1 Change log file for Exim from version 4.21
2 ------------------------------------------
3 This document describes *changes* to previous versions, that might
4 affect Exim's operation, with an unchanged configuration file. For new
5 options, and new features, see the NewStuff file next to this ChangeLog.
11 JH/01 Replace the store_release() internal interface with store_newblock(),
12 which internalises the check required to safely use the old one, plus
13 the allocate and data copy operations duplicated in both (!) of the
16 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
17 modifier. This matches the restriction on the commandline.
19 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
20 Previously only the last row was returned.
22 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
23 we assumed that tags in the header were well-formed, and parsed the
24 element content after inspecting only the first char of the tag.
25 Assumptions at that stage could crash the receive process on malformed
28 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
29 While running the DKIM ACL we operate on the Permanent memory pool so that
30 variables created with "set" persist to the DATA ACL. Also (at any time)
31 DNS lookups that fail create cache records using the Permanent pool. But
32 expansions release any allocations made on the current pool - so a dnsdb
33 lookup expansion done in the DKIM ACL releases the memory used for the
34 DNS negative-cache, and bad things result. Solution is to switch to the
35 Main pool for expansions.
36 While we're in that code, add checks on the DNS cache during store_reset,
37 active in the testsuite.
38 Problem spotted, and debugging aided, by Wolfgang Breyha.
40 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
41 When none of the hosts presented to a transport match an already-open
42 connection, close it and proceed with the list. Previously we would
43 queue the message. Spotted by Lena with Yahoo, probably involving
46 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
47 Previously a spurious "250 OK id=" response was appended to the proper
50 JH/08 The "support for" informational output now, which built with Content
51 Scanning support, has a line for the malware scanner interfaces compiled
52 in. Interface can be individually included or not at build time.
54 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
55 by the template makefile "src/EDITME". The "STREAM" support for an older
56 ClamAV interface method is removed.
58 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
59 rows affected is given instead).
61 JH/11 The runtime Berkeley DB library version is now additionally output by
62 "exim -d -bV". Previously only the compile-time version was shown.
64 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
65 SMTP connection. Previously, when one had more receipients than the
66 first, an abortive onward connection was made. Move to full support for
67 multiple onward connections in sequence, handling cutthrough connection
68 for all multi-message initiating connections.
70 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
71 routers. Previously, a multi-recipient message would fail to match the
72 onward-connection opened for the first recipient, and cause its closure.
74 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
75 a timeout on read on a GnuTLS initiating connection, resulting in the
76 initiating connection being dropped. This mattered most when the callout
77 was marked defer_ok. Fix to keep the two timeout-detection methods
80 JH/15 Relax results from ACL control request to enable cutthrough, in
81 unsupported situations, from error to silently (except under debug)
82 ignoring. This covers use with PRDR, frozen messages, queue-only and
85 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
86 metadata, resulting in a crash in free().
88 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
89 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
90 Broken also in d185889f4, with init system revamp.
92 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
93 Previously we abruptly closed the connection after reading a malware-
94 found indication; now we go on to read the "scan ok" response line,
101 JH/01 Rework error string handling in TLS interface so that the caller in
102 more cases is responsible for logging. This permits library-sourced
103 string to be attached to addresses during delivery, and collapses
104 pairs of long lines into single ones.
106 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
107 during configuration. Wildcards are allowed and expanded.
109 JH/02 Rework error string handling in DKIM to pass more info back to callers.
110 This permits better logging.
112 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
113 do not close it down and have the child transport start it up again on
114 the passed-on TCP connection. Instead, proxy the child (and any
115 subsequent ones) for TLS via a unix-domain socket channel. Logging is
116 affected: the continued delivery log lines do not have any DNSSEC, TLS
117 Certificate or OCSP information. TLS cipher information is still logged.
119 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
120 identical IP addresses on different listening ports. Will also affect
123 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
124 add noisy ifdef guards to special-case this sillyness.
125 Patch from Bernd Kuhls.
127 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
128 than 255 are no longer allowed.
130 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
131 Disable the session-cache too, which might reduce our load. Since we
132 currrectly use a new context for every connection, both as server and
133 client, there is no benefit for these.
134 GnuTLS appears to not support tickets server-side by default (we don't
135 call gnutls_session_ticket_enable_server()) but client side is enabled
136 by default on recent versions (3.1.3 +) unless the PFS priority string
139 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
140 <https://reproducible-builds.org/specs/source-date-epoch/>.
142 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
143 the check for any unsuccessful recipients did not notice the limit, and
144 erroneously found still-pending ones.
146 JH/08 Pipeline CHUNKING command and data together, on kernels that support
147 MSG_MORE. Only in-clear (not on TLS connections).
149 JH/09 Avoid using a temporary file during transport using dkim. Unless a
150 transport-filter is involved we can buffer the headers in memory for
151 creating the signature, and read the spool data file once for the
152 signature and again for transmission.
154 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
155 4.77 as the kernel support then wasn't solid, having issues in 64bit
156 mode. Now, it's been long enough. Add support for FreeBSD also.
158 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
159 case where the routing stage had gathered several addresses to send to
160 a host before calling the transport for the first, we previously failed
161 to close down TLS in the old transport process before passing the TCP
162 connection to the new process. The new one sent a STARTTLS command
163 which naturally failed, giving a failed delivery and bloating the retry
164 database. Investigation and fix prototype from Wolfgang Breyha.
166 JH/12 Fix check on SMTP command input synchronisation. Previously there were
167 false-negatives in the check that the sender had not preempted a response
168 or prompt from Exim (running as a server), due to that code's lack of
169 awareness of the SMTP input buffering.
171 PP/04 Add commandline_checks_require_admin option.
172 Exim drops privileges sanely, various checks such as -be aren't a
173 security problem, as long as you trust local users with access to their
174 own account. When invoked by services which pass untrusted data to
175 Exim, this might be an issue. Set this option in main configuration
176 AND make fixes to the calling application, such as using `--` to stop
179 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
180 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
181 responses to those, into a single TLS record each way (this usually means
182 a single packet). As a side issue, smtp_enforce_sync now works on TLS
185 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
186 affects you only if you're dancing at the edge of the param size limits.
187 If you are, and this message makes sense to you, then: raise the
188 configured limit or use OpenSSL 1.1. Nothing we can do for older
191 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
192 cmdline element to get the documented default one. Previously it was
195 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
196 are unsuable for non-message-specific reasons - eg. connection timeout,
197 banner-time rejection.
199 JH/16 Fix logging of delivery remote port, when specified by router, under
202 PP/06 Repair manualroute's ability to take options in any order, even if one
203 is the name of a transport.
206 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
208 JH/17 Change the list-building routines interface to use the expanding-string
209 triplet model, for better allocation and copying behaviour.
211 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
212 Previously it was constructed the first time a possibly-matching string
213 was met in the configuration file input during startup; now it is done
216 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
217 compatible one, to avoid the (poorly documented) possibility of a config
218 file in the working directory redirecting the DB files, possibly correpting
219 some existing file. CVE-2017-10140 assigned for BDB.
221 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
222 cache-hot. Previously, although the result was properly cached, the
223 initial verify call returned a defer.
225 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
226 the main verify for receipient in uncached-mode.
228 JH/22 Retire historical build files to an "unsupported" subdir. These are
229 defined as "ones for which we have no current evidence of testing".
231 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
232 if present. Previously it was ignored.
234 JH/24 Start using specified-initialisers in C structure init coding. This is
235 a C99 feature (it's 2017, so now considered safe).
237 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
238 if was a fixed-sized field and bitmask ops via macros; it is now more
241 PP/07 GitHub PR 56: Apply MariaDB build fix.
242 Patch provided by Jaroslav Škarvada.
244 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
245 during Coverity cleanups [4.87 JH/47]
246 Diagnosis and fix provided by Michael Fischer v. Mollard.
248 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
249 the right size to place the terminating semicolon on its own folded
250 line, the header hash was calculated to an incorrect value thanks to
251 the (relaxed) space the fold became.
253 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
254 and confused the parent.
256 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
257 which could crash as a result. This could lead to undeliverable messages.
259 JH/28 Logging: "next input sent too soon" now shows where input was truncated
262 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
263 matters on fast-turnover and PID-randomising systems, which were getting
264 out-of-order delivery.
266 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
267 a possibly-overlapping copy. The symptom was that "Remote host closed
268 connection in response to HELO" was logged instead of the actual 4xx
271 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
272 Previously only that bufferd was discarded, resulting in SYMTP command
275 JH/32 DKIM: when a message has multiple signatures matching an identity given
276 in dkim_verify_signers, run the dkim acl once for each. Previously only
277 one run was done. Bug 2189.
279 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
280 "panic the current process" to "deliberately defer". The panic log is
281 still written with the problem list name; the mail and reject logs now
282 get a temp-reject line for the message that was being handled, saying
283 something like "domains check lookup or other defer". The SMTP 451
284 message is still "Temporary local problem".
286 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
287 A crafted sequence of BDAT commands could result in in-use memory beeing
288 freed. CVE-2017-16943.
290 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
291 from SMTP input. Previously it was always done; now only done for DATA
292 and not BDAT commands. CVE-2017-16944.
294 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
295 to the message (such as an overlong header line). Previously this was
296 not done and we did not exit BDAT mode. Followon from the previous item
297 though a different problem.
303 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
304 than -2003 did; needs libidn2 in addition to libidn.
306 JH/02 The path option on a pipe transport is now expanded before use.
308 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
309 Patch provided by "Björn", documentation fix added too.
311 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
312 missing a wire-to-host endian conversion.
314 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
315 close after a BDAT command line could be taken as a following command,
316 giving a synch failure. Fix by only checking for synch immediately
317 before acknowledging the chunk.
319 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
320 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
321 macro. Patches provided by Josh Soref.
323 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
324 Previously we did not; the RFC seems ambiguous and VRFY is not listed
325 by IANA as a service extension. However, John Klensin suggests that we
328 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
329 the dkim code may be unix-mode line endings rather than smtp wire-format
330 CRLF, so prepend a CR to any bare LF.
332 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
333 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
335 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
336 the first were themselves being wrongly included in the feed into dkim
337 processing; with most chunk sizes in use this resulted in an incorrect
338 body hash calculated value.
340 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
341 DKIM signature block, for verification. Although advised against by
342 standards it is specifically not ruled illegal.
344 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
346 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
347 missing a body hash (the bh= tag).
349 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
350 It seems that HAProxy sends the Proxy Protocol information in clear and
351 only then does a TLS startup, so do the same.
353 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
354 TCP connections (such as for Spamd) unless the daemon successfully set
355 Fast Open mode on its listening sockets. This fixes breakage seen on
356 too-old kernels or those not configured for Fast Open, at the cost of
357 requiring both directions being enabled for TFO, and TFO never being used
358 by non-daemon-related Exim processes.
360 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
361 endings, at least on the first header line. Try to canonify any that get
362 past that check, despite the cost.
364 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
365 now limited to an arbitrary five deep, while parsing addresses with the
366 strip_excess_angle_brackets option enabled.
368 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
369 instead leave the unprompted TLS handshake in socket buffer for the
370 TLS library to consume.
372 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
374 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
376 JH/16 Drop variables when they go out of scope. Memory management drops a whole
377 region in one operation, for speed, and this leaves assigned pointers
378 dangling. Add checks run only under the testsuite which checks all
379 variables at a store-reset and panics on a dangling pointer; add code
380 explicitly nulling out all the variables discovered. Fixes one known
381 bug: a transport crash, where a dangling pointer for $sending_ip_address
382 originally assigned in a verify callout, is re-used.
384 PP/06 Drop '.' from @INC in various Perl scripts.
386 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
388 PP/08 Reduce a number of compilation warnings under clang; building with
389 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
390 should be warning-free.
392 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
394 HS/01 Fix portability problems introduced by PP/08 for platforms where
395 realloc(NULL) is not equivalent to malloc() [SunOS et al].
397 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
398 chunk. This allows us to accept broken chunked messages. We need a more
399 general solution here.
401 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
402 already-broken messages in the queue.
404 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
406 JH/19 Fix reference counting bug in routing-generated-address tracking.
412 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
413 supports it and a size is available (ie. the sending peer gave us one).
415 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
416 years of being deprecated). The replacements are the ACLs
417 acl_smtp_mime and acl_not_smtp_mime.
419 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
420 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
421 or even an in-clear connection were permitted. Now, if the host lookup
422 was dnssec and dane was requested then the host is only used if the
423 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
424 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
425 if one fails this test.
