268 20. There is now an overall timeout for performing a callout verification. It
5269 defaults to 4 times the callout timeout, which applies to individual SMTP
5270 commands during the callout. The overall timeout applies when there is more
5271 than one host that can be tried. The timeout is checked before trying the
5272 next host. This prevents very long delays if there are a large number of
5273 hosts and all are timing out (e.g. when the network connections are timing
5274 out). The value of the overall timeout can be changed by specifying an
5275 additional sub-option for "callout", called "maxwait". For example:
5277 verify = sender/callout=5s,maxwait=20s
5279 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5280 to the end before writing, but this should make it even safer).
5282 22. Exim was forgetting that it had advertised PIPELINING for the second and
5283 subsequent messages on an SMTP connection. It was also not resetting its
5284 memory on STARTTLS and an internal HELO.
5286 23. When Exim logs an SMTP synchronization error within a session, it now
5287 records whether PIPELINING has been advertised or not.
5289 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5290 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5291 rather than long int.
5293 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5295 26. Added the never_mail option to autoreply.
5301 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5302 crash if the getsockname() call failed; this can happen if a connection is
5303 closed very soon after it is established. The problem was simply in the
5304 order in which certain operations were done, causing Exim to try to write
5305 to the SMTP stream before it had set up the file descriptor. The bug has
5306 been fixed by making things happen in the correct order.
5312 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5313 before the connection was closed, thus losing the rejection response.
5315 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5316 some early Solaris releases, but causes trouble in current releases where
5317 socklen_t is defined.
5319 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5322 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5325 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5326 that releasing the top of it at the end releases what was used for sub-
5327 expansions (unless the block got too big). However, discard this block if
5328 the first thing is a variable or header, so that we can use its block when
5329 it is dynamic (useful for very large $message_headers, for example).
5331 6. Lookups now cache *every* query, not just the most recent. A new, separate
5332 store pool is used for this. It can be recovered when all lookup caches are
5333 flushed. Lookups now release memory at the end of their result strings.
5334 This has involved some general refactoring of the lookup sources.
5336 7. Some code has been added to the store_xxx() functions to reduce the amount
5337 of flapping under certain conditions.
5339 8. log_incoming_interface used to affect only the <= reception log lines. Now
5340 it causes the local interface and port to be added to several more SMTP log
5341 lines, for example "SMTP connection from", and rejection lines.
5343 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5345 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5347 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5348 could be overwritten at the end of the current message (or the start of a
5349 new message if it was set in a HELO ACL). The value is now preserved for
5350 the duration of the SMTP connection.
5352 12. If a transport had a headers_rewrite setting, and a matching header line
5353 contained an unqualified address, that address was qualified, even if it
5354 did not match any rewriting rules. The underlying bug was that the values
5355 of the flags that permit the existence of unqualified sender and recipient
5356 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5357 non-local messages, and by -bnq for local messages) were not being
5358 preserved with the message after it was received.
5360 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5361 "next input=" as part of the text comprising the host identity instead of
5362 the correct text. The code was using the same buffer for two different
5363 strings. However, depending on which order the printing function evaluated
5364 its arguments, the bug did not always show up. Under Linux, for example, my
5365 test suite worked just fine.
5367 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5368 doesn't work with some older versions of Perl. It has been changed to "my",
5369 which in any case is probably the better facility to use.
5371 15. A really picky compiler found some instances of statements for creating
5372 error messages that either had too many or two few arguments for the format
5375 16. The size of the buffer for calls to the DNS resolver has been increased
5376 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5377 for addresses that have a lot of PTR records. This alleviates a problem; it
5378 does not fully solve it.
5380 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5381 buffer now truncates the list and logs the incident, which is the same
5382 action as happens when Exim is looking up a host name and its aliases.
5383 Previously in this situation something unpredictable would happen;
5384 sometimes it was "internal error: store_reset failed".
5386 18. If a server dropped the connection unexpectedly when an Exim client was
5387 using GnuTLS and trying to read a response, the client delivery process
5388 crashed while trying to generate an error log message.
5390 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5391 string, for example:
5393 warn message = H1: something\nH2: something
5395 the text was added as a single header line from Exim's point of view
5396 though it ended up OK in the delivered message. However, searching for the
5397 second and subsequent header lines using $h_h2: did not work. This has been
5398 fixed. Similarly, if a system filter added multiple headers in this way,
5399 the routers could not see them.
5401 20. Expanded the error message when iplsearch is called with an invalid key to
5402 suggest using net-iplsearch in a host list.
5404 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5405 ACLs are no longer actually imposed (and a message to that effect is
5408 22. If a "gecos" field in a passwd entry contained escaped characters, in
5409 particular, if it contained a \" sequence, Exim got it wrong when building
5410 a From: or a Sender: header from that name. A second bug also caused
5411 incorrect handling when an unquoted " was present following a character
5412 that needed quoting.
5414 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5415 was not being matched caselessly.
5417 24. Arranged for all hyphens in the exim.8 source to be escaped with
5420 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5421 properly. Recipient callout cache records were still being keyed to include
5422 the sender, even when use_sender was set false. This led to far more
5423 callouts that were necessary. The sender is no longer included in the key
5424 when use_sender is false.
5426 26. Added "control = submission" modifier to ACLs.
5428 27. Added the ${base62d: operator to decode base 62 numbers.
5430 28. dnsdb lookups can now access SRV records.
5432 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5433 the configuration file.
5435 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5436 (-v) mode. This makes the output for a verbose queue run more intelligible.
5438 31. Added a use_postmaster feature to recipient callouts.
5440 32. Added the $body_zerocount variable, containing the number of binary zero
5441 bytes in the message body.
5443 33. The time of last modification of the "new" subdirectory is now used as the
5444 "mailbox time last read" when there is a quota error for a maildir
5447 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5449 35. Added +ignore_unknown as a special item in host lists.
5451 36. Code for decoding IPv6 addresses in host lists is now included, even if
5452 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5453 address was recognized as an IP address, but was then not correctly decoded
5454 into binary, causing unexpected and incorrect effects when compared with
5461 1. Very minor rewording of debugging text in manualroute to say "list of
5462 hosts" instead of "hostlist".
5464 2. If verify=header_syntax was set, and a header line with an unqualified
5465 address (no domain) and a large number of spaces between the end of the
5466 name and the colon was received, the reception process suffered a buffer
5467 overflow, and (when I tested it) crashed. This was caused by some obsolete
5468 code that should have been removed. The fix is to remove it!
5470 3. When running in the test harness, delay a bit after writing a bounce
5471 message to get a bit more predictability in the log output.
5473 4. Added a call to search_tidyup() just before forking a reception process. In
5474 theory, someone could use a lookup in the expansion of smtp_accept_max_
5475 per_host which, without the tidyup, could leave open a database connection.
5477 5. Added the variables $recipient_data and $sender_data which get set from a
5478 lookup success in an ACL "recipients" or "senders" condition, or a router
5479 "senders" option, similar to $domain_data and $local_part_data.
5481 6. Moved the writing of debug_print from before to after the "senders" test
5484 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5485 problems for message scanning, either using a data ACL, or using
5486 local_scan() because the Received: header was not generated till after they
5487 were called (in order to set the time as the time of reception completion).
5488 I have revised the way this works. The header is now generated after the
5489 body is received, but before the ACL or local_scan() are called. After they
5490 are run, the timestamp in the header is updated.
5496 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5497 before starting a queue runner without re-exec. This happened only when
5498 deliver_drop_privilege was set or when the Exim user was set to root. The
5499 effect of the bug was that timeouts during subsequent deliveries caused
5500 crashes instead of being properly handled. The handler is now left at its
5501 default (and expected) setting.
5503 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5504 message, again when deliver_drop_privilege is set or Exim is run as root.