426 This means that a poorly-configured remote DNS will make it incommunicado;
427 but it protects against a DNS-interception attack on it.
429 JH/04 Bug 1810: make continued-use of an open smtp transport connection
430 non-noisy when a race steals the message being considered.
432 JH/05 If main configuration option tls_certificate is unset, generate a
433 self-signed certificate for inbound TLS connections.
435 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
436 in rewrites and routers.
438 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
439 and logged a warning sing 4.83; now they are a configuration file error.
441 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
442 (lacking @domain). Apply the same qualification processing as RCPT.
444 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
446 JH/10 Support ${sha256:} applied to a string (as well as the previous
449 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
450 a cutthrough deliver is pending, as we always want to make a connection.
451 This also avoids re-routing the message when later placing the cutthrough
452 connection after a verify cache hit.
453 Do not update it with the verify result either.
455 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
456 when routing results in more than one destination address.
458 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
459 signing (which inhibits the cutthrough capability). Previously only
460 the presence of an option was tested; now an expansion evaluating as
461 empty is permissible (obviously it should depend only on data available
462 when the cutthrough connection is made).
464 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
465 the relevant preceding SMTP command did not note the pipelining mode.
467 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
468 Previously they were not counted.
470 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
471 as one having no matching records. Previously we deferred the message
472 that needed the lookup.
474 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
475 distinguished as "(=".
477 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
478 for missing MX records. Previously it only worked for missing A records.
480 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
482 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
483 after the data-go-ahead and data-ack. Patch from Jason Betts.
485 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
486 even for a "none" policy. Patch from Tony Meyer.
488 JH/22 Fix continued use of a connection for further deliveries. If a port was
489 specified by a router, it must also match for the delivery to be
492 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
493 When one of the recipients of a message was unsuitable for the connection
494 (has no matching addresses), we lost track of needing to mark it
495 deferred. As a result mail would be lost.
497 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
499 JH/25 Decoding ACL controls is now done using a binary search; the source code
500 takes up less space and should be simpler to maintain. Merge the ACL
501 condition decode tables also, with similar effect.
503 JH/26 Fix problem with one_time used on a redirect router which returned the
504 parent address unchanged. A retry would see the parent address marked as
505 delivered, so not attempt the (identical) child. As a result mail would
508 JH/27 Fix a possible security hole, wherein a process operating with the Exim
509 UID can gain a root shell. Credit to http://www.halfdog.net/ for
510 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
513 JH/28 Enable {spool,log} filesystem space and inode checks as default.
514 Main config options check_{log,spool}_{inodes,space} are now
515 100 inodes, 10MB unless set otherwise in the configuration.
517 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
518 Previously it only applied to the main-section connection policy
521 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
523 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
524 by me. Added RFC7919 DH primes as an alternative.
526 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
527 are not in the system include path.
529 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
530 GnuTLS, when a session startup failed (eg because the client disconnected)
531 Exim did stdio operations after fclose. This was exposed by a recent
532 change which nulled out the file handle after the fclose.
534 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
535 signed directly by the cert-signing cert, rather than an intermediate
536 OCSP-signing cert. This is the model used by LetsEncrypt.
538 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
540 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
541 an incoming connection.
543 HS/02 Bug 1802: Do not half-close the connection after sending a request
546 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
547 fallback to "prime256v1".
549 JH/34 SECURITY: Use proper copy of DATA command in error message.
550 Could leak key material. Remotely exploitable. CVE-2016-9963.
556 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
557 and 3.4.4 - once the server is enabled to respond to an OCSP request
558 it does even when not requested, resulting in a stapling non-aware
559 client dropping the TLS connection.
561 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
562 support variable-length bit vectors. No functional change.
564 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
565 The I= interface field on outgoing lines is now after the H= remote
566 host field, same as incoming lines. There is a separate
567 outgoing_interface log selector which allows you to disable the
570 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
571 If not running log_selector +smtp_connection the mainlog would be held
572 open indefinitely after a "too many connections" event, including to a
573 deleted file after a log rotate. Leave the per net connection logging
574 leaving it open for efficiency as that will be quickly detected by the
575 check on the next write.
577 HS/01 Bug 1671: Fix post transport crash.
578 Processing the wait-<transport> messages could crash the delivery
579 process if the message IDs didn't exist for some reason. When
580 using 'split_spool_directory=yes' the construction of the spool
581 file name failed already, exposing the same netto behaviour.
583 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
584 mime_regex ACL conditions.
586 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
587 to DSN fail messages (bounces): remote IP, remote greeting, remote response
588 to HELO, local diagnostic string.
590 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
591 log line to debug. Even when configured with a tls authenticator many
592 client connections are expected to not authenticate in this way, so
593 an authenticate fail is not an error.
595 HS/02 Add the Exim version string to the process info. This way exiwhat
596 gives some more detail about the running daemon.
598 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
599 matter for fast-change records such as DNSBLs.
601 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
602 retry record, even if constant. There may be multiple transports with
603 different interface settings and the retry behaviour needs to be kept
606 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
608 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
610 JH/10 Bug 840: fix log_defer_output option of pipe transport
612 JH/11 Bug 830: use same host for all RCPTS of a message, even under
613 hosts_randomize. This matters a lot when combined with mua_wrapper.
615 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
616 ${quote_pgsql:<string>} operator.
618 JH/13 Bug 1708: avoid misaligned access in cached lookup.
620 JH/14 Change header file name for freeradius-client. Relevant if compiling
621 with Radius support; from the Gentoo tree and checked under Fedora.
623 JH/15 Bug 1712: Introduce $prdr_requested flag variable
625 JH/16 Bug 1714: Permit an empty string as expansion result for transport
626 option transport_filter, meaning no filtering.
628 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
630 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
631 defaults to "*" (all hosts). The variable is now available when not built
632 with TLS, default unset, mainly to enable keeping the testsuite sane.
633 If a server certificate is not supplied (via tls_certificate) an error is
634 logged, and clients will find TLS connections fail on startup. Presumably
635 they will retry in-clear.
636 Packagers of Exim are strongly encouraged to create a server certificate
637 at installation time.
639 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
640 with the $config_file variable.
642 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
643 in transport context, after the attempt, and per-recipient. The latter type
644 is per host attempted. The event data is the error message, and the errno
645 information encodes the lookup type (A vs. MX) used for the (first) host,
646 and the trailing two digits of the smtp 4xx response.
648 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
649 to write to mainlog (or rejectlog, paniclog) in the window between file
650 creation and permissions/ownership being changed. Particularly affects
651 installations where exicyclog is run as root, rather than exim user;
652 result is that the running daemon panics and dies.
654 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
656 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
657 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
658 "pri" and "weight". Note that the previous implicit priority given by the
659 list order is no longer honoured.
661 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
664 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
665 by defining SUPPORT_SOCKS.
667 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
668 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
669 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
670 variables are renamed to proxy_{local,external}_{address,port}.
672 JH/27 Move Internationalisation support from Experimental to mainline, enabled
673 for a build by defining SUPPORT_I18N
675 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
676 of the query string, and make ${quote_redis:} do that quoting.
678 JH/29 Move Events support from Experimental to mainline, enabled by default
679 and removable for a build by defining DISABLE_EVENT.
681 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
683 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
684 cached by the daemon.
686 JH/32 Move Redis support from Experimental to mainline, enabled for a build
687 by defining LOOKUP_REDIS. The libhiredis library is required.
689 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
690 keys are given for lookup.
692 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
693 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
694 only supported when built with TLS support. The PolarSSL SHA routines
695 are still used when the TLS library is too old for convenient support.
697 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
698 openssl_options), for security. OpenSSL forces this from version 1.1.0
699 server-side so match that on older versions.
701 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
702 allocation for $value could be released as the expansion processing
703 concluded, but leaving the global pointer active for it.
705 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
706 and to use the domains and local_parts ACL conditions.
708 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
709 incorrectly not doubled on cutthrough transmission, hence seen as a
710 body-termination at the receiving system - resulting in truncated mails.
711 Commonly the sender saw a TCP-level error, and retransmitted the message
712 via the normal store-and-forward channel. This could result in duplicates
713 received - but deduplicating mailstores were liable to retain only the
714 initial truncated version.
716 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
718 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
720 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
721 we're in there, support oversigning also; bug 1309.
723 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
725 HS/04 Add support for keep_environment and add_environment options.
727 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
728 either intentional arithmetic overflow during PRNG, or testing config-
731 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
732 delivery resulted in actual delivery. Cancel cutthrough before DATA
735 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
736 rejects a recipient: pass the reject to the originator.
738 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
739 Many were false-positives and ignorable, but it's worth fixing the
742 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
743 for the new environment-manipulation done at startup. Move the routines
744 from being local to tls.c to being global via the os.c file.
746 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
747 an extract embedded as result-arg for a map, the first arg for extract
748 is unavailable so we cannot tell if this is a numbered or keyed
749 extraction. Accept either.
755 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
758 JH/02 The smtp transport option "multi_domain" is now expanded.
760 JH/03 The smtp transport now requests PRDR by default, if the server offers
763 JH/04 Certificate name checking on server certificates, when exim is a client,
764 is now done by default. The transport option tls_verify_cert_hostnames
765 can be used to disable this per-host. The build option
766 EXPERIMENTAL_CERTNAMES is withdrawn.
768 JH/05 The value of the tls_verify_certificates smtp transport and main options
769 default to the word "system" to access the system default CA bundle.
770 For GnuTLS, only version 3.0.20 or later.
772 JH/06 Verification of the server certificate for a TLS connection is now tried
773 (but not required) by default. The verification status is now logged by
774 default, for both outbound TLS and client-certificate supplying inbound
777 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
780 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
781 Status Notification (bounce) messages are now MIME format per RFC 3464.
782 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
783 under the control of the dsn_advertise_hosts option, and routers may
784 have a dsn_lasthop option.
786 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
787 default, modifiable by a malware= option. The list separator for
788 the options can now be changed in the usual way. Bug 68.
790 JH/10 The smtp_receive_timeout main option is now expanded before use.
792 JH/11 The incoming_interface log option now also enables logging of the
793 local interface on delivery outgoing connections.
795 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
796 if the interface and destination host and port all match.
798 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
801 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
802 Patch from Andrew Lewis.
804 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
805 now supports optional time-restrictions, weighting, and priority
806 modifiers per server. Patch originally by <rommer@active.by>.
808 JH/16 The spamd_address main option now supports a mixed list of local
809 and remote servers. Remote servers can be IPv6 addresses, and
810 specify a port-range.
812 JH/17 Bug 68: The spamd_address main option now supports an optional
813 timeout value per server.
815 JH/18 Bug 1581: Router and transport options headers_add/remove can
816 now have the list separator specified.
818 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
821 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
824 JH/21 Support for the A6 type of dns record is withdrawn.
826 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
827 rather than the verbs used.
829 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
830 from 255 to 1024 chars.
832 JH/24 Verification callouts now attempt to use TLS by default.
834 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
835 are generic router options now. The defaults didn't change.
837 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
838 Original patch from Alexander Shikoff, worked over by JH.
840 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
841 files if ZCAT_COMMAND is not executable.
843 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
845 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
847 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
848 Normally benign, it bites when the pair was led to by a CNAME;
849 modern usage is to not canonicalize the domain to a CNAME target
850 (and we were inconsistent anyway for A-only vs AAAA+A).
852 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
854 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
855 when evaluating $sender_host_dnssec.
857 JH/31 Check the HELO verification lookup for DNSSEC, adding new
858 $sender_helo_dnssec variable.
860 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
862 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
864 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
866 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
867 documented as working, but never had. Support all but $spam_report.
869 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
870 added for tls authenticator.
872 HS/03 Add perl_taintmode main config option
878 TL/01 When running the test suite, the README says that variables such as
879 no_msglog_check are global and can be placed anywhere in a specific
880 test's script, however it was observed that placement needed to be near
881 the beginning for it to behave that way. Changed the runtest perl
882 script to read through the entire script once to detect and set these
883 variables, reset to the beginning of the script, and then run through
884 the script parsing/test process like normal.
886 TL/02 The BSD's have an arc4random API. One of the functions to induce
887 adding randomness was arc4random_stir(), but it has been removed in
888 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
889 function when detected.
891 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
892 cause callback expansion.
894 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
895 syntax errors in an expansion can be treated as a string instead of
896 logging or causing an error, due to the internal use of bool_lax
897 instead of bool when processing it.
899 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
900 server certificates when making smtp deliveries.
902 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
904 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
906 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
908 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
909 Merged patch from Sebastian Wiedenroth.
911 JH/05 Fix results-pipe from transport process. Several recipients, combined
912 with certificate use, exposed issues where response data items split
913 over buffer boundaries were not parsed properly. This eventually
914 resulted in duplicates being sent. This issue only became common enough
915 to notice due to the introduction of connection certificate information,
916 the item size being so much larger. Found and fixed by Wolfgang Breyha.
918 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
919 size buffer was used, resulting in syntax errors when an expansion
922 JH/07 Add support for directories of certificates when compiled with a GnuTLS
923 version 3.3.6 or later.
925 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
926 is EXPERIMENTAL_EVENT, the main-configuration and transport options
927 both become "event_action", the variables become $event_name, $event_data
928 and $event_defer_errno. There is a new variable $verify_mode, usable in
929 routers, transports and related events. The tls:cert event is now also
930 raised for inbound connections, if the main configuration event_action
933 TL/06 In test suite, disable OCSP for old versions of openssl which contained
934 early OCSP support, but no stapling (appears to be less than 1.0.0).
936 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
937 server certificate names available under the smtp transport option
938 "tls_verify_cert_hostname" now do not permit multi-component wildcard
941 JH/10 Time-related extraction expansions from certificates now use the main
942 option "timezone" setting for output formatting, and are consistent
943 between OpenSSL and GnuTLS compilations. Bug 1541.
945 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
946 encoded parameter in the incoming message. Bug 1558.
948 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
949 include certificate info, eximon was claiming there were spoolfile
952 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
954 JH/14 Log delivery-related information more consistently, using the sequence
955 "H=<name> [<ip>]" wherever possible.
957 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
958 are problematic for Debian distribution, omit them from the release
961 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
963 JH/16 Fix string representation of time values on 64bit time_t architectures.
966 JH/17 Fix a null-indirection in certextract expansions when a nondefault
967 output list separator was used.
972 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
973 checkers that were complaining about end of non-void function with no
976 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
977 This was a regression introduced in 4.83 by another bugfix.
979 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
981 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
982 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
988 TF/01 Correctly close the server side of TLS when forking for delivery.
990 When a message was received over SMTP with TLS, Exim failed to clear up
991 the incoming connection properly after forking off the child process to
992 deliver the message. In some situations the subsequent outgoing
993 delivery connection happened to have the same fd number as the incoming
994 connection previously had. Exim would try to use TLS and fail, logging
995 a "Bad file descriptor" error.
997 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
998 utilities have not been installed.
1000 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1001 temporary space as the ACL may create new global variables.
1003 TL/01 LDAP support uses per connection or global context settings, depending
1004 upon the detected version of the libraries at build time.
1006 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1007 to extract and use the src ip:port in logging and expansions as if it
1008 were a direct connection from the outside internet. PPv2 support was
1009 updated based on HAProxy spec change in May 2014.
1011 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1013 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1014 Properly escape header and check for NULL return.
1016 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1019 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1021 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1022 characters in header names, implemented as a verify condition.
1023 Contributed by Michael Fischer v. Mollard.
1025 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1026 results permerror and temperror. Previous values are deprecated but
1027 still accepted. In a future release, err_perm and err_temp will be
1028 completely removed, which will be a backward incompatibility if the
1029 ACL tests for either of these two old results. Patch contributed by
1030 user bes-internal on the mailing list.
1032 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1034 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1035 selectors, in both main and reject logs.
1037 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1040 JH/07 Add malware type "sock" for talking to simple daemon.
1042 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1044 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1045 routers/transports under cutthrough routing.
1047 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1048 numbers. Touch up "bool" conditional to keep the same definition.
1050 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1052 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1054 JH/12 Expand items in router/transport headers_add or headers_remove lists
1055 individually rather than the list as a whole. Bug 1452.
1057 Required for reasonable handling of multiple headers_ options when
1058 they may be empty; requires that headers_remove items with embedded
1059 colons must have them doubled (or the list-separator changed).
1061 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1062 view the policy declared in the DMARC record. Currently, $dmarc_status
1063 is a combined value of both the record presence and the result of the
1066 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1068 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1069 dnslookup router and the smtp transport (applying to the forward
1072 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1073 of ldap servers used for a specific lookup. Patch provided by Heiko
1076 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1077 New variable $lookup_dnssec_authenticated for observability.
1079 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1080 Patch submitted by Lars Timman.
1082 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1084 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1085 Requires trusted mode and valid format message id, aborts otherwise.
1086 Patch contributed by Heiko Schlichting.
1088 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1089 certextract with support for various fields. Bug 1358.
1091 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1092 is requested by default, modifiable by smtp transport option
1095 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1096 operate on certificate variables to give certificate fingerprints
1097 Also new ${sha256:cert_variable}.
1099 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1101 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1102 Christian Aistleitner.
1104 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1106 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1107 file. Patch from Wolfgang Breyha.
1109 JH/25 Expand the coverage of the delivery $host and $host_address to
1110 client authenticators run in verify callout. Bug 1476.
1112 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1113 align with daemon_smtp_ports. Bug 72.
1115 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1116 support and error reporting did not work properly.
1118 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1119 and is readable. Patch from Andrew Colin Kissa.
1121 TL/14 Enhance documentation of ${run expansion and how it parses the
1122 commandline after expansion, particularly in the case when an
1123 unquoted variable expansion results in an empty value.
1125 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1127 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1128 by renaming away from it. Addresses GH issue 15, reported by
1131 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1133 TL/15 SECURITY: prevent double expansion in math comparison functions
1134 (can expand unsanitized data). Not remotely exploitable.
1141 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1143 PP/02 Make -n do something, by making it not do something.
1144 When combined with -bP, the name of an option is not output.
1146 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1149 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1150 $sender_host_name and config options to manage this, and basic check
1153 PP/05 DSCP support for outbound connections and control modifier for inbound.
1155 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1156 (Only plugin which currently uses this is kerberos4, which nobody should
1157 be using, but we should make it available and other future plugins might
1158 conceivably use it, even though it would break NAT; stuff *should* be
1159 using channel bindings instead).
1161 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1162 name; added for Sendmail compatibility; requires admin caller.
1163 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1164 just ignore it); requires trusted caller.
1165 Also parse but ignore: -Ac -Am -X<logfile>
1168 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1170 TL/02 Add +smtp_confirmation as a default logging option.
1172 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1173 Patch by Magnus Holmgren from 2007-02-20.
1175 TL/04 Bugzilla 1281 - Spec typo.
1176 Bugzilla 1283 - Spec typo.
1177 Bugzilla 1290 - Spec grammar fixes.
1179 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1181 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1183 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1184 mailing list by by Dmitry Isaikin.
1186 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1188 JH/02 Support "G" suffix to numbers in ${if comparisons.
1190 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1192 NM/01 Bugzilla 1197 - Spec typo
1193 Bugzilla 1196 - Spec examples corrections
1195 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1197 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1198 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1201 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1202 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1204 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1205 "acl {{name}{arg}...}", and optional args on acl condition
1208 JH/05 Permit multiple router/transport headers_add/remove lines.
1210 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1212 JH/07 Avoid using a waiting database for a single-message-only transport.
1213 Performance patch from Paul Fisher. Bugzilla 1262.
1215 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1218 JH/09 Add $headers_added variable, with content from use of ACL modifier
1219 add_header (but not yet added to the message). Bugzilla 199.
1221 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1222 Pulled from Bugzilla 817 by Wolfgang Breyha.
1224 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1226 (nb: this is the same fix as in Exim 4.80.1)
1228 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1231 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1234 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1236 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1237 advertise SMTP AUTH mechanism to us, instead of a generic
1238 protocol violation error. Also, make Exim more robust to bad
1239 data from the Dovecot auth socket.
1241 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1243 When a queue runner is handling a message, Exim first routes the
1244 recipient addresses, during which it prunes them based on the retry
1245 hints database. After that it attempts to deliver the message to
1246 any remaining recipients. It then updates the hints database using
1249 So if a recipient address works intermittently, it can get repeatedly
1250 deferred at routing time. The retry hints record remains fresh so the
1251 address never reaches the final cutoff time.
1253 This is a fairly common occurrence when a user is bumping up against
1254 their storage quota. Exim had some logic in its local delivery code
1255 to deal with this. However it did not apply to per-recipient defers
1256 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1258 This change adds a proper retry rule check during routing so that the
1259 final cutoff time is checked against the message's age. We only do
1260 this check if there is an address retry record and there is not a
1261 domain retry record; this implies that previous attempts to handle
1262 the address had the retry_use_local_parts option turned on. We use
1263 this as an approximation for the destination being like a local
1264 delivery, as in LMTP.
1266 I suspect this new check makes the old local delivery cutoff check
1267 redundant, but I have not verified this so I left the code in place.
1269 TF/02 Correct gecos expansion when From: is a prefix of the username.
1271 Test 0254 submits a message to Exim with the header
1275 When I ran the test suite under the user fanf2, Exim expanded
1276 the header to contain my full name, whereas it should have added
1277 a Resent-Sender: header. It erroneously treats any prefix of the
1278 username as equal to the username.
1280 This change corrects that bug.
1282 GF/01 DCC debug and logging tidyup
1283 Error conditions log to paniclog rather than rejectlog.
1284 Debug lines prefixed by "DCC: " to remove any ambiguity.
1286 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1288 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1289 Bug spotted by Jeremy Harris; was flawed since initial commit.
1290 Would have resulted in OCSP responses post-SNI triggering an Exim
1291 NULL dereference and crash.
1293 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1295 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1296 Bug detection, analysis and fix by Samuel Thibault.
1297 Bugzilla 1331, Debian bug #698092.
1299 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1301 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1302 Server implementation by Todd Lyons, client by JH.
1303 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1304 config variable "prdr_enable" controls whether the server
1305 advertises the facility. If the client requests PRDR a new
1306 acl_data_smtp_prdr ACL is called once for each recipient, after
1307 the body content is received and before the acl_smtp_data ACL.
1308 The client is controlled by both of: a hosts_try_prdr option
1309 on the smtp transport, and the server advertisement.
1310 Default client logging of deliveries and rejections involving
1311 PRDR are flagged with the string "PRDR".
1313 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1314 fclose(). Diagnosis by Todd Lyons.
1316 PP/17 Update configure.default to handle IPv6 localhost better.
1317 Patch by Alain Williams (plus minor tweaks).
1320 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1321 This is now consistent with GnuTLS, and is now documented: the
1322 previous undocumented portable approach to treating the option as
1323 unset was to force an expansion failure. That still works, and
1324 an empty string is now equivalent.
1326 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1327 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1328 not performing validation itself.
1330 PP/20 Added force_command boolean option to pipe transport.
1331 Patch from Nick Koston, of cPanel Inc.
1333 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1336 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1338 PP/21 Fix eximon continuous updating with timestamped log-files.
1339 Broken in a format-string cleanup in 4.80, missed when I repaired the
1340 other false fix of the same issue.
1341 Report and fix from Heiko Schlichting.
1344 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1345 Report from Prashanth Katuri.
1347 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1348 It's SecureTransport, so affects any MacOS clients which use the
1349 system-integrated TLS libraries, including email clients.
1351 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1352 using a MIME ACL for non-SMTP local injection.
1353 Report and assistance in diagnosis by Warren Baker.
1355 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1357 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1359 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1360 last id that failed so it may be referenced in subsequent ACL's.
1362 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1365 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1366 ldap library initialization, allowing self-signed CA's to be
1367 used. Also properly sets require_cert option later in code by
1368 using NULL (global ldap config) instead of ldap handle (per
1369 session). Bug diagnosis and testing by alxgomz.
1371 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1372 the src/util/ subdirectory.
1374 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1375 renamed to Transport Post Delivery Action by Jeremy Harris, as
1378 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1379 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1380 redis_servers = needs to be configured which will be used by the redis
1381 lookup. Patch from Warren Baker, of The Packet Hub.
1383 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1385 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1386 hostname or reverse DNS when processing a host list. Used suggestions
1387 from multiple comments on this bug.
1389 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1391 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1392 Missed a few lines, added it to make the runtest require no keyboard
1395 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1396 contains upper case chars. Make router use caseful_local_part.
1398 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1399 support when GnuTLS has been built with p11-kit.
1405 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1407 This, or similar/improved, will also be change PP/11 of 4.82.