5505 The bug described in (1) was not present in this case, but the tidying up
5506 of the other signals was missing. I have made the two cases consistent.
5508 3. The ignore_target_hosts setting on a manualroute router was being ignored
5509 for hosts that were looked up using the /MX notation.
5511 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5514 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5515 operated on the sender address. After changing the $sender_address to <>
5516 for the sender address verify, Exim was re-instated it as the original
5517 (before rewriting) address, but remembering that it had rewritten it, so it
5518 wasn't rewriting it again. This bug also had the effect of breaking the
5519 sender address verification caching when the sender address was rewritten.
5521 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5522 This has been changed so that if the ip literal address matches
5523 ignore_target_hosts, the router declines.
5525 7. Added expansion conditions match_domain, match_address, and match_local_
5526 part (NOT match_host).
5528 8. The placeholder for the Received: header didn't have a length field set.
5530 9. Added code to Exim itself and to exim_lock to test for a specific race
5531 condition that could lead to file corruption when using MBX delivery. The
5532 issue is with the lockfile that is created in /tmp. If this file is removed
5533 after a process has opened it but before that process has acquired a lock,
5534 there is the potential for a second process to recreate the file and also
5535 acquire a lock. This could lead to two Exim processes writing to the file
5536 at the same time. The added code performs the same test as UW imapd; it
5537 checks after acquiring the lock that its file descriptor still refers to
5538 the same named file.
5540 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5541 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5544 11. Added the smtp_active_hostname option. If used, this will typically be made
5545 to depend on the incoming interface address. Because $interface_address is
5546 not set up until the daemon has forked a reception process, error responses
5547 that can happen earlier (such as "too many connections") no longer contain
5550 12. If an expansion in a condition on a "warn" statement fails because a lookup
5551 defers, the "warn" statement is abandoned, and the next ACL statement is
5552 processed. Previously this caused the whole ACL to be aborted.
5554 13. Added the iplsearch lookup type.
5556 14. Added ident_timeout as a log selector.
5558 15. Added tls_certificate_verified as a log selector.
5560 16. Added a global option tls_require_ciphers (compare the smtp transport
5561 option of the same name). This controls incoming TLS connections.
5563 17. I finally figured out how to make tls_require_ciphers do a similar thing
5564 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5565 before starting the TLS session.
5567 18. Tabs are now shown as \t in -bP output.
5569 19. If the log selector return_path_on_delivery was set, Exim crashed when
5570 bouncing a message because it had too many Received: header lines.
5572 20. If two routers both had headers_remove settings, and the first one included
5573 a superfluous trailing colon, the final name in the first list and the
5574 first name in the second list were incorrectly joined into one item (with a
5575 colon in the middle).
5581 1. Added -C and -D options to the exinext utility, mainly to make it easier
5582 to include in the automated testing, but these could be helpful when
5583 multiple configurations are in use.
5585 2. The exinext utility was not formatting the output nicely when there was
5586 an alternate port involved in the retry record key, nor when there was a
5587 message id as well (for retries that were specific to a specific message
5588 and a specific host). It was also confused by IPv6 addresses, because of
5589 the additional colons they contain. I have fixed the IPv4 problem, and
5590 patched it up to do a reasonable job for IPv6.
5592 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5593 delivery, the log line now contains "pipelined" if PIPELINING was used.
5595 4. An SMTP transport process used to panic and die if the bind() call to set
5596 an explicit outgoing interface failed. This has been changed; it is now
5597 treated in the same way as a connect() failure.
5599 5. A reference to $sender_host_name in the part of a conditional expansion
5600 that was being skipped was still causing a DNS lookup. This no longer
5603 6. The def: expansion condition was not recognizing references to header lines
5604 that used bh_ and bheader_.
5606 7. Added the _cache feature to named lists.
5608 8. The code for checking quota_filecount in the appendfile transport was
5609 allowing one more file than it should have been.
5611 9. For compatibility with Sendmail, the command line option
5619 and sets the incoming protocol and host name (for trusted callers). The
5620 host name and its colon can be omitted when only the protocol is to be set.
5621 Note the Exim already has two private options, -pd and -ps, that refer to
5622 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5623 "s", but I don't think that's a major issue.
5625 10. A number of refactoring changes to the code, none of which should affect
5628 (a) The number of logging options was getting close to filling up the
5629 32-bit word that was used as a bit map. I have split them into two classes:
5630 those that are passed in the argument to log_write(), and those that are
5631 only ever tested independently outside of that function. These are now in
5632 separate 32-bit words, so there is plenty of room for expansion again.
5633 There is no change in the user interface or the logging behaviour.
5635 (b) When building, for example, log lines, the code previously used a
5636 macro that called string_cat() twice, in order to add two strings. This is
5637 not really sufficiently general. Furthermore, there was one instance where
5638 it was actually wrong because one of the argument was used twice, and in
5639 one call a function was used. (As it happened, calling the function twice
5640 did not affect the overall behaviour.) The macro has been replaced by a
5641 function that can join an arbitrary number of extra strings onto a growing
5644 (c) The code for expansion conditions now uses a table and a binary chop
5645 instead of a serial search (which was left over from when there were very
5646 few conditions). Also, it now recognizes conditions like "pam" even when
5647 the relevant support is not compiled in: a suitably worded error message is
5648 given if an attempt is made to use such a condition.
5650 11. Added ${time_interval:xxxxx}.
5652 12. A bug was causing one of the ddress fields not to be passed back correctly
5653 from remote delivery subprocesses. The field in question was not being
5654 subsequently used, so this caused to problems in practice.
5656 13. Added new log selectors queue_time and deliver_time.
5658 14. Might have fixed a bug in maildirsizefile handling that threw up
5659 "unexpected character" debug warnings, and recalculated the data
5660 unnecessarily. In any case, I expanded the warning message to give more
5663 15. Added the message "Restricted characters in address" to the statements in
5664 the default ACL that block characters like @ and % in local parts.
5666 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5667 Three changes have been made:
5669 (a) There was a serious bug; a negative response to MAIL caused the whole
5670 recipient domain to be cached as invalid, thereby blocking all messages
5671 to all local parts at the same domain, from all senders. This bug has
5672 been fixed. The domain is no longer cached after a negative response to
5673 MAIL if the sender used is not empty.
5675 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5678 (c) A new callout option, "use_sender" has been added for people who want
5679 the modified behaviour.
5685 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5688 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5689 indeed breaks things for older releases.
5691 3. Added additional logging to the case where there is a problem reading data
5692 from a filter that is running in a subprocess using a pipe, in order to
5693 try to track down a specific problem.
5695 4. Testing facility fudge: when running in the test harness and attempting
5696 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5697 getting "No route to host". Convert this to a timeout.
5699 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5702 6. Some OS don't have socklen_t but use size_t instead. This affects the
5703 fifth argument of getsockopt() amongst other things. This is now
5704 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5705 can be set for individual OS. I have set it for SunOS5, OSF1, and
5706 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5707 some earlier ones do not.
5709 7. Change 4.30/15 was not doing the test caselessly.
5711 8. The standard form for an IPv6 address literal was being rejected by address
5712 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5713 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5714 this, as well as the form without the "IPv6" on the front (but only when
5715 address literals are enabled, of course).
5717 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5719 10. Exim crashed if a message with an empty sender address specified by -f
5720 encountered a router with an errors_to setting. This could be provoked only
5721 by a command such as
5725 where an empty string was supplied; "<>" did not hit this bug.
5727 11. Installed PCRE release 4.5.
5729 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5730 remained set. It is now erased.
5732 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5733 times from message ids (which are base 36 rather than the normal 62).
5735 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5736 were being counted as actual protocol errors, and logged if the log
5737 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5738 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5739 and DATA following a set of rejected RCPTs do not count as protocol errors.