1413 PP/01 Handle short writes when writing local log-files.
1414 In practice, only affects FreeBSD (8 onwards).
1415 Bugzilla 1053, with thanks to Dmitry Isaikin.
1417 NM/01 Bugzilla 949 - Documentation tweak
1419 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1422 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1424 PP/02 Implemented gsasl authenticator.
1426 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1428 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1429 `pkg-config foo` for cflags/libs.
1431 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1432 with rest of GSASL and with heimdal_gssapi.
1434 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1435 `pkg-config foo` for cflags/libs for the TLS implementation.
1437 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1438 properties get this fed in as external SSF. A number of robustness
1439 and debugging improvements to the cyrus_sasl authenticator.
1441 PP/08 cyrus_sasl server now expands the server_realm option.
1443 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1444 Patch by Jeremy Harris.
1446 PP/10 Added dbmjz lookup type.
1448 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1450 PP/12 MAIL args handles TAB as well as SP, for better interop with
1451 non-compliant senders.
1452 Analysis and variant patch by Todd Lyons.
1454 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1455 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1456 Patch from Dirk Mueller <dmueller@suse.com>
1458 PP/13 tls_peerdn now print-escaped for spool files.
1459 Observed some $tls_peerdn in wild which contained \n, which resulted
1460 in spool file corruption.
1462 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1463 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1464 or write after TLS renegotiation, which otherwise led to messages
1467 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1468 as a tracking header (ie: a signed header comes before the signature).
1469 Patch from Wolfgang Breyha.
1471 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1472 comma-sep list; embedded commas doubled.
1474 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1476 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1478 Report and patch from Dmitry Banschikov.
1480 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1481 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1482 failures. We appear to now support TLS1.1+ with Exim.
1484 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1485 lets Exim select keys and certificates based upon TLS SNI from client.
1486 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1487 before an outbound SMTP session. New log_selector, +tls_sni.
1489 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1490 NULL dereference. Report and patch from Alun Jones.
1492 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1493 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1494 Not seeing resolver debug output on NetBSD, but suspect this is a
1495 resolver implementation change.
1497 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1498 Left warnings. Added "eximon gdb" invocation mode.
1500 PP/21 Defaulting "accept_8bitmime" to true, not false.
1502 PP/22 Added -bw for inetd wait mode support.
1504 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1505 locate the relevant includes and libraries. Made this the default.
1507 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1508 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1510 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1511 This may cause build issues on older platforms.
1513 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1514 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1515 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1516 Added SNI support via GnuTLS too.
1517 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1519 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1521 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1522 Applied second patch from Janne, implementing suggestion to default
1523 multiple-strings-in-record handling to match SPF spec.
1525 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1527 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1528 read-only, out of scope).
1529 Patch from Wolfgang Breyha, report from Stuart Northfield.
1531 PP/29 Fix three issues highlighted by clang analyser static analysis.
1532 Only crash-plausible issue would require the Cambridge-specific
1533 iplookup router and a misconfiguration.
1534 Report from Marcin Mirosław.
1536 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1538 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1539 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1540 As part of this, removing so much warning spew let me fix some minor
1541 real issues in debug logging.
1543 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1544 assignment on my part. Fixed.
1546 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1547 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1548 Janne Snabb (who went above and beyond: thank you).
1550 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1551 string otherwise requires a connection and a bunch more work and it's
1552 relatively easy to get wrong. Should also expose TLS library linkage
1555 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1556 64-bit ${eval} (JH/03).
1558 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1559 GNU libc to support some of the 64-bit stuff, should not lead to
1560 conflicts. Defined before os.h is pulled in, so if a given platform
1561 needs to override this, it can.
1563 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1564 protection layer was required, which is not implemented.
1565 Bugzilla 1254, patch from Wolfgang Breyha.
1567 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1568 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1569 tls_dhparam take prime identifiers. Also unbreak combination of
1570 OpenSSL+DH_params+TLSSNI.
1572 PP/39 Disable SSLv2 by default in OpenSSL support.
1578 PP/01 Solaris build fix for Oracle's LDAP libraries.
1579 Bugzilla 1109, patch from Stephen Usher.
1581 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1583 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1586 TF/02 Fix a couple more cases where we did not log the error message
1587 when unlink() failed. See also change 4.74-TF/03.
1589 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1590 lock up or crash if it happened to be inside a call to libc when it
1591 got a SIGUSR1 from exiwhat.
1593 The SIGUSR1 handler appends the current process status to the process
1594 log which is later printed by exiwhat. It used to use the general
1595 purpose logging code to do this, but several functions it calls are
1596 not safe for signals.
1598 The new output code in the SIGUSR1 handler is specific to the process
1599 log, and simple enough that it's easy to inspect for signal safety.
1600 Removing some special cases also simplifies the general logging code.
1601 Removing the spurious timestamps from the process log simplifies
1604 TF/04 Improved ratelimit ACL condition.
1606 The /noupdate option has been deprecated in favour of /readonly which
1607 has clearer semantics. The /leaky, /strict, and /readonly update modes
1608 are mutually exclusive. The update mode is no longer included in the
1609 database key; it just determines when the database is updated. (This
1610 means that when you upgrade Exim will forget old rate measurements.)
1612 Exim now checks that the per_* options are used with an update mode that
1613 makes sense for the current ACL. For example, when Exim is processing a
1614 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1615 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1616 must specify per_mail/readonly. If you omit the update mode it defaults to
1617 /leaky where that makes sense (as before) or /readonly where required.
1619 The /noupdate option is now undocumented but still supported for
1620 backwards compatibility. It is equivalent to /readonly except that in
1621 ACLs where /readonly is required you may specify /leaky/noupdate or
1622 /strict/noupdate which are treated the same as /readonly.
1624 A useful new feature is the /count= option. This is a generalization
1625 of the per_byte option, so that you can measure the throughput of other
1626 aggregate values. For example, the per_byte option is now equivalent
1627 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1629 The per_rcpt option has been generalized using the /count= mechanism
1630 (though it's more complicated than the per_byte equivalence). When it is
1631 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1632 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1633 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1634 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1635 Note that using per_rcpt with a non-readonly update mode in more than
1636 one ACL will cause the recipients to be double-counted. (The per_mail
1637 and per_byte options don't have this problem.)
1639 The handling of very low rates has changed slightly. If the computed rate
1640 is less than the event's count (usually one) then this event is the first
1641 after a long gap. In this case the rate is set to the same as this event's
1642 count, so that the first message of a spam run is counted properly.
1644 The major new feature is a mechanism for counting the rate of unique
1645 events. The new per_addr option counts the number of different
1646 recipients that someone has sent messages to in the last time period. It
1647 behaves like per_rcpt if all the recipient addresses are different, but
1648 duplicate recipient addresses do not increase the measured rate. Like
1649 the /count= option this is a general mechanism, so the per_addr option
1650 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1651 example, measure the rate that a client uses different sender addresses
1652 with the options per_mail/unique=$sender_address. There are further
1653 details in the main documentation.
1655 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1657 TF/06 Removed a few PCRE remnants.
1659 TF/07 Automatically extract Exim's version number from tags in the git
1660 repository when doing development or release builds.
1662 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1663 Bugzilla 879. Patch from Paul Fisher.
1665 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1666 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1669 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1671 PP/05 Variable $av_failed, true if the AV scanner deferred.
1672 Bugzilla 1078. Patch from John Horne.
1674 PP/06 Stop make process more reliably on build failure.
1675 Bugzilla 1087. Patch from Heiko Schlittermann.
1677 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1678 Bugzilla 1089. Patch from Heiko Schlittermann.
1680 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1681 Bugzilla 1131. Patch from Holger Weiß.
1683 PP/09 Handle IPv6 addresses with SPF.
1684 Bugzilla 860. Patch from Wolfgang Breyha.
1686 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1688 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1691 PP/11 match_* no longer expand right-hand-side by default.
1692 New compile-time build option, EXPAND_LISTMATCH_RHS.
1693 New expansion conditions, "inlist", "inlisti".
1695 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1697 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1699 PP/14 fix log_write() format string regression from TF/03.
1700 Bugzilla 1152. Patch from Dmitry Isaikin.
1706 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1708 PP/02 Harmonised TLS library version reporting; only show if debugging.
1709 Layout now matches that introduced for other libraries in 4.74 PP/03.
1711 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1713 PP/04 New "dns_use_edns0" global option.
1715 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1718 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1719 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1721 TK/01 Updated PolarSSL code to 0.14.2.
1722 Bugzilla 1097. Patch from Andreas Metzler.
1724 PP/07 Catch divide-by-zero in ${eval:...}.
1725 Fixes bugzilla 1102.
1727 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1730 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1731 format-string attack -- SECURITY: remote arbitrary code execution.
1733 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1734 time unintentionally subject to list matching rules, letting the header
1735 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1736 arbitrary string expansion). This allowed for information disclosure.
1738 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1739 INT_MIN/-1 -- value coerced to INT_MAX.
1745 NM/01 Workaround for PCRE version dependency in version reporting
1748 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1749 This fixes portability to compilers other than gcc, notably
1750 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1752 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1753 makefiles for portability to HP-UX and POSIX correctness.
1755 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1756 Also via indented variable definition in the Makefile.
1757 (Debugging by Oliver Heesakkers).
1759 PP/02 Restore caching of spamd results with expanded spamd_address.
1760 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1762 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1763 Improves build reliability. Fix from: Frank Elsner
1765 NM/02 Fix wide character breakage in the rfc2047 coding
1766 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1768 NM/03 Allow underscore in dnslist lookups
1769 Fixes bug 1026. Patch from Graeme Fowler
1771 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1772 Code patches from Adam Ciarcinski of NetBSD.
1774 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1777 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1778 is logged, to avoid truncation. Patch from John Horne.
1780 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1781 Patch from Jakob Hirsch.
1783 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1784 SQL string expansion failure details.
1785 Patch from Andrey Oktyabrski.
1787 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1788 Patch from Simon Arlott.
1790 PP/09 New lookups functionality failed to compile on old gcc which rejects
1791 extern declarations in function scope.
1792 Patch from Oliver Fleischmann
1794 PP/10 Use sig_atomic_t for flags set from signal handlers.
1795 Check getgroups() return and improve debugging.
1796 Fixed developed for diagnosis in bug 927 (which turned out to be
1799 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1800 Patch from Mark Zealey.
1802 PP/12 Bugzilla 1056: Improved spamd server selection.
1803 Patch from Mark Zealey.
1805 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1806 Based on patch from Heiko Schlittermann.
1808 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1809 Patch from Uwe Doering, sign-off by Michael Haardt.
1811 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1812 variable declaration deep within a block. Bug and patch from
1815 PP/15 lookups-Makefile IRIX compatibility coercion.
1817 PP/16 Make DISABLE_DKIM build knob functional.
1819 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1820 Patch by Simon Arlott
1822 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1823 variable argument macros. Our copy now differs from upstream.
1829 TF/01 Failure to get a lock on a hints database can have serious
1830 consequences so log it to the panic log.
1832 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1833 controlled using the smtp_confirmation log selector.
1835 TF/03 Include the error message when we fail to unlink a spool file.
1837 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1838 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1839 for maintaining out-of-tree patches for some time.
1841 PP/01 Bugzilla 139: Documentation and portability issues.
1842 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1843 Handle per-OS dynamic-module compilation flags.
1845 PP/02 Let /dev/null have normal permissions.
1846 The 4.73 fixes were a little too stringent and complained about the
1847 permissions on /dev/null. Exempt it from some checks.
1848 Reported by Andreas M. Kirchwitz.
1850 PP/03 Report version information for many libraries, including
1851 Exim version information for dynamically loaded libraries. Created
1852 version.h, now support a version extension string for distributors
1853 who patch heavily. Dynamic module ABI change.
1855 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1856 privilege escalation vulnerability whereby the Exim run-time user
1857 can cause root to append content of the attacker's choosing to
1860 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1863 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1864 If dropping privileges for untrusted macros, we disabled normal logging
1865 on the basis that it would fail; for the Exim run-time user, this is not
1866 the case, and it resulted in successful deliveries going unlogged.
1867 Fixed. Reported by Andreas Metzler.
1873 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1874 only prepend for the Resent-* case. Fixes regression introduced in
1875 Exim 4.70 by NM/22 for Bugzilla 607.
1877 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1878 increasing numbers of administrators be bitten by this.
1880 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1882 PP/03 Bugzilla 994: added openssl_options main configuration option.
1884 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1886 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1888 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1890 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1891 without a peer certificate, leading to a segfault because of an
1892 assumption that peers always have certificates. Be a little more
1893 paranoid. Problem reported by Martin Tscholak.