5740 In other words, Exim assumes they were pipelined, though this may not
5741 actually be the case. Of course, in all cases the client gets an
5742 appropriate error code.
5744 15. If a lookup fails in an ACL condition, a message about the failure may
5745 be available; it is used if testing the ACL cannot continue, because most
5746 such messages specify what the cause of the deferral is. However, some
5747 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5748 that caused an old message to be retained and used if a later statement
5749 caused a defer, replacing the real cause of the deferral.
5751 16. If an IP address had so many PTR records that the DNS lookup buffer
5752 was not large enough to hold them, Exim could crash while trying to process
5753 the truncated data. It now detects and logs this case.
5755 17. Further to 4.21/58, another change has been made: if (and only if) the
5756 first line of a message (the first header line) ends with CRLF, a bare LF
5757 in a subsequent header line has a space inserted after it, so as not to
5758 terminate the header.
5760 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5761 unnecessarily, used atoi() instead of strtol(), and didn't check the
5762 termination when getting file sizes from file names by regex.
5764 19. Completely re-implemented the support for maildirsize files, in the light
5765 of a number of problems with the previous contributed implementation
5766 (4.30/29). In particular:
5768 . If the quota is zero, the maildirsize file is maintained, but no quota is
5771 . If the maildir directory does not exist, it is created before any attempt
5772 to write a maildirsize file.
5774 . The quota value in the file is just a cache; if the quota is changed in
5775 the transport, the new value overrides.
5777 . A regular expression is available for excluding directories from the
5780 20. The autoreply transport checks the characters in options that define the
5781 message's headers; it allows continued headers, but it was checking with
5782 isspace() after an embedded newline instead of explicitly looking for a
5785 21. If all the "regular" hosts to which an address was routed had passed their
5786 expiry times, and had not reached their retry times, the address was
5787 bounced, even if fallback hosts were defined. Now Exim should go on to try
5790 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5791 equivalent code in the SMTP transport. Some hosts send humungous responses
5792 to HELO/EHLO, more than 1024 it seems.
5794 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5795 gives compiler warnings in some environments. I've now done it "properly",
5798 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5799 (because of problems with the built-in one) was declared to return uschar *
5800 instead of char *, causing compiler failure.
5802 25. Fixed a file descriptor leak when processing alias/forward files.
5804 26. Fixed a minor format string issue in dbfn.c.
5806 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5808 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5809 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5812 29. When a sender address is verified, it is cached, to save repeating the test
5813 when there is more than one recipient in a message. However, when the
5814 verification involves a callout, it is possible for different callout
5815 options to be set for different recipients. It is too complicated to keep
5816 track of this in the cache, so now Exim always runs a verification when a
5817 callout is required, relying on the callout cache for the optimization.
5818 The overhead is duplication of the address routing, but this should not be
5821 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5822 to be verified with callout=postmaster, and the main callout worked but the
5823 postmaster check failed, the verification correctly failed. However, if a
5824 subsequent RCPT command asked for sender verification *without* the
5825 postmaster check, incorrect caching caused this verification also to fail,
5828 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5829 it was not caching the DNS options (qualify_single, search_parents) that
5830 were used when the lookup failed. A subsequent lookup with different
5831 options therefore always gave the same answer, though there were cases
5832 where it should not have. (Example: a "domains = !$mx_any" option on a
5833 dnslookup router: the "domains" option is always processed without any
5834 widening, but the router might have qualify_single set.) Now Exim uses the
5835 cached value only when the same options are set.
5837 32. Added John Jetmore's "exipick" utility to the distribution.
5839 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5840 than a timeout (e.g. a certificate is required, and is not provided), an
5841 Exim server now closes the connection immediately. Previously it waited for
5842 the client to close - but if the client is SSL, it seems that they each
5843 wait for each other, leading to a delay before one of them times out.
5845 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5846 maintained 0.8.x compatibility because I don't think many are using it, and
5847 it is clearly obsolete.
5849 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5852 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5853 client certificate was expired. A simple patch fixes this, though I don't
5854 understand the full logic of why the verify callback is called multiple
5857 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5858 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5859 which causes problems with some clients (such as the Certicom SSL Plus
5860 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5861 disables the coutermeasure allowing Eudora to connect."
5863 38. Exim was not checking that a write() to a log file succeeded. This could
5864 lead to Bad Things if a log got too big, in particular if it hit a file
5865 size limit. Exim now panics and dies if it cannot write to a log file, just
5866 as it does if it cannot open a log file.
5868 39. Modified OS/Makefile-Linux so that it now contains
5870 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5872 The two -D definitions ensure that Exim is compiled with large file
5873 support, which makes it possible to handle log files that are bigger than
5876 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5877 instance) a domain was checked against a named list that involved a lookup,
5878 causing $domain_data to be set, then another domain was checked against the
5879 same list, then the first domain was re-checked, the value of $domain_data
5880 after the final check could be wrong. In particular, if the second check
5881 failed, it could be set empty. This bug probably also applied to
5884 41. The strip_trailing_dot option was not being applied to the address given
5885 with the -f command-line option.
5887 42. The code for reading a message's header from the spool was incrementing
5888 $received_count, but never initializing it. This meant that the value was
5889 incorrect (doubled) while delivering a message in the same process in which
5890 it was received. In the most common configuration of Exim, this never
5891 happens - a fresh exec is done - but it can happen when
5892 deliver_drop_privilege is set.
5894 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5895 it now includes up to 150 characters of the unexpected data in the log
5898 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5899 and building data strings. The size of both of these buffers was 10 000
5900 bytes - far larger than anybody would *ever* want, thought I. Needless to
5901 say, somebody hit the limit. I have increased the maximum line length to
5902 20 000 and the maximum data length of concatenated lines to 100 000. I have
5903 also fixed two bugs, because there was no checking on these buffers. Tsk,
5904 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5905 buffer is too small.
5907 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5908 lsearch lookups. Now it does.
5910 46. When parsing a route_list item in a manualroute router, a fixed-length
5911 buffer was used for the list of hosts. I made this 1024 bytes long,
5912 thinking that nobody would ever have a list of hosts that long. Wrong.
5913 Somebody had a whole pile of complicated expansion conditions, and the
5914 string was silently truncated, leading to an expansion error. It turns out
5915 that it is easier to change to an unlimited length (owing to other changes
5916 that have happened since this code was originally written) than to build
5917 structure for giving a limitation error. The length of the item that
5918 expands into the list of hosts is now unlimited.
5920 47. The lsearch lookup could not handle data where the length of text line was
5921 more than 4095 characters. Such lines were truncated, leading to shortened
5922 data being returned. It should now handle lines of any length.
5924 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5925 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5928 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5929 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5930 no longer happens in comment lines. A list of replacements is now placed
5931 at the head of all of the source files, except those whose only change is
5932 to replace PERL_COMMAND in the very first #! line.
5934 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5935 messages on the queue, with a bottom-up merge sort, using code contributed
5936 by Michael Haardt. This should make operations like -bp somewhat faster on
5937 large queues. It won't affect queue runners, except when queue_run_in_order
5940 51. Installed eximstats 1.31 in the distribution.
5942 52. Added support for SRV lookups to the dnslookup router.
5944 53. If an ACL referred to $message_body or $message_body_end, the value was not
5945 reset for any messages that followed in the same SMTP session.
5947 54. The store-handling optimization for building very long strings was not
5948 differentiating between the different store pools. I don't think this
5949 actually made any difference in practice, but I've tidied it.
5951 55. While running the routers to verify a sender address, $sender_address
5952 was still set to the sender address. This is wrong, because when routing to
5953 send a bounce to the sender, it would be empty. Therefore, I have changed
5954 it so that, while verifying a sender address, $sender_address is set to <>.
5955 (There is no change to what happens when verifying a recipient address.)
5957 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5958 target A or AAAA records (if not already returned) without resetting the
5959 qualify_single or search_parents options of the DNS resolver. These are
5960 inappropriate in this case because the targets of MX and SRV records must
5961 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5962 target that, when qualified, matched something in the local domain. These
5963 two options are now turned off when doing these lookups.
5965 57. It seems that at least some releases of Reiserfs (which does not have the
5966 concept of a fixed number of inodes) returns zero and not -1 for the
5967 number of available inodes. This interacted badly with check_spool_inodes,
5968 which assumed that -1 was the "no such thing" setting. What I have done is
5969 to check that the total number of inodes is greater than zero before doing
5970 the test of how many are available.
5972 58. When a "warn" ACL statement has a log_message modifier, the message is
5973 remembered, and not repeated. This is to avoid a lot of repetition when a
5974 message has many recipients that cause the same warning to be written.
5975 However, Exim was preserving the list of already written lines for an
5976 entire SMTP session, which doesn't seem right. The memory is now reset if a
5977 new message is started.
5979 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5980 debugging output unless log_rewrite was also set.
5982 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5983 of (int)(handle) when we know that handle contains (void *)(-1).
5985 61. The Exim daemon panic-logs an error return when it closes the incoming
5986 connection. However "connection reset by peer" seems to be common, and
5987 isn't really an error worthy of noting specially, so that particular error
5990 62. When Exim is trying to find all the local interfaces, it used to panic and
5991 die if the ioctl to get the interface flags failed. However, it seems that
5992 on at least one OS (Solaris 9) it is possible to have an interface that is
5993 included in the list of interfaces, but for which you get a failure error
5994 for this call. This happens when the interface is not "plumbed" into a
5995 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5996 failure of the "get flags" call assumes that the interface is down.
5998 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5999 makes life easier for people who are doing arithmetic on fields extracted
6000 from dates, where you often get leading zeros that should not be
6001 interpreted as octal.
6003 64. Added qualify_domain to the redirect router, to override the global
6006 65. If a pathologically long header line contained very many addresses (the
6007 report of this problem mentioned 10 000) and each of them was rewritten,
6008 Exim could use up a very large amount of memory. (It kept on making new
6009 copies of the header line as it rewrote, and never released the old ones.)
6010 At the expense of a bit more processing, the header rewriting function has
6011 been changed so that it no longer eats memory in this way.
6013 66. The generation of the Received: header has been moved from the time that a
6014 message starts to be received, to the time that it finishes. The timestamp
6015 in the Received: header should now be very close to that of the <= log
6016 line. There are two side-effects of this change:
6018 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6019 logged header lines no longer include the local Received: line, because
6020 it has not yet been created. The same applies to a copy of the message
6021 that is returned to a non-SMTP sender when a message is rejected.
6023 (b) When a filter file is tested using -bf, no additional Received: header
6024 is added to the test message. After some thought, I decided that this
6027 This change does not affect the value of $received_for. It is still set
6028 after address rewriting, but before local_scan() is called.
6030 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6032 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6033 gave an unhelpful panic error message, and a defer error. I have managed to
6034 change this behaviour so that it now rejects any supplied certificate,
6035 which seems right, as the list of acceptable certificates is empty.
6037 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6038 gave an unhelpful defer error. I have not managed to make this reject any
6039 supplied certificates, but the error message it gives is "no certificate
6040 supplied", which is not helpful.
6042 70. exigrep's output now also includes lines that are not associated with any
6043 message, but which match the given pattern. Implemented by a patch from
6044 Martin Sluka, which also tidied up the Perl a bit.
6046 71. Recipient callout verification, like sender verification, was using <> in
6047 the MAIL FROM command. This isn't really the right thing, since the actual
6048 sender may affect whether the remote host accepts the recipient or not. I
6049 have changed it to use the actual sender in the callout; this means that
6050 the cache record is now keyed on a recipient/sender pair, not just the
6051 recipient address. There doesn't seem to be a real danger of callout loops,
6052 since a callout by the remote host to check the sender would use <>.
6053 [SEE ABOVE: changed after hitting problems.]
6055 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6056 temporary errors. However, in the case of such a code being given after
6057 the end of a data transmission (i.e. after ".") Exim was failing to write
6058 a retry record for the message. (Yes, there was some broken host that was
6059 actually sending 8xx at this point.)
6061 73. An unknown lookup type in a host list could cause Exim to panic-die when
6062 the list was checked. (An example that provoked this was putting <; in the
6063 middle of a list instead of at the start.) If this happened during a DATA
6064 ACL check, a -D file could be left lying around. This kind of configuration
6065 error no longer causes Exim to die; instead it causes a defer error. The
6066 incident is still logged to the main and panic logs.
6068 74. Buglet left over from Exim 3 conversion. The message "too many messages
6069 in one connection" was written to the rejectlog but not the mainlog, except
6070 when address rewriting (yes!) was being logged.
6072 75. Added write_rejectlog option.
6074 76. When a system filter was run not as root (that is, when system_filter_user
6075 was set), the values of the $n variables were not being returned to the
6076 main process; thus, they were not subsequently available in the $sn
6079 77. Added +return_path_on_delivery log selector.
6081 78. A connection timeout was being treated differently from recipients deferred
6082 when testing hosts_max_try with a message that was older than the host's
6083 retry timeout. (The host should not be counted, thus allowing all hosts to
6084 be tried at least once before bouncing.) This may have been the cause of an
6085 occasionally reported bug whereby a message would remain on the queue
6086 longer than the retry timeout, but would be bounced if a delivery was
6087 forced. I say "may" because I never totally pinned down the problem;
6088 setting up timeout/retry tests is difficult. See also the next item.
6090 79. The ultimate address timeout was not being applied to errors that involved
6091 a combination of host plus message (for example, a timeout on a MAIL
6092 command). When an address resolved to a number of possible hosts, and they
6093 were not all tried for each delivery (e.g. because of hosts_max_try), a
6094 message could remain on the queue longer than the retry timeout.
6096 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6099 81. Fixed an obscure SMTP outgoing bug which required at least the following
6100 conditions: (a) there was another message waiting for the same server;
6101 (b) the server returned 5xx to all RCPT commands in the first message so
6102 that the message was not completed; (c) the server dropped the connection
6103 or gave a negative response to the RSET that Exim sends to abort the
6104 transaction. The observed case was a dropped connection after DATA that had
6105 been sent in pipelining mode. That is, the server had advertised PIPELINING
6106 but was not implementing it correctly. The effect of the bug was incorrect
6107 behaviour, such as trying another host, and this could lead to a crash.
6113 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6114 and daemon.c were passed as pointers to ints; they should have been
6115 pointers to socklen_t variables (which are typically unsigned ints).
6117 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6120 3. Fixed a really odd bug that affected only the testing scheme; patching a
6121 certain fixed string in the binary changed the value of another string that
6122 happened to be identical to the end of the original first string.
6124 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6125 name", it returns that address as the IP address. On some operating
6126 systems (e.g. Solaris), it also passes back the IP address string as the
6127 "host name". However, on others (e.g. Linux), it passes back an empty
6128 string. Exim wasn't checking for this, and was changing the host name to an
6129 empty string, assuming it had been canonicalized.
6131 5. Although rare, it is permitted to have more than one PTR record for a given
6132 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6133 all the names associated with an address, because they do in Solaris.
6134 However, it seems that they do not in Linux for data that comes from the
6135 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6136 I found this out when I moved to a new Linux workstation and tried to run
6137 the Exim test suite.
6139 To get round this problem I have changed the code so that it now does its
6140 own call to the DNS to look up PTR records when searching for a host name.
6141 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6142 addresses that are only in /etc/hosts are still found.