1895 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1896 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1897 NB: ClamAV planning to remove STREAM in "middle of 2010".
1898 CL also introduces -bmalware, various -d+acl logging additions and
1899 more caution in buffer sizes.
1901 PP/09 Implemented reverse_ip expansion operator.
1903 PP/10 Bugzilla 937: provide a "debug" ACL control.
1905 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1907 PP/12 Bugzilla 973: Implement --version.
1909 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1911 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1913 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1915 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1916 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1917 ignore trailing whitespace.
1919 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1921 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1924 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1925 Notification from Dr Andrew Aitchison.
1927 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1928 ExtendedDetectionInfo response format.
1929 Notification from John Horne.
1931 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1934 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1935 XSL and documented dependency on system catalogs, with examples of how
1938 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1941 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1942 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1943 configuration file which is writeable by the Exim user or group.
1945 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1946 of configuration files to cover files specified with the -C option if
1947 they are going to be used with root privileges, not just the default
1950 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1951 option (effectively making it always true).
1953 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1954 files to be used while preserving root privileges.
1956 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1957 that rogue child processes cannot use them.
1959 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1960 run-time user, instead of root.
1962 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1963 Exim run-time user without dropping privileges.
1965 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1966 result string, instead of calling string_vformat() twice with the same
1969 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1970 for other users. Others should always drop root privileges if they use
1971 -C on the command line, even for a whitelisted configure file.
1973 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1975 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1981 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1982 $data_path, and $header_path variables; fixed documentation bugs and
1985 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1986 exipick to access non-standard spools, including the "frozen" queue
1989 NM/01 Bugzilla 965: Support mysql stored procedures.
1990 Patch from Alain Williams
1992 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1994 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1995 Patch from Andreas Metzler
1997 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1998 Patch from Kirill Miazine
2000 NM/05 Bugzilla 671: Added umask to procmail example.
2002 JJ/03 installed exipick 20100323.0, fixing doc bug
2004 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2005 directory. Notification and patch from Dan Rosenberg.
2007 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2009 TK/02 Improve log output when DKIM signing operation fails.
2011 MH/01 Treat the transport option dkim_domain as a colon separated
2012 list, not as a single string, and sign the message with each element,
2013 omitting multiple occurences of the same signer.
2015 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2016 Bugzilla 985, 986. Patch by Simon Arlott
2018 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2019 Patch by Simon Arlott
2021 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2022 MBX locking. Notification from Dan Rosenberg.
2028 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2030 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2032 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2034 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2036 NM/04 Bugzilla 915: Flag broken perl installation during build.
2042 TK/01 Added patch by Johannes Berg that expands the main option
2043 "spamd_address" if it starts with a dollar sign.
2045 TK/02 Write list of recipients to X-Envelope-Sender header when building
2046 the mbox-format spool file for content scanning (suggested by Jakob
2049 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2050 (http://www.dcc-servers.net/) support via dccifd. Activated by
2051 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2053 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2054 by Mark Daniel Reidel <mr@df.eu>.
2056 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2057 When building exim an external PCRE library is now needed -
2058 PCRE is a system library on the majority of modern systems.
2059 See entry on PCRE_LIBS in EDITME file.
2061 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2062 conversation. Added nologin parameter to request.
2063 Patch contributed by Kirill Miazine.
2065 TF/01 Do not log submission mode rewrites if they do not change the address.
2067 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2069 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2070 log files in place. Contributed by Roberto Lima.
2072 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2074 TF/03 Bugzilla 615: When checking the local_parts router precondition
2075 after a local_part_suffix or local_part_prefix option, Exim now
2076 does not use the address's named list lookup cache, since this
2077 contains cached lookups for the whole local part.
2079 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2080 Robert Millan. Documentation is in experimental-spec.txt.
2082 TF/04 Bugzilla 668: Fix parallel build (make -j).
2084 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2086 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2087 Patch provided by Jan Srzednicki.
2089 TF/05 Leading white space used to be stripped from $spam_report which
2090 wrecked the formatting. Now it is preserved.
2092 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2093 that they are available at delivery time.
2095 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2097 TF/08 TLS error reporting now respects the incoming_interface and
2098 incoming_port log selectors.
2100 TF/09 Produce a more useful error message if an SMTP transport's hosts
2101 setting expands to an empty string.
2103 NM/06 Bugzilla 744: EXPN did not work under TLS.
2104 Patch provided by Phil Pennock.
2106 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2107 Patch provided by Richard Godbee.
2109 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2110 acl_smtp_notquit, added index entry.
2112 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2113 Patch provided by Eugene Bujak.
2115 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2116 accept(). Patch provided by Maxim Dounin.
2118 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2119 Patch provided by Phil Pennock.
2121 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2123 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2124 Patch provided by Brad "anomie" Jorsch.
2126 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2127 Patch provided by Dean Brooks.
2129 TK/05 Add native DKIM support (does not depend on external libraries).
2131 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2132 Patch provided by Graeme Fowler.
2134 NM/16 Bugzilla 851: Documentation example syntax fix.
2136 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2138 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2141 NM/19 Bugzilla 745: TLS version reporting.
2142 Patch provided by Phil Pennock.
2144 NM/20 Bugzilla 167: bool: condition support.
2145 Patch provided by Phil Pennock.
2147 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2148 clients. Patch provided by Phil Pennock.
2150 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2151 Patch provided by Brad "anomie" Jorsch.
2153 NM/23 Bugzilla 687: Fix misparses in eximstats.
2154 Patch provided by Heiko Schlittermann.
2156 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2157 Patch provided by Heiko Schlittermann.
2159 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2160 plus update to original patch.
2162 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2164 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2165 Patch provided by David Brownlee.
2167 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2169 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2171 NM/30 Bugzilla 888: TLS documentation bugfixes.
2173 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2175 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2176 Unlike the original bugzilla I have changed all shell scripts in src tree.
2178 NM/33 Bugzilla 898: Transport filter timeout fix.
2179 Patch by Todd Rinaldo.
2181 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2182 Patch by Serge Demonchaux.
2184 NM/35 Bugzilla 39: Base64 decode bug fixes.
2185 Patch by Jakob Hirsch.
2187 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2189 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2191 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2193 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2199 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2200 ALT-N's libdkim that I have put here:
2201 http://duncanthrax.net/exim-experimental/
2203 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2204 were called 'true' and it seems that C99 defines that as a reserved
2205 keyword to be used with 'bool' variable types. That means you could
2206 not include C99-style headers which use bools without triggering
2207 build errors in sieve.c.
2209 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2210 as mailq or other aliases. Changed the --help handling significantly
2211 to do whats expected. exim_usage() emits usage/help information.
2213 SC/01 Added the -bylocaldomain option to eximstats.
2215 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2217 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2219 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2225 PH/01 Another patch from the Sieve maintainer.
2227 PH/02 When an IPv6 address is converted to a string for single-key lookup
2228 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2229 dots are used instead of colons so that keys in lsearch files need not
2230 contain colons. This was done some time before quoting was made available
2231 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2232 (notated using the quote facility) so as to distinguish them from IPv4
2233 keys. This meant that lookups for IP addresses in host lists did not work
2234 for iplsearch lookups.
2236 This has been fixed by arranging for IPv6 addresses to be expressed with
2237 colons if the lookup type is iplsearch. This is not incompatible, because
2238 previously such lookups could never work.
2240 The situation is now rather anomalous, since one *can* have colons in
2241 ordinary lsearch keys. However, making the change in all cases is
2242 incompatible and would probably break a number of configurations.
2244 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2247 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2248 conversion specification without a maximum field width, thereby enabling
2249 a rogue spamd server to cause a buffer overflow. While nobody in their
2250 right mind would setup Exim to query an untrusted spamd server, an
2251 attacker that gains access to a server running spamd could potentially
2252 exploit this vulnerability to run arbitrary code as the Exim user.
2254 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2255 $primary_hostname instead of what libspf2 thinks the hosts name is.
2257 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2258 a directory entry by the name of the lookup key. Previously, if a
2259 symlink pointed to a non-existing file or a file in a directory that
2260 Exim lacked permissions to read, a lookup for a key matching that
2261 symlink would fail. Now it is enough that a matching directory entry
2262 exists, symlink or not. (Bugzilla 503.)
2264 PH/03 The body_linecount and body_zerocount variables are now exported in the
2267 PH/04 Added the $dnslist_matched variable.
2269 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2270 This means they are set thereafter only if the connection becomes
2273 PH/06 Added the client_condition to authenticators so that some can be skipped
2274 by clients under certain conditions.
2276 PH/07 The error message for a badly-placed control=no_multiline_responses left
2277 "_responses" off the end of the name.
2279 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2281 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2282 (without spaces) instead of just copying the configuration text.
2284 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2286 PH/11 Added $max_received_linelength.
2288 PH/12 Added +ignore_defer and +include_defer to host lists.
2290 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2291 way in which PCRE > 7.0 is built.
2293 PH/14 Implemented queue_only_load_latch.
2295 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2296 MAIL command. The effect was to mangle the value on 64-bit systems.
2298 PH/16 Another patch from the Sieve maintainer.
2300 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2302 PH/18 If a system quota error occurred while trying to create the file for
2303 a maildir delivery, the message "Mailbox is full" was not appended to the
2304 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2305 only to a quota excession during the actual writing of the file.
2307 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2308 characters?) which causes problems in log lines. The DN values are now
2309 passed through string_printing() before being added to log lines.
2311 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2312 and InterBase are left for another time.)
2314 PH/21 Added message_body_newlines option.
2316 PH/22 Guard against possible overflow in moan_check_errorcopy().
2318 PH/23 POSIX allows open() to be a macro; guard against that.
2320 PH/24 If the recipient of an error message contained an @ in the local part
2321 (suitably quoted, of course), incorrect values were put in $domain and
2322 $local_part during the evaluation of errors_copy.
2328 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2329 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2332 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2333 issue a MAIL command.
2335 PH/02 In an ACL statement such as
2337 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2339 if a client was not listed at all, or was listed with a value other than
2340 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2341 the condition was not true (as it should be), so access was not denied.
2342 The bug was that the ! inversion was incorrectly passed on to the second
2343 item. This has been fixed.
2345 PH/03 Added additional dnslists conditions == and =& which are different from
2346 = and & when the dns lookup returns more than one IP address.
2348 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2349 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2351 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2352 FSYNC, which compiles an option called disable_fsync that allows for
2353 bypassing fsync(). The documentation is heavily laced with warnings.
2355 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2357 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2358 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2359 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2360 including adding "make clean"; (3) Added -fPIC when compiling the test
2361 dynamically loaded module, to get rid of a warning.
2363 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2364 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2365 = 0s. The bug is otherwise harmless.
2367 PH/07 There was a bug in the dovecot authenticator such that the value of
2368 $auth1 could be overwritten, and so not correctly preserved, after a
2369 successful authentication. This usually meant that the value preserved by
2370 the server_setid option was incorrect.
2372 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2374 PH/09 Installed PCRE release 7.0.
2376 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2377 run for batched SMTP input. It is now run at the start of every message
2378 in the batch. While fixing this I discovered that the process information
2379 (output by running exiwhat) was not always getting set for -bs and -bS
2380 input. This is fixed, and it now also says "batched" for BSMTP.
2382 PH/11 Added control=no_pipelining.
2384 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2385 patch, slightly modified), and move the expansion of helo_data till after
2386 the connection is made in the smtp transport (so it can use these
2389 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2391 PH/14 Added log_selector = +pid.
2393 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2395 PH/16 Add ${if forany and ${if forall.
2397 PH/17 Added dsn_from option to vary the From: line in DSNs.
2399 PH/18 Flush SMTP output before performing a callout, unless control =
2400 no_callout_flush is set.
2402 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2403 was true (the default) a successful delivery failed to delete the retry
2404 item, thus causing premature timeout of the address. The bug is now
2407 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2409 PH/21 Long custom messages for fakedefer and fakereject are now split up
2410 into multiline responses in the same way that messages for "deny" and
2411 other ACL rejections are.
2413 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2414 with slight modification.
2416 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2417 draft, changing the syntax and factoring some duplicate code".
2419 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2420 for deliveries of the second and subsequent messages over the same SMTP
2423 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2424 ${reduce, with only minor "tidies".
2426 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2428 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2429 expansion side effects.