6144 This behaviour is, however, controlled by an option called host_lookup_
6145 order, which defaults to "bydns:byaddr". If people want to use the other
6146 order, or indeed, just use one or the other means of lookup, they can
6147 specify it in this variable.
6149 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6150 some operating systems, this comes back from gethostbyaddr() as an empty
6151 string, and this is what Exim used to test for. However, it seems that in
6152 other systems, "." is yielded. Exim now tests for this case too.
6154 7. The values of check_spool_space and check_log_space are now held internally
6155 as a number of kilobytes instead of an absolute number of bytes. If a
6156 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6157 kilobyte. This means that much larger values can be stored.
6159 8. Exim monitor: an attempt to get the action menu when not actually pointing
6160 at a message produces an empty menu entitled "No message selected". This
6161 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6162 no entries in it ("Shell widget menu has zero width and/or height"). So I
6163 have added a single, blank menu entry in this case.
6165 9. Added ${quote_local_part.
6167 10. MIME decoding is now applied to the contents of Subject: header lines when
6170 11. Now that a reference to $sender_host_address automatically causes a reverse
6171 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6172 host lookup before query-style lookups in lists that might use this
6173 variable. This has therefore been abolished, and the "net-" prefix is no
6174 longer necessary for query-style lookups.
6176 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6177 have been a typo for LFLAGS, so it has been changed.
6179 13. The install script calls Exim with "-C /dev/null" in order to find the
6180 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6181 to be output. However, since Exim outputs its version number before the
6182 error, it didn't break the script. It just looked ugly. I fixed this by
6183 always allowing "-C /dev/null" if the caller is root.
6185 14. Ignore overlarge ACL variable number when reading spool file - insurance
6186 against a later release with more variables having written the file.
6188 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6189 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6190 this, as well as the form without the "IPv6" on the front.
6192 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6193 OS/Makefile-Darwin file.
6195 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6196 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6198 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6199 variable. However, some OS use ints for inodes. I've added cast to long int
6200 to get rid of the compiler warning.
6202 19. I had forgotten to lock out "/../" in configuration file names when
6203 ALT_CONFIG_PREFIX was set.
6205 20. Routers used for verification do not need to specify transports. However,
6206 if such a router generated a host list, and callout was configured, Exim
6207 crashed, because it could not find a port number from the (non-existent)
6208 transport. It now assumes port 25 in this circumstance.
6210 21. Added the -t option to exigrep.
6212 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6213 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6214 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6215 them is set, LOOKUP_LSEARCH is forced.
6217 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6219 24. Added sender and host information to the "rejected by local_scan()" log
6220 line; previously there was no indication of these.
6222 25. Added .include_if_exists.
6224 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6225 Linux. It turns out that not all file systems support this. Apparently some
6226 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6227 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6228 not supported on the file descriptor, is now ignored when Exim is trying to
6229 sync a directory. This applies only to Linux.
6231 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6233 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6234 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6237 29. Added support for maildirsize files from supplied patch (modified a bit).
6239 30. The use of :fail: followed by an empty string could lead Exim to respond to
6240 sender verification failures with (e.g.):
6242 550 Verification failed for <xxx>
6243 550 Sender verify failed
6245 where the first response line was missing the '-' that indicates it is not
6246 the final line of the response.
6248 31. The loop for finding the name of the user that called Exim had a hardwired
6249 limit of 10; it now uses the value of finduser_retries, which is used for
6250 all other user lookups.
6252 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6255 33. Exim was neglecting to zero errno before one call of strtol() when
6256 expanding a string and expecting an integer value. On some systems this
6257 resulted in spurious "integer overflow" errors. Also, it was casting the
6258 result into an int without checking.
6260 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6261 did not work. The code looks as if it has *never* worked, though it appears
6262 to have been documented since at least release 1.62. I have made it work.
6264 35. The "timeout_DNS" error in retry rules, also documented since at least
6265 1.62, also never worked. As it isn't clear exactly what this means, and
6266 clearly it isn't a major issue, I have abolished the feature by treating it
6267 as "timeout", and writing a warning to the main and panic logs.
6269 36. The display of retry rules for -brt wasn't always showing the error code
6272 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6273 timeout_connect_A, timeout_connect_MX.
6275 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6276 to the empty sender.
6278 39. The daemon was not analysing the content of -oX till after it had closed
6279 stderr and disconnected from the controlling terminal. This meant that any
6280 syntax errors were only noted on the panic log, and the return code from
6281 the command was 0. By re-arranging the code a little, I've made the
6282 decoding happen first, so such errors now appear on stderr, and the return
6283 code is 1. However, the actual setting up of the sockets still happens in
6284 the disconnected process, so errors there are still only recorded on the
6287 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6288 connections (as happens on some IP stacks) was logged at start up time as
6289 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6290 it from "IPv6 and IPv4", which means that two separate sockets are being
6293 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6294 says whether AF_INET or AF_INET6 was passed as an argument.
6296 42. Exiwhat output was messed up when time zones were included in log
6299 43. Exiwhat now gives more information about the daemon's listening ports,
6300 and whether -tls-on-connect was used.
6302 44. The "port" option of the smtp transport is now expanded.
6304 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6305 silently ignored. Now an error message is written to the main and panic
6308 46. There's a new ACL modifier called "logwrite" which writes to a log file
6309 as soon as it is encountered.
6311 47. Added $local_user_uid and $local_user_gid at routing time.
6313 48. Exim crashed when trying to verify a sender address that was being
6316 49. Exim was recognizing only a space character after ".include". It now also
6317 recognizes a tab character.
6319 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6320 extracting the relevant information from the specification. The man page no
6321 longer contains scrambled data for the -d option, and I've added a section
6322 at the front about calling Exim under different names.
6324 51. Added "extra_headers" argument to the "mail" command in filter files.
6326 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6329 53. Installed eximstats 1.29.
6331 54. Added transport_filter_timeout as a generic transport option.
6333 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6334 Cc: header lines. This was required by RFC 822, but it not required by RFC
6337 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6338 incoming messages that did not have them. Now it does so only if the
6339 message originates locally, that is, if there is no associated remote host
6340 address. When Resent- header lines are present, this applies to the Resent-
6341 lines rather than the non-Resent- lines.
6343 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6344 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6346 58. Messages for configuration errors now include the name of the main
6347 configuration file - useful now that there may be more than one file in a
6348 list (.included file names were always shown).
6350 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6351 for those rare installations that do not start the daemon as root or run it
6352 setuid root. I've cut out the call to initgroups() if the daemon is not
6355 60. The Exim user and group can now be bound into the binary as text strings
6356 that are looked up at the start of Exim's processing.
6358 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6360 62. Added $mailstore_basename variable.
6362 63. Installed patch to sieve.c from Michael Haardt.
6364 64. When Exim failed to open the panic log after failing to open the main log,
6365 the original message it was trying to log was written to stderr and debug
6366 output, but if they were not available (the usual case in production), it
6367 was lost. Now it is written to syslog before the two lines that record the
6368 failures to open the logs.
6370 65. Users' Exim filters run in subprocesses under the user's uid. It is
6371 possible for a "deliver" command or an alias in a "personal" command to
6372 provoke an address rewrite. If logging of address rewriting is configured,
6373 this fails because the process is not running as root or exim. There may be
6374 a better way of dealing with this, but for the moment (because 4.30 needs
6375 to be released), I have disabled address rewrite logging when running a
6376 filter in a non-root, non-exim process.
6382 1. The buildconfig auxiliary program wasn't quoting the value set for
6383 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6384 not defined. This bug was masked in 4.22 by the effect that was fixed in
6387 2. Some messages that were rejected after a message id was allocated were
6388 shown as "incomplete" by exigrep. It no longer does this for messages that
6389 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6391 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6392 have allow_domain_literals set, the ID did not get logged in the <= line.
6393 Domain literals are now always recognized in Message-ID: header lines.
6395 4. The first argument for a ${extract expansion item is the key name or field
6396 number. Leading and trailing spaces in this item were not being ignored,
6397 causing some misleading effects.
6399 5. When deliver_drop_privilege was set, single queue runner processes started
6400 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6401 same command in the process it spins off) were not dropping privilege.
6403 6. When the daemon running as "exim" started a queue runner, it always
6404 re-executed Exim in the spun-off process. This is a waste of effort when
6405 deliver_drop_privilege is set. The new process now just calls the
6406 queue-runner function directly.
6412 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6415 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6416 ignored. Though the use of -oP was forcing the writing of a pid file, it
6417 was always written to the default place.
6419 3. If the message "no IP address found for host xxxx" is generated during
6420 incoming verification, it is now followed by identification of the incoming
6421 connection (so you can more easily find what provoked it).
6423 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6425 5. Added some features to "harden" Exim a bit more against certain attacks:
6427 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6428 be put in Local/Makefile. This is like the never_users runtime option,
6429 but it cannot be overridden. The default setting is "root".
6431 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6432 prefix string with which any file named in a -C command line option
6435 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6436 is retained for -C and -D only if the caller of Exim is root. Without
6437 it, the exim user may also use -C and -D and retain privilege.
6439 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6440 command line option is disabled.
6442 6. Macro names set by the -D option must start with an upper case letter, just
6443 like macro names defined in the configuration file.
6445 7. Added "dereference=" facility to LDAP.
6447 8. Two instances of the typo "uknown" in the source files are fixed.
6449 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6450 the Configure-Makefile script screwed up while processing it.
6452 10. Incorporated PCRE 4.4.
6454 11. The SMTP synchronization check was not operating right at the start of an
6455 SMTP session. For example, it could not catch a HELO sent before the client
6456 waited for the greeting. There is now a check for outstanding input at the
6457 point when the greeting is written. Because of the duplex, asynchronous
6458 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6459 way, but not yet received, when the check is performed.
6461 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6462 on TCP/IP sockets, because this apparently causes some broken clients to
6465 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6466 unchanged) from the Cygwin maintainer.
6468 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6469 was supported instead of testing for mbx. Effectively a typo.
6471 15. The spa authenticator server code was not checking that the input it
6472 received was valid base64.
6474 16. The debug output line for the "set" modifier in ACLs was not showing the
6475 name of the variable that was being set.
6477 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6479 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6480 lookup on its own. Something else was needed to trigger the lookup. For
6481 example, a match in host_lookup or the need for a host name in a host list.
6482 Now, if $sender_host_name is referenced and the host name has not yet been
6483 looked up, a lookup is performed. If the lookup fails, the variable remains
6484 empty, and $host_lookup_failed is set to "1".
6486 19. Added "eqi" as a case-independent comparison operator.
6488 20. The saslauthd authentication condition could segfault if neither service
6489 nor realm was specified.
6491 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6492 error message that was logged was misleading, and incoming SMTP
6493 connections were dropped. The message is now more accurate, and temporary
6494 errors are given to SMTP connections.
6496 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6497 (except RSET) with a 421 error, until QUIT is received. However, it was
6498 failing to send a response to QUIT.
6500 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6501 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6502 ACL is now run after the helo_try_verify_hosts code.
6504 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6505 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6506 case-independent, so other case variants are also recognized). Apparently
6507 some systems use these upper case variants.
6509 25. If more than two messages were waiting for the same host, and a transport
6510 filter was specified for the transport, Exim sent two messages over the
6511 same TCP/IP connection, and then failed with "socket operation on non-
6512 socket" when it tried to send the third.
6514 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6516 27. The extern definition of crypt16() in expand.c was not being excluded when
6517 the OS had its own crypt16() function.
6519 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6520 as a preferred synonym for return_size_limit, both as an option and as an
6523 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6525 30. Changed the default configuration ACL to relax the local part checking rule
6526 for addresses that are not in any local domains. For these addresses,
6527 slashes and pipe symbols are allowed within local parts, but the sequence
6528 /../ is explicitly forbidden.
6530 31. SPA server authentication was not clearing the challenge buffer before
6533 32. log_message in a "warn" ACL statement was writing to the reject log as
6534 well as to the main log, which contradicts the documentation and doesn't
6535 seem right (because no rejection is happening). So I have stopped it.
6537 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6538 However, I am unable to do any testing of this.
6540 34. Fixed an infelicity in the appendfile transport. When checking directories
6541 for a mailbox, to see if any needed to be created, it was accidentally
6542 using path names with one or more superfluous leading slashes; tracing
6543 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6545 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6546 contents are added to the log line that is written for every discarded
6547 recipient. (Previously a log_message setting was ignored.)
6549 36. The ${quote: operator now quotes the string if it is empty.
6551 37. The install script runs exim in order to find its version number. If for
6552 some reason other than non-existence or emptiness, which it checks, it
6553 could not run './exim', it was installing it with an empty version number,
6554 i.e. as "exim-". This error state is now caught, and the installation is
6557 38. An argument was missing from the function that creates an error message
6558 when Exim fails to connect to the socket for saslauthd authentication.
6559 This could cause Exim to crash, or give a corrupted message.
6561 39. Added isip, isip4, and isip6 to ${if conditions.
6563 40. The ACL variables $acl_xx are now saved with the message, and can be
6564 accessed later in routers, transports, and filters.
6566 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6567 strings in the file are not string-expanded.
6569 42. If a MAIL command specified a SIZE value that was too large to fit into an
6570 int variable, the check against message_size_limit failed. Such values are
6571 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6572 day this will have to be increased, but I don't think I want to be around
6573 when emails are that large.
6580 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6581 iconv() is not standard in FreeBSD.
6583 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6584 IPv6 enabled. The observed symptom was a segmentation fault on return from
6585 the function os_common_find_running_interfaces() in src/os.c.
6587 3. In the check_special_case() function in daemon.c I had used "errno" as an
6588 argument name, which causes warnings on some systems. This was basically a
6589 typo, since it was named "eno" in the comments!
6591 4. The code that waits for the clock to tick (at a resolution of some fraction
6592 of a second) so as to ensure message-id uniqueness was always waiting for
6593 at least one whole tick, when it could have waited for less. [This is
6594 almost certainly not relevant at current processor speeds, where it is
6595 unlikely to ever wait at all. But we try to future-proof.]
6597 5. The function that sleeps for a time interval that includes fractions of a
6598 second contained a race. It did not block SIGALRM between setting the
6599 timer, and suspending (a couple of lines later). If the interval was short
6600 and the sigsuspend() was delayed until after it had expired, the suspension
6601 never ended. On busy systems this could lead to processes getting stuck for
6604 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6605 process, before it forks any delivery processes. The open lookup caching
6606 mechanism meant that the open file or database connection was passed into
6607 the delivery process. The problem was that delivery processes always tidy
6608 up cached lookup data. This could cause a problem for the next delivery
6609 process started by the queue runner, because the external queue runner
6610 process does not know about the closure. So the next delivery process
6611 still has data in the lookup cache. In the case of a file lookup, there was
6612 no problem because closing a file descriptor in a subprocess doesn't affect
6613 the parent. However, if the lookup was caching a connection to a database,
6614 the connection was closed, and the second delivery process was likely to
6615 see errors such as "PGSQL: query failed: server closed the connection
6616 unexpectedly". The problem has been fixed by closing all cached lookups
6617 in a queue runner before running a delivery process.
6619 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6620 seem to have the "const" qualifier which it has on other OS. I've
6623 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6624 *of the same type* (client or server) with the same public name that an
6625 error should be diagnosed.