2431 PH/27 When a message times out after an over-quota error from an Exim-imposed
2432 quota, the bounce message says "mailbox is full". This message was not
2433 being given when it was a system quota that was exceeded. It now should
2436 MH/03 Made $recipients available in local_scan(). local_scan() already has
2437 better access to the recipient list through recipients_list[], but
2438 $recipients can be useful in postmaster-provided expansion strings.
2440 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2441 in the case of a MAIL command with additional options following the
2442 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2443 were accidentally chopped off.
2445 PH/29 SMTP synchronization checks are implemented when a command is read -
2446 there is a check that no more input is waiting when there shouldn't be
2447 any. However, for some commands, a delay in an ACL can mean that it is
2448 some time before the response is written. In this time, more input might
2449 arrive, invalidly. So now there are extra checks after an ACL has run for
2450 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2451 pipelining has not been advertised.
2453 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2455 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2456 returned in the SMTP response when smtp_return_error_details was false.
2457 This has been fixed.
2459 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2460 instead of the C I/O that was originally supplied, because problems were
2461 reported on Solaris.
2463 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2464 Exim which did not show up earlier: it was assuming that a call to
2465 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2466 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2467 was a macro that became an assignment, so it seemed to work. This has
2468 changed to a proper function call with a void return, hence the compile
2469 error. Exim's code has been fixed.
2471 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2474 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2476 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2478 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2479 because it turns out to be too big or there is a timeout) there is no
2480 "Completed" line in the log. When some messages of this type were
2481 selected by exigrep, they were listed as "not completed". Others were
2482 picked up by some special patterns. I have improved the selection
2483 criteria to be more general.
2485 PH/38 The host_find_failed option in the manualroute router can now be set
2486 to "ignore", to completely ignore a host whose IP address cannot be
2487 found. If all hosts are ignored, the behaviour is controlled by the new
2488 host_all_ignored option.
2490 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2491 homing or because of multiple MX records with /mx) generated more than
2492 one IP address, and the following item turned out to be the local host,
2493 all the secondary addresses of the first item were incorrectly removed
2494 from the list, along with the local host and any following hosts (which
2495 is what is supposed to happen).
2497 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2498 whoever called Exim into the -H file. In the case of the daemon it was
2499 behaving confusingly. When first started, it used values for whoever
2500 started the daemon, but after a SIGHUP it used the Exim user (because it
2501 calls itself on a restart). I have changed the code so that it now always
2504 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2505 message are rejected with the same error (e.g. no authentication or bad
2506 sender address), and a DATA command is nevertheless sent (as can happen
2507 with PIPELINING or a stupid MUA), the error message that was given to the
2508 RCPT commands is included in the rejection of the DATA command. This is
2509 intended to be helpful for MUAs that show only the final error to their
2512 PH/42 Another patch from the Sieve maintainer.
2514 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2515 Eximstats - Fixed some broken HTML links and added missing column headers
2517 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2518 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2520 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2522 PH/43 Yet another patch from the Sieve maintainer.
2524 PH/44 I found a way to check for a TCP/IP connection going away before sending
2525 the response to the final '.' that terminates a message, but only in the
2526 case where the client has not sent further data following the '.'
2527 (unfortunately, this is allowed). However, in many cases there won't be
2528 any further data because there won't be any more messages to send. A call
2529 to select() can be used: if it shows that the input is "ready", there is
2530 either input waiting, or the socket has been closed. An attempt to read
2531 the next input character can distinguish the two cases. Previously, Exim
2532 would have sent an OK response which the client would never have see.
2533 This could lead to message repetition. This fix should cure that, at
2534 least in a lot of common cases.
2536 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2537 advertised in response to EHLO.
2543 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2544 fixed by 4.65/MH/01 (is this a record?) are fixed:
2546 (i) An empty string was always treated as zero by the numeric comparison
2547 operators. This behaviour has been restored.
2549 (ii) It is documented that the numeric comparison operators always treat
2550 their arguments as decimal numbers. This was broken in that numbers
2551 starting with 0 were being interpreted as octal.
2553 While fixing these problems I realized that there was another issue that
2554 hadn't been noticed. Values of message_size_limit (both the global option
2555 and the transport option) were treated as octal if they started with 0.
2556 The documentation was vague. These values are now always treated as
2557 decimal, and I will make that clear in the documentation.
2563 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2564 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2567 MH/01 Don't check that the operands of numeric comparison operators are
2568 integers when their expansion is in "skipping" mode (fixes bug
2569 introduced by 4.64-PH/07).
2571 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2572 child addresses, Exim now panics and dies. Previously, because the count
2573 is held in a short int, deliveries were likely to be lost. As such a
2574 large number of recipients for a single message is ridiculous
2575 (performance will be very, very poor), I have chosen to impose a limit
2576 rather than extend the field.
2582 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2583 leftover -K file (the existence of which was triggered by #402).
2584 While we were at it, introduced process PID as part of the -K
2585 filename. This should rule out race conditions when creating
2588 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2589 processing considerably. Previous code took too long for large mails,
2590 triggering a timeout which in turn triggers #401.
2592 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2593 in the DK code in transports.c. sendfile() is not really portable,
2594 hence the _LINUX specificness.
2596 TF/01 In the add_headers option to the mail command in an Exim filter,
2597 there was a bug that Exim would claim a syntax error in any
2598 header after the first one which had an odd number of characters
2601 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2602 callout verification, Exim cached a "reject" for the entire domain. This
2603 is correct for most verifications, but it is not correct for a recipient
2604 verification with use_sender or use_postmaster set, because in that case
2605 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2606 case of MAIL FROM:<> rejection from other early rejections (e.g.
2607 rejection of HELO). When verifying a recipient using a non-null MAIL
2608 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2609 Whatever the result of the callout, the value of the domain cache is
2610 left unchanged (for any other kind of callout, getting as far as trying
2611 RCPT means that the domain itself is ok).
2613 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2616 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2617 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2618 session, the daemon ignores ECONNECT errors and logs others; it now
2619 ignores EPIPE as well.
2621 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2622 (quoted-printable decoding).
2624 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2625 later the small subsequent patch to fix an introduced bug.
2627 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2629 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2631 PH/08 An error is now given if message_size_limit is specified negative.
2633 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2634 to be given (somewhat) arbitrary names.
2636 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2639 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2640 miscellaneous code fixes
2642 PH/10 Added the log_reject_target ACL modifier to specify where to log
2645 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2646 hostname. This is wrong, because it relates to the incoming message (and
2647 probably the interface on which it is arriving) and not to the outgoing
2648 callout (which could be using a different interface). This has been
2649 changed to use the value of the helo_data option from the smtp transport
2650 instead - this is what is used when a message is actually being sent. If
2651 there is no remote transport (possible with a router that sets up host
2652 addresses), $smtp_active_hostname is used.
2654 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2655 tweaks were necessary in order to get it to work (see also 21 below):
2656 (a) The code assumed that strncpy() returns a negative number on buffer
2657 overflow, which isn't the case. Replaced with Exim's string_format()
2659 (b) There were several signed/unsigned issues. I just did the minimum
2660 hacking in of casts. There is scope for a larger refactoring.
2661 (c) The code used strcasecmp() which is not a standard C function.
2662 Replaced with Exim's strcmpic() function.
2663 (d) The code set only $1; it now sets $auth1 as well.
2664 (e) A simple test gave the error "authentication client didn't specify
2665 service in request". It would seem that Dovecot has changed its
2666 interface. Fortunately there's a specification; I followed it and
2667 changed what the client sends and it appears to be working now.
2669 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2672 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2673 address A is aliased to B and C, where B exists and C does not. Without
2674 -v the output is "A verified" because verification stops after a
2675 successful redirection if more than one address is generated. However,
2676 with -v the child addresses are also verified. Exim was outputting "A
2677 failed to verify" and then showing the successful verification for C,
2678 with its parentage. It now outputs "B failed to verify", showing B's
2679 parentage before showing the successful verification of C.
2681 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2682 look up a TXT record in a specific list after matching in a combined
2685 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2686 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2687 they consult the DNS. I had assumed they would set it the way they
2688 wanted; and indeed my experiments on Linux seem to show that in some
2689 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2690 To be on the safe side, however, I have now made the interface to
2691 host_find_byname() similar to host_find_bydns(), with an argument
2692 containing the DNS resolver options. The host_find_byname() function now
2693 sets these options at its start, just as host_find_bydns() does. The smtp
2694 transport options dns_qualify_single and dns_search_parents are passed to
2695 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2696 of host_find_byname() use the default settings of RES_DEFNAMES
2697 (qualify_single) but not RES_DNSRCH (search_parents).
2699 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2700 spool_read_header() do less string testing, by means of a preliminary
2701 switch on the second character of optional "-foo" lines. (This is
2702 overdue, caused by the large number of possibilities that now exist.
2703 Originally there were few.) While I was there, I also converted the
2704 str(n)cmp tests so they don't re-test the leading "-" and the first
2705 character, in the hope this might squeeze out yet more improvement.
2707 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2708 flag allowing group syntax was set by the header_syntax check but not
2709 turned off, possible causing trouble later; (2) The flag was not being
2710 set at all for the header_verify test, causing "group"-style headers to
2711 be rejected. I have now set it in this case, and also caused header_
2712 verify to ignore an empty address taken from a group. While doing this, I
2713 came across some other cases where the code for allowing group syntax
2714 while scanning a header line wasn't quite right (mostly, not resetting
2715 the flag correctly in the right place). These bugs could have caused
2716 trouble for malformed header lines. I hope it is now all correct.
2718 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2719 with the "reply" argument non-NULL. The code, however (which originally
2720 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2721 but it didn't always do it. This confused somebody who was copying the
2722 code for some other use. I have removed all the tests.
2724 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2725 feature that was used to support insecure browsers during the U.S. crypto
2726 embargo. It requires special client support, and Exim is probably the
2727 only MTA that supported it -- and would never use it because real RSA is
2728 always available. This code has been removed, because it had the bad
2729 effect of slowing Exim down by computing (never used) parameters for the
2730 RSA_EXPORT functionality.
2732 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2733 authenticator to fail if there's a tab character in the incoming data
2734 (there should never be unless someone is messing about, as it's supposed
2735 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2736 if the connection is using TLS or if the remote IP is the same as the
2737 local IP, and the "valid-client-cert option" if a client certificate has
2740 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2741 authenticators. This can be used for authorization after authentication
2742 succeeds. (In the case of plaintext, it servers for both authentication
2745 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2746 if any retry times were supplied.
2748 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2749 connection, where there is no client IP address to check. In this
2750 situation, the verify now always succeeds.
2752 PH/25 Applied John Jetmore's -Mset patch.
2754 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2756 PH/27 In a string expansion for a processed (not raw) header when multiple
2757 headers of the same name were present, leading whitespace was being
2758 removed from all of them, but trailing whitespace was being removed only
2759 from the last one. Now trailing whitespace is removed from each header
2760 before concatenation. Completely empty headers in a concatenation (as
2761 before) are ignored.
2763 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2764 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2766 PH/29 [Removed. This was a change that I later backed out, and forgot to
2767 correct the ChangeLog entry (that I had efficiently created) before
2768 committing the later change.]
2770 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2771 address errors (4xx response to RCPT) over the same connection as other
2772 messages routed to the same hosts. Such deliveries are always "forced",
2773 so retry times are not inspected. This resulted in far too many retries
2774 for the affected addresses. The effect occurred only when there were more
2775 hosts than the hosts_max_try setting in the smtp transport when it had
2776 the 4xx errors. Those hosts that it had tried were not added to the list
2777 of hosts for which the message was waiting, so if all were tried, there
2778 was no problem. Two fixes have been applied:
2780 (i) If there are any address or message errors in an SMTP delivery, none
2781 of the hosts (tried or untried) are now added to the list of hosts
2782 for which the message is waiting, so the message should not be a
2783 candidate for sending over the same connection that was used for a
2784 successful delivery of some other message. This seems entirely
2785 reasonable: after all the message is NOT "waiting for some host".
2786 This is so "obvious" that I'm not sure why it wasn't done
2787 previously. Hope I haven't missed anything, but it can't do any
2788 harm, as the worst effect is to miss an optimization.
2790 (ii) If, despite (i), such a delivery is accidentally attempted, the
2791 routing retry time is respected, so at least it doesn't keep
2792 hammering the server.
2794 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2795 in ${readsocket because some servers need this prod.
2797 PH/32 Added some extra debug output when updating a wait-xxx database.
2799 PH/33 The hint "could be header name not terminated by colon", which has been
2800 given for certain expansion errors for a long time, was not being given
2801 for the ${if def:h_colon_omitted{... case.
2803 PH/34 The spec says: "With one important exception, whenever a domain list is
2804 being scanned, $domain contains the subject domain." There was at least
2805 one case where this was not true.
2807 PH/35 The error "getsockname() failed: connection reset by peer" was being
2808 written to the panic log as well as the main log, but it isn't really
2809 panic-worthy as it just means the connection died rather early on. I have
2810 removed the panic log writing for the ECONNRESET error when getsockname()
2813 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2814 runs only) independently of the message's sender address. This meant
2815 that, if the 4xx error was in fact related to the sender, a different
2816 message to the same recipient with a different sender could confuse
2817 things. In particular, this can happen when sending to a greylisting
2818 server, but other circumstances could also provoke similar problems.
2819 I have changed the default so that the retry time for these errors is now
2820 based a combination of the sender and recipient addresses. This change
2821 can be overridden by setting address_retry_include_sender=false in the
2824 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2825 remote server are returned as part of bounce messages. This was not
2826 happening for LMTP over a pipe (the lmtp transport), but now it is the
2827 same for both kinds of LMTP.
2829 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2830 in header lines that were in fact CNAMEs. This is no longer the case.
2832 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2833 and queue runs started by the daemon processed all messages. This has
2834 been fixed so that -R and -S can now usefully be given with -q<time>.
2836 PH/40 Import PCRE release 6.7 (fixes some bugs).
2838 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2840 PH/42 Give an error if -q is specified more than once.
2842 PH/43 Renamed the variables $interface_address and $interface_port as
2843 $received_ip_address and $received_port, to make it clear that these
2844 values apply to message reception, and not to the outgoing interface when
2845 a message is delivered. (The old names remain recognized, of course.)
2847 PH/44 There was no timeout on the connect() call when using a Unix domain
2848 socket in the ${readsocket expansion. There now is.
2850 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2851 be meaningful with "accept".
2853 SC/01 Eximstats V1.43
2854 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2856 SC/02 Eximstats V1.44
2857 Use a glob alias rather than an array ref in the generated
2858 parser. This improves both readability and performance.
2860 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2861 Collect SpamAssassin and rejection statistics.
2862 Don't display local sender or destination tables unless
2863 there is data to show.
2864 Added average volumes into the top table text output.
2866 SC/04 Eximstats V1.46
2867 Collect data on the number of addresses (recipients)
2868 as well as the number of messages.
2870 SC/05 Eximstats V1.47
2871 Added 'Message too big' to the list of mail rejection
2872 reasons (thanks to Marco Gaiarin).
2874 SC/06 Eximstats V1.48
2875 Mainlog lines which have GMT offsets and are too short to
2876 have a flag are now skipped.
2878 SC/07 Eximstats V1.49 (Alain Williams)
2879 Added the -emptyok flag.
2881 SC/08 Eximstats V1.50
2882 Fixes for obtaining the IP address from reject messages.
2884 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2885 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2886 whitespace changes from 4.64-PH/27
2888 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2891 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2892 are found, allow negative numbers in numeric criteria)
2894 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2896 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2897 to match changes made in 4.64-PH/43
2899 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2901 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2902 "domain" is actually an IP address, and force a failure. This locks out
2903 those revolvers/nameservers that support "A-for-A" lookups, in
2904 contravention of the specifications.
2906 PH/48 When a host name was looked up from an IP address, and the subsequent
2907 forward lookup of the name timed out, the host name was left in
2908 $sender_host_name, contrary to the specification.
2910 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2911 restricted to single-key lookups, Exim was not diagnosing an error if
2912 * or *@ was used with a query-style lookup.
2914 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2916 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2917 long ago, but noone interested enough thought of it. Let's just say that
2918 the "1.1" means that there are some new functions that weren't there at
2919 some point in the past.
2921 PH/51 Error processing for expansion failure of helo_data from an smtp
2922 transport during callout processing was broken.
2924 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2925 tested/used via the -bh/-bhc/-bs options.
2927 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2928 bug, fixed in subsequent PCRE releases).
2930 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2931 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2933 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2939 SC/01 Use a glob alias rather than an array ref in eximstats generated
2940 parser. This improves both readability and performance.
2942 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2943 Don't display local sender or destination tables in eximstats unless
2944 there is data to show.
2945 Added average volumes into the eximstats top table text output.
2947 SC/03 Collect data on the number of addresses (recipients) as well
2948 as the number of messages in eximstats.
2950 TF/01 Correct an error in the documentation for the redirect router. Exim
2951 does (usually) call initgroups() when daemonizing.
2953 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2954 with consistent privilege compared to when running as a daemon.
2956 TF/03 Note in the spec that $authenticated_id is not set for local
2957 submissions from trusted users.
2959 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2960 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2962 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2963 by adding some example configuration directives to the default
2964 configuration file. A little bit of work is required to uncomment the
2965 directives and define how usernames and passwords are checked, but
2966 there is now a framework to start from.
2968 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2969 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2970 without this. I don't know how relevant this is to other LDAP libraries.
2972 PH/02 Add the verb name to the "unknown ACL verb" error.
2974 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2976 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2978 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2979 directory not expanded when it should be if an expanded home directory
2980 was set for the address (which is overridden by the transport).
2982 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2985 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2986 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2987 because it is too late at that time, and has no effect.
2989 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2990 security issue with \' (bugzilla #107). I could not use the
2991 PQescapeStringConn() function, because it needs a PGconn value as one of
2994 PH/08 When testing addresses using -bt, indicate those final addresses that
2995 are duplicates that would not cause an additional delivery. At least one
2996 person was confused, thinking that -bt output corresponded to deliveries.
2997 (Suppressing duplicates isn't a good idea as you lose the information
2998 about possibly different redirections that led to the duplicates.)
3000 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3001 systems where poll() doesn't work, in particular OS X.
3003 PH/10 Added more information to debugging output for retry time not reached.
3005 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3006 operations in malware.c.
3008 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3011 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3012 syslog_duplication set false, log lines that would normally be written
3013 both the the main log and to the reject log were not written to syslog at
3016 PH/14 In the default configuration, change the use of "message" in ACL warn
3017 statements to "add_header".
3019 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3020 not followed by a command (e.g. "seen endif").
3022 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3023 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3026 PH/17 Added extra conditions to the default value of delay_warning_condition
3030 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3031 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3032 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3035 The Auto-Submitted: and various List- headers are standardised, whereas I
3036 don't think Precedence: ever was.
3038 PH/18 Refactored debugging code in route_finduser() to show more information,
3039 in particular, the error code if getpwnam() issues one.
3041 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3042 This is apparently needed in addition to the PH/07 change above to avoid
3043 any possible encoding problems.
3045 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3046 but not after initializing Perl.
3048 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3049 output them only if debugging. By default they are written stderr,
3050 apparently, which is not desirable.
3052 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3055 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3058 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3060 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3061 authenticated or an ident call has been made. Suppress the default
3062 values for $authenticated_id and $authenticated_sender (but permit -oMai
3063 and -oMas) when testing with -bh.
3065 PH/24 Re-jigged the order of the tests in the default configuration so that the
3066 tests for valid domains and recipients precede the DNS black list and CSA
3067 tests, on the grounds that those ones are more expensive.
3069 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3070 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3071 This bug exists in every version of Exim that I still have, right back to
3074 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3075 However, an attempt to turn on case-sensitivity in a regex key by
3076 including (?-i) didn't work because the subject string was already
3077 lowercased, and the effects were non-intuitive. It turns out that a
3078 one-line patch can be used to allow (?-i) to work as expected.
3084 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3085 other effects) broke the use of negated acl sub-conditions.
3087 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3090 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3091 "Deny" causes Exim to reject the incoming connection with a 554 error.
3092 Unfortunately, if there is a major crisis, such as a disk failure,
3093 tcp-wrappers gives "deny", whereas what one would like would be some
3094 kind of temporary error. A kludge has been added to help with this.
3095 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3096 554 error is used if errno is still zero or contains ENOENT (which occurs
3097 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3100 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3102 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3103 errors. Otherwise a message that provokes a temporary error (when other
3104 messages do not) can cause a whole host to time out.
3106 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3107 addresses were routed directly to files or pipes from a redirect router.
3108 File deliveries just didn't batch; pipe deliveries might have suffered
3111 PH/06 A failure to get a lock for a hints database would erroneously always say
3112 "Failed to get write lock", even when it was really a read lock.
3114 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3115 of 0600. This has been changed to use the value of the lockfile_mode
3116 option (which defaults to 0600).
3118 PH/08 Applied small patch from the Sieve maintainer.
3120 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3121 folder from quota calculations, a direct delivery into this folder messed
3122 up the contents of the maildirsize file. This was because the regex was
3123 used only to exclude .Trash (or whatever) when the size of the mailbox
3124 was calculated. There was no check that a delivery was happening into an
3125 excluded directory. This bug has been fixed by ignoring all quota
3126 processing for deliveries into excluded directories.
3128 PH/10 Added the maildirfolder_create_regex option to appendfile.
3134 PH/01 The code for finding all the local interface addresses on a FreeBSD
3135 system running IPv6 was broken. This may well have applied to all BSD
3136 systems, as well as to others that have similar system calls. The broken
3137 code found IPv4 interfaces correctly, but gave incorrect values for the
3138 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3139 that it would not match correctly against @[] and not recognize the IPv6
3142 PH/02 The ipliteral router was not recognizing addresses of the form user@
3143 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3145 PH/03 Added disable_ipv6.
3147 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3148 decoded form, because it is most often used to construct To: headers
3149 lines in autoreplies, and the decoded form may well be syntactically
3150 invalid. However, $reply_address has leading white space removed, and all
3151 newlines turned into spaces so that the autoreply transport does not
3154 PH/05 If group was specified without a user on a router, and no group or user
3155 was specified on a transport, the group from the router was ignored.
3157 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3158 for visible compile-time settings that can be used to change these
3159 numbers, for those that want even more. Backwards compatibility with old
3160 spool files has been maintained. However, going back to a previous Exim
3161 release will lost any variables that are in spool files.
3163 PH/07 Two small changes when running in the test harness: increase delay when
3164 passing a TCP/IP connection to a new process, in case the original
3165 process has to generate a bounce, and remove special handling of
3166 127.0.0.2 (sic), which is no longer necessary.
3168 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3169 be the same on different OS.
3171 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3174 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3175 whether --show-vars was specified or not
3177 JJ/02 exipick: Added support for new ACL variable spool format introduced
3180 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3181 syntactically invalid From: or Reply-to: line, and a filter used this to
3182 generate an autoreply, and therefore failed to obtain an address for the
3183 autoreply, Exim could try to deliver to a non-existent relative file
3184 name, causing unrelated and misleading errors. What now happens is that
3185 it logs this as a hard delivery error, but does not attempt to create a
3188 PH/11 The exinext utility has a -C option for testing purposes, but although
3189 the given file was scanned by exinext itself; it wasn't being passed on
3190 when Exim was called.
3192 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3193 an end-of-file indication when reading a command response.
3195 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3196 compiled. In many other places in Exim, IPv6 addresses are always
3197 recognized, so I have changed this. It also means that IPv4 domain
3198 literals of the form [IPV4:n.n.n.n] are now always recognized.
3200 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3201 used if the router is not running as root, for example, when verifying at
3202 ACL time, or when using -bh. The debugging output from this situation was
3203 non-existent - all you got was a failure to exec. I have made two
3206 (a) Failures to set uid/gid, the current directory, or a process leader
3207 in a subprocess such as that created by queryprogram now generate
3208 suitable debugging output when -d is set.
3210 (b) The queryprogram router detects when it is not running as root,
3211 outputs suitable debugging information if -d is set, and then runs
3212 the subprocess without attempting to change uid/gid.
3214 PH/15 Minor change to Makefile for building test_host (undocumented testing
3217 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3218 additional section of a DNS packet that returns MX or SRV records.
3219 Instead, it always explicitly searches for A/AAAA records. This avoids
3220 major problems that occur when a DNS server includes only records of one
3221 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3222 fixed another bug: if SRV records were looked up and the corresponding
3223 address records were *not* found in the additional section, the port
3224 values from the SRV records were lost.
3226 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3227 using the correct key (the original address) when searching the retry
3228 rules in order to find which one to use for generating the retry hint.
3230 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3231 adding the default one. Similarly, if it contains a Reply-To: header, the
3232 errors_reply_to option, if set, is not used.
3234 PH/19 When calculating a retry time, Exim used to measure the "time since
3235 failure" by looking at the "first failed" field in the retry record. Now
3236 it does not use this if it is later than than the arrival time of the
3237 message. Instead it uses the arrival time. This makes for better
3238 behaviour in cases where some deliveries succeed, thus re-setting the
3239 "first failed" field. An example is a quota failure for a huge message
3240 when small messages continue to be delivered. Without this change, the
3241 "time since failure" will always be short, possible causing more frequent
3242 delivery attempts for the huge message than are intended.
3243 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3245 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3246 $1, $2, $3) because the numerical variables can be reset during some
3247 expansion items (e.g. "match"), thereby losing the authentication data.
3249 PH/21 Make -bV show the size of off_t variables so that the test suite can
3250 decide whether to run tests for quotas > 2G.
3252 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3253 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3254 filecount value is greater than 2G or if a quota value is greater than 2G
3255 on a system where the size of off_t is not greater than 4, a panic error
3258 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3259 never match. The debug and -bh output now contains an explicit error
3260 message indicating a malformed IPv4 address or mask.
3262 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3263 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3264 PH/23 above applies.
3266 PH/25 Do not write to syslog when running in the test harness. The only
3267 occasion when this arises is a failure to open the main or panic logs
3268 (for which there is an explicit test).
3270 PH/26 Added the /no_tell option to "control=freeze".
3272 PH/27 If a host name lookup failed very early in a connection, for example, if
3273 the IP address matched host_lookup and the reverse lookup yielded a name
3274 that did not have a forward lookup, an error message of the form "no IP
3275 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3276 could be logged. Now it outputs the IP address instead of "NULL".
3278 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3279 allows the sender and the authenticated sender to be set when
3280 submitting a message from within Exim. Since child_open_exim() is
3281 documented for local_scan(), the new function should be too.
3283 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3284 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3285 results in an empty string is now treated as unset.
3287 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3289 PH/31 Added sender_verify_fail logging option.
3291 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3292 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3293 tidied the source and removed it altogether.
3295 PH/33 When a queue run was abandoned because the load average was too high, a
3296 log line was always written; now it is written only if the queue_run log
3297 selector is set. In addition, the log line for abandonment now contains
3298 information about the queue run such as the pid. This is always present
3299 in "start" and "stop" lines but was omitted from the "abandon" line.
3301 PH/34 Omit spaces between a header name and the colon in the error message that
3302 is given when verify = headers_syntax fails (if there are lots of them,
3303 the message gets confusing).
3305 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3306 names, as there are now some PTR records that contain slashes. This check
3307 is only to protect against broken name servers that fall over on strange
3308 characters, so the fact that it applies to all lookups doesn't matter.
3310 PH/36 Now that the new test suite is complete, we can remove some of the
3311 special code in Exim that was needed for the old test suite. For example,
3312 sorting DNS records because real resolvers return them in an arbitrary
3313 order. The new test suite's fake resolver always returns records in the
3316 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3317 bounces) except when queue_only is set, to avoid logging races between
3318 the different processes.
3320 PH/38 Panic-die if .include specifies a non-absolute path.
3322 PH/39 A tweak to the "H" retry rule from its user.
3324 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3325 a label. They prevented compilation on older perls.
3327 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3328 a warning to be raised on newish perls.
3330 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3331 on queue. Changes to match documented behaviour of showing count of
3332 messages matching specified criteria.
3334 PH/40 Changed the default ident timeout from 30s to 5s.
3336 PH/41 Added support for the use of login_cap features, on those BSD systems
3337 that have them, for controlling the resources used by pipe deliveries.
3339 PH/42 The content-scanning code uses fopen() to create files in which to put
3340 message data. Previously it was not paying any attention to the mode of
3341 the files. Exim runs with umask(0) because the rest of the code creates
3342 files with open(), and sets the required mode explicitly. Thus, these
3343 files were ending up world-writeable. This was not a big issue, because,
3344 being within the spool directory, they were not world-accessible. I have
3345 created a function called modefopen, which takes an additional mode
3346 argument. It sets umask(777), creates the file, chmods it to the required
3347 mode, then resets the umask. All the relevant calls to fopen() in the
3348 content scanning code have been changed to use this function.
3350 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3351 to 24 hours. This avoids potential overflow problems when processing G
3352 and H retry rules. I suspect nobody ever tinkers with this value.
3354 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3356 PH/45 When the plaintext authenticator is running as a client, the server's
3357 challenges are checked to ensure they are valid base64 strings. By
3358 default, the authentication attempt is cancelled if an invalid string is
3359 received. Setting client_ignore_invalid_base64 true ignores these errors.
3360 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3361 they are received. Thus, the responses can be made to depend on the
3362 challenges. If an invalid string is ignored, an empty string is placed in
3365 PH/46 Messages that are created by the autoreply transport now contains a
3366 References: header, in accordance with RFCs 2822 and 3834.
3368 PH/47 Added authenticated_sender_force to the smtp transport.
3370 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3372 PH/49 Installed latest patch from the Sieve maintainer.
3374 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3375 was also set, the appendfile transport was unnecessarily scanning a
3376 directory of message files (e.g. for maildir delivery) to find the count
3377 of files (along with the size), even though it did not need this
3378 information. It now does the scan only if it needs to find either the
3379 size of the count of files.
3381 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3383 PH/52 Two bugs concerned with error handling when the smtp transport is
3386 (i) Exim was not creating retry information for temporary errors given
3387 for individual recipients after the DATA command when the smtp transport
3388 was used in LMTP mode. This meant that they could be retried too
3389 frequently, and not timed out correctly.
3391 (ii) Exim was setting the flag that allows error details to be returned
3392 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3393 recipients that were returned after the DATA command.
3395 PH/53 This is related to PH/52, but is more general: for any failing address,
3396 when detailed error information was permitted to be returned to the
3397 sender, but the error was temporary, then after the final timeout, only
3398 "retry timeout exceeded" was returned. Now it returns the full error as
3399 well as "retry timeout exceeded".
3401 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3402 do this, and (what is worse) MTAs that accept it.
3404 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3405 will now be deprecated.
3407 PH/56 New os.c-cygwin from the Cygwin maintainer.
3409 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3410 formats (previously only available in exim formats via -bpr, -bpru,
3411 and -bpra. Now also available in native and exiqgrep formats)
3413 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3414 with very large, slow to parse queues
3416 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3418 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3420 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3421 responses to authentication challenges, though it was showing the
3422 challenges; (ii) I've removed the CR characters from the debug output for
3425 PH/58 Allow for the insertion of a newline as well as a space when a string
3426 is turned into more than one encoded-word during RFC 2047 encoding. The
3427 Sieve code now uses this.
3429 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3430 data_4xx, lost_connection, tls_required.
3432 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3433 message was being sent as an SMTP response.
3435 PH/61 Add -l and -k options to exicyclog.
3437 PH/62 When verifying, if an address was redirected to one new address, so that
3438 verification continued, and the new address failed or deferred after
3439 having set something in $address_data, the value of $address_data was not
3440 passed back to the ACL. This was different to the case when no
3441 redirection occurred. The value is now passed back in both cases.
3443 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3444 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3445 use login_cap.h, so on its own it isn't the distinguishing feature. The
3446 new name refers directly to the setclassresources() function.
3448 PH/65 Added configuration files for NetBSD3.
3450 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3452 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3453 is preferred over IPv4.
3455 PH/68 The bounce_return_message and bounce_return_body options were not being
3456 honoured for bounces generated during the reception of non-SMTP messages.
3457 In particular, this applied to messages rejected by the ACL. This bug has
3458 been fixed. However, if bounce_return_message is true and bounce_return_
3459 body is false, the headers that are returned for a non-SMTP message
3460 include only those that have been read before the error was detected.
3461 (In the case of an ACL rejection, they have all been read.)
3463 PH/69 The HTML version of the specification is now built in a directory called
3464 spec_html instead of spec.html, because the latter looks like a path with
3465 a MIME-type, and this confuses some software.
3467 PH/70 Catch two compiler warnings in sieve.c.
3469 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3470 function verify_get_ident() calls ip_connect() to connect a socket, but
3471 if the "connect()" function timed out, ip_connect() used to close the
3472 socket. However, verify_get_ident() also closes the socket later, and in
3473 between Exim writes to the log, which may get opened at this point. When
3474 the socket was closed in ip_connect(), the log could get the same file
3475 descriptor number as the socket. This naturally causes chaos. The fix is
3476 not to close the socket in ip_connect(); the socket should be closed by
3477 the function that creates it. There was only one place in the code where
3478 this was missing, in the iplookup router, which I don't think anybody now
3479 uses, but I've fixed it anyway.
3481 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3482 well as to direct DNS lookups. Otherwise the handling of names in host
3483 lists is inconsistent and therefore confusing.
3489 PH/01 Two changes to the default runtime configuration:
3491 (1) Move the checks for relay_from_hosts and authenticated clients from
3492 after to before the (commented out) DNS black list checks.
3494 (2) Add control=submission to the relay_from_hosts and authenticated
3495 clients checks, on the grounds that messages accepted by these
3496 statements are most likely to be submissions.
3498 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3500 (1) Generate an error if the third argument for the ${prvs expansion is
3503 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3506 (3) Reset the variables that are obtained from the first argument of
3507 ${prvscheck and used in the second argument before leaving the code,
3508 because their memory is reclaimed, so using them afterwards may do
3511 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3512 one (it's much easier than Tom thought :-).
3514 (5) Because of (4), we can now allow for the use of $prvscheck_result
3515 inside the third argument.
3517 PH/03 For some reason, the default setting of PATH when running a command from
3518 a pipe transport was just "/usr/bin". I have changed it to
3521 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3522 anything to be listed in the output from -bV.
3524 PH/05 When a filter generated an autoreply, the entire To: header line was
3525 quoted in the delivery log line, like this:
3527 => >A.N.Other <ano@some.domain> <original@ddress> ...
3529 This has been changed so that it extracts the operative address. There
3530 may be more than one such address. If so, they are comma-separated, like
3533 => >ano@some.domain,ona@other.domain <original@ddress> ...
3535 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3536 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3537 not being looked up in the rDNS to get a host name, Exim was showing the
3538 IP address twice in Received: lines, even though the IP addresses were
3539 identical. For example:
3541 Received: from [1.2.3.4] (helo=[1.2.3.4])
3543 However, if the real host name was known, it was omitting the HELO data
3544 if it matched the actual IP address. This has been tidied up so that it
3545 doesn't show the same IP address twice.
3547 PH/07 When both +timestamp and +memory debugging was on, the value given by
3548 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3549 called by the debug printing, thereby overwriting the timestamp buffer.
3550 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3552 PH/08 When the original message was included in an autoreply transport, it
3553 always said "this is a copy of the message, including all the headers",
3554 even if body_only or headers_only was set. It now gives an appropriate
3557 PH/09 Applied a patch from the Sieve maintainer which:
3559 o fixes some comments
3560 o adds the (disabled) notify extension core
3561 o adds some debug output for the result of if/elsif tests
3562 o points to the current vacation draft in the documentation
3563 and documents the missing references header update
3567 o fixes a bug in processing the envelope test (when testing
3568 multiple envelope elements, the last element determined the
3571 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3572 Electronic Mail") by including:
3574 Auto-submitted: auto-generated
3576 in the messages that it generates (bounce messages and others, such as
3577 warnings). In the case of bounce messages for non-SMTP messages, there was
3578 also a typo: it was using "Auto_submitted" (underscore instead of
3579 hyphen). Since every message generated by Exim is necessarily in response
3580 to another message, thes have all been changed to:
3582 Auto-Submitted: auto-replied
3584 in accordance with these statements in the RFC:
3586 The auto-replied keyword:
3588 - SHOULD be used on messages sent in direct response to another
3589 message by an automatic process,
3591 - MUST NOT be used on manually-generated messages,
3593 - MAY be used on Delivery Status Notifications (DSNs) and Message
3594 Disposition Notifications (MDNs),
3596 - MUST NOT be used on messages generated by automatic or periodic
3597 processes, except for messages which are automatic responses to
3600 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3601 to the default Received: header definition.
3603 PH/12 Added log selector acl_warn_skipped (default on).
3605 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3606 variables because (a) they are in the wrong storage pool and (b) even if
3607 they were copied, it wouldn't work properly because of the caching.
3609 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3610 checking when decoding. Apparently there are clients that generate
3611 overlong encoded strings. Why am I not surprised?
3613 PH/15 If the first argument of "