6627 9. When Exim looked up a host name for an IP address, but failed to find the
6628 original IP address when looking up the host name (a safety check), it
6629 output the message "<ip address> does not match any IP for NULL", which was
6630 confusing, to say the least. The bug was that the host name should have
6631 appeared instead of "NULL".
6633 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6634 user that is built into the binary, and the -C or -D options is used, root
6635 privilege is dropped before the configuration file is read. In addition,
6636 logging is switched to stderr instead of the normal log files. If the
6637 configuration then re-defines the Exim user, the unprivileged environment
6638 is probably not what is expected, so Exim logs a panic warning message (but
6641 However, if deliver_drop_privilege is set, the unprivileged state may well
6642 be exactly what is intended, so the warning has been cut out in that case,
6643 and Exim is allowed to try to write to its normal log files.
6649 1. smtp_return_error_details was not giving details for temporary sender
6650 or receiver verification errors.
6652 2. Diagnose a configuration error if two authenticators have the same public
6655 3. Exim used not to create the message log file for a message until the first
6656 delivery attempt. This could be confusing when incoming messages were held
6657 for policy or load reasons. The message log file is now created at the time
6658 the message is received, and an initial "Received" line is written to it.
6660 4. The automatically generated man page for command line options had a minor
6661 bug that caused no ill effects; however, a more serious problem was that
6662 the procedure for building the man page automatically didn't always
6663 operate. Consequently, release 4.20 contains an out-of-date version. This
6664 shouldn't happen again.
6666 5. When building Exim with embedded Perl support, the script that builds the
6667 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6668 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6670 6. The freeze_tell option was not being used for messages that were frozen on
6671 arrival, either by an ACL or by local_scan().
6673 7. Added the smtp_incomplete_transaction log selector.
6675 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6676 was accepting AUTH without a new EHLO.
6678 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6679 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6682 10. Logging of TCP/IP connections (when configured) now happens in the main
6683 daemon process instead of the child process, so that the TCP/IP connection
6684 count is more accurate (but it can never be perfect).
6686 11. The use of "drop" in a nested ACL was not being handled correctly in the
6687 outer ACL. Now, if condition failure induced by the nested "drop" causes
6688 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6689 or "require"), the connection is dropped.
6691 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6692 that yield "discard" can now be used with an "accept" or a "discard" verb,
6693 but an error is generated for any others (because I can't see a useful way
6694 to define what should happen).
6696 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6697 are now processed in the same way as lines in the Exim configuration file.
6698 In particular, continuation lines are supported.
6700 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6702 15. Added -ti meaning -t -i.
6704 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6705 domains, and warn by logging if others are found.
6707 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6708 returned from GIFCONF when Exim is trying to find the list of interfaces on
6709 a host. The code in os.c has been modified to copy each ifreq to an aligned
6710 structure in all cases.
6712 Also, in some cases, the returned ifreq's were being copied to a 'struct
6713 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6714 means the last couple of bytes of an IPv6 address could be chopped if the
6715 ifreq contained only a normal sockaddr (14 bytes storage).
6717 18. Named domain lists were not supported in the hosts_treat_as_local option.
6718 An entry such as +xxxx was not recognized, and was treated as a literal
6721 19. Ensure that header lines added by a DATA ACL are included in the reject log
6722 if the ACL subsequently rejects the message.
6724 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6725 MD5 (which is deprecated).
6727 21. When testing a filter file using -bf, Exim was writing a message when it
6728 took the sender from a "From " line in the message, but it was not doing so
6729 when it took $return_path from a Return-Path: header line. It now does.
6731 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6732 with a valid header line field name (a series of printing characters
6733 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6735 23. Changed "disc" in the source to "disk" to conform to the documentation and
6736 the book and for uniformity.
6738 24. Ignore Sendmail's -Ooption=value command line item.
6740 25. When execve() failed while trying to run a command in a pipe transport,
6741 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6742 could be confused with a return value of 69 from the command itself. This
6743 has been changed to 127, the value the shell returns if it is asked to run
6744 a non-existent command. The wording for the related log line suggests a
6745 non-existent command as the problem.
6747 26. If received_header_text expands to an empty string, do not add a Received:
6748 header line to the message. (Well, it adds a token one on the spool, but
6749 marks it "old" so that it doesn't get used or transmitted.)
6751 27. Installed eximstats 1.28 (addition of -nt option).
6753 28. There was no check for failure on the call to getsockname() in the daemon
6754 code. This can fail if there is a shortage of resources on the system, with
6755 ENOMEM, for example. A temporary error is now given on failure.
6757 29. Contrary to the C standard, it seems that in some environments, the
6758 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6759 program. Exim now does this explicitly; it affects the formatting of
6760 timestamps using strftime().
6762 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6763 complaints. I've now initialized all the variables, to avoid this.
6765 32. Header lines added by a system filter were not being "seen" during
6766 transport-time rewrites.
6768 33. The info_callback() function passed to OpenSSL is set up with type void
6769 (*)(SSL *, int, int), as described somewhere. However, when calling the
6770 function (actually a macro) that sets it up, the type void(*)() is
6771 expected. I've put in a cast to prevent warnings from picky compilers.
6773 34. If a DNS black list lookup found a CNAME record, but there were no A
6774 records associated with the domain it pointed at, Exim crashed.
6776 35. If a DNS black list lookup returned more than one A record, Exim ignored
6777 all but the first. It now scans all returned addresses if a particular IP
6778 value is being sought. In this situation, the contents of the
6779 $dnslist_value variable are a list of all the addresses, separated by a
6782 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6783 to accept a host name and all its aliases if the forward lookup for any of
6784 them yielded the IP address of the incoming connection. Now it accepts only
6785 those names whose forward lookup yields the correct IP address. Any other
6786 names are discarded. This closes a loophole whereby a rogue DNS
6787 administrator could create reverse DNS records to break through a
6788 wildcarded host restriction in an ACL.
6790 37. If a user filter or a system filter that ran in a subprocess used any of
6791 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6792 the wrong values were passed to the pipe command ($thisaddress had the
6793 value of $0, $0 had the value of $1, etc). This bug was introduced by
6794 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6796 38. Improved the line breaking for long SMTP error messages from ACLs.
6797 Previously, if there was no break point between 40 and 75 characters, Exim
6798 left the rest of the message alone. Two changes have been made: (a) I've
6799 reduced the minimum length to 35 characters; (b) if it can't find a break
6800 point between 35 and 75 characters, it looks ahead and uses the first one
6801 that it finds. This may give the occasional overlong line, but at least the
6802 remaining text gets split now.
6804 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6805 file descriptors might be low, and that setting 1000 would always raise it.
6806 It turns out that in some environments, the limit is already over 1000 and
6807 that lowering it causes trouble. So now Exim takes care not to decrease it.
6809 40. When delivering a message, the value of $return_path is set to $sender_
6810 address at the start of routing (routers may change the value). By an
6811 oversight, this default was not being set up when an address was tested by
6812 -bt or -bv, which affected the outcome if any router or filter referred to
6815 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6816 log only when "message" or "log_message" are set. However, if one of the
6817 conditions was an address verification, or a call to a nested ACL, the
6818 messages generated by the underlying test were being passed through. This
6819 no longer happens. The underlying message is available in $acl_verify_
6820 message for both "message" and "log_message" expansions, so it can be
6821 passed through if needed.
6823 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6824 new expansion $bh_ to give the encoded byte string without charset
6825 translation. Translation happens only if iconv() is available; HAVE_ICONV
6826 indicates this at build time. HEADERS_CHARSET gives the charset to
6827 translate to; headers_charset can change it in the configuration, and
6828 "headers charset" can change it in an individual filter file.
6830 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6831 that creates RFC 2047 encoded "words" labels them as that charset instead
6832 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6833 expansion operator; (ii) when Exim creates a From: line for a local
6834 message; (iii) when a header line is rewritten to include a "phrase" part.
6836 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6837 buggy, causing it to skip the first lines of messages whose message ID
6838 ended in 'D'. This would not have bitten before Exim release 4.14, because
6839 message IDs were unlikely to end in 'D' before then. The effect was to have
6840 incorrect size information for certain domains.
6842 45. #include "config.h" was missing at the start of the crypt16.c module. This
6843 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6846 46. If there was a timeout during a "random" callout check, Exim treated it as
6847 a failure of the random address, and carried on sending RSET and the real
6848 address. If the delay was just some slowness somewhere, the response to the
6849 original RCPT would be taken as a response to RSET and so on, causing
6850 mayhem of various kinds.
6852 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6853 when I implemented it. It didn't allow for the fact that some option values
6854 may legitimately be negative (e.g. size_addition), and it didn't even do
6855 the right test for positive values.
6857 48. Domain names in DNS records are case-independent. Exim always looks them up
6858 in lower case. Some resolvers return domain names in exactly the case they
6859 appear in the zone file, that is, they may contain uppercase letters. Not
6860 all resolvers do this - some return always lower case. Exim was treating a
6861 change of case by a resolver as a change of domain, similar to a widening
6862 of a domain abbreviation. This triggered its re-routing code and so it was
6863 trying to route what was effectively the same domain again. This normally
6864 caused routing to fail (because the router wouldn't handle the domain
6865 twice). Now Exim checks for this case specially, and just changes the
6866 casing of the domain that it ultimately uses when it transmits the message
6869 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6872 50. If a filter generated a file delivery with a non-absolute name (possible if
6873 no home directory exists for the router), the forbid_file option was not
6876 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6877 the existing equality matching.
6879 52. Exim was using ints instead of ino_t variables in some places where it was
6880 dealing with inode numbers.
6882 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6883 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6884 called TMPDIR, and if it finds it is different, it changes its value.
6886 54. The smtp_printf() function is now made available to local_scan() so
6887 additional output lines can be written before returning. There is also an
6888 smtp_fflush() function to enable the detection of a dropped connection.
6889 The variables smtp_input and smtp_batched_input are exported to
6892 55. Changed the default runtime configuration: the message "Unknown user"
6893 has been removed from the ACL, and instead placed on the localuser router,
6894 using the cannot_route_message feature. This means that any verification
6895 failures that generate their own messages won't get overridden. Similarly,
6896 the "Unrouteable address" message that was in the ACL for unverifiable
6897 relay addresses has also been removed.
6899 56. Added hosts_avoid_esmtp to the smtp transport.
6901 57. The exicyclog script was not checking for the esoteric option
6902 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6903 will work only if exicyclog is run under the appropriate euid.
6905 58. Following a discussion on the list, the rules by which Exim recognises line
6906 endings on incoming messages have been changed. The -dropcr and drop_cr
6907 options are now no-ops, retained only for backwards compatibility. The
6908 following line terminators are recognized: LF CRLF CR. However, special
6909 processing applies to CR:
6911 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6912 nor a local message in the state where . is a terminator.
6914 (ii) If a bare CR is encountered in a header line, an extra space is added
6915 after the line terminator so as not to end the header. The reasoning
6916 behind this is that bare CRs in header lines are most likely either
6917 to be mistakes, or people trying to play silly games.
6919 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6920 was being incorrectly given as 18 bytes larger than it should have been.
6921 This is a VOB (very old bug).
6923 60. This may never have affected anything current, but just in case it has:
6924 When the local host is found other than at the start of a list of hosts,
6925 the local host, those with the same MX, and any that follow, are discarded.
6926 When the list in question was part of a longer list of hosts, the following
6927 hosts (not currently being processed) were also being discarded. This no
6928 longer happens. I'm not sure if this situation could ever has previously
6931 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6934 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6935 header. This is something that is recommended in a new Internet Draft, and
6936 is something that is documented as being done by Sendmail. There are two
6937 possible values. For messages generated by the autoreply transport, Exim
6940 Auto-Submitted: auto-replied
6942 whereas for all other generated messages (e.g. bounces) it adds
6944 Auto-Submitted: auto-generated
6946 63. The "personal" condition in filters now includes a test for the
6947 Auto-Submitted: header. If it contains the string "auto-" the message it
6948 not considered personal.
6950 64. Added rcpt_include_affixes as a generic transport option.
6952 65. Added queue_only_override (default true).
6954 66. Added the syslog_duplication option.
6956 67. If what should have been the first header line of a message consisted of
6957 a space followed by a colon, Exim was mis-interpreting it as a header line.
6958 It isn't of course - it is syntactically invalid and should therefore be
6959 treated as the start of the message body. The misbehaviour could have
6960 caused a number of strange effects, including loss of data in subsequent
6961 header lines, and spool format errors.
6963 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6964 client host had authenticated. This control can now be exercised by an ACL
6965 for more flexibility.
6967 69. By default, callouts do not happen when testing with -bh. There is now a
6968 variant, -bhc, which does actually run the callout code, including
6969 consulting and updating the callout cache.
6971 70. Added support for saslauthd authentication, courtesy of Alexander
6974 71. If statvfs() failed on the spool or log directories while checking their
6975 size for availability, Exim confusingly gave the error "space shortage".
6976 Furthermore, in debugging mode it crashed with a floating point exception.
6977 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6978 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6979 serious problem, Exim now writes to the main and panic logs when this
6980 happens, with details of the failure. It then refuses to accept the
6981 incoming message, giving the message "spool directory problem" or "log
6982 directory problem" with a 421 code for SMTP messages.
6984 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6985 0, 1, and 2 exist, because some OS complain for execs without them (see
6986 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6987 descriptors. However, the code omitted to check that the open succeeded,
6988 causing mysterious errors if for some reason the permissions on /dev/null
6989 got screwed. Now Exim writes a message to the main and panic logs, and
6990 bombs out if it can't open /dev/null.
6992 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6993 interact so that it is all more flexible. It is supposed to remain
6994 backwards compatible. Also added extra_local_interfaces.
6996 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6997 to bomb out with an assertion failure - to the client this appears as a
6998 connection drop. This problem occurs in the part of the code that was taken
6999 from the Samba project. Fortunately, the assertion is in a very simple
7000 function, so I have fixed this by reproducing the function inline in the
7001 one place where it is called, and arranging for authentication to fail
7002 instead of killing the process with assert().
7004 75. The SPA client code was not working when the server requested OEM rather
7005 than Unicode encoding.
7007 76. Added code to make require_files with a specific uid setting more usable in
7008 the case where statting the file as root fails - usually a non-root-mounted
7009 NFS file system. When this happens and the failure is EACCES, Exim now
7010 forks a subprocess and does the per-uid checking as the relevant uid.
7012 77. Added process_log_path.
7014 78. If log_file_path was not explicitly set, a setting of check_log_space or
7015 check_log_inodes was ignored.
7017 79. If a space check for the spool or log partitions fails, the incident is now
7018 logged. Of course, in the latter case the data may get lost...
7020 80. Added the %p formatting code to string_format() so that it can be used to
7021 print addresses in debug_print(). Adjusted all the address printing in the
7022 debugging in store.c to use %p rather than %d.
7024 81. There was a concern that a line of code in smtp_in.c could overflow a
7025 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7026 initially expressed, the concern was not well-founded, because trailing
7027 spaces are removed early. However, if the trailing spaces were followed by
7028 a NULL, they did not get removed, so the overflow was possible. Two fixes
7031 (a) I re-wrote the offending code in a cleaner fashion.
7032 (b) If an incoming SMTP command contains a NULL character, it is rejected
7035 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7036 runs initgroups(), so that if the Exim user is in any additional groups,
7037 they will be used during message reception.
7043 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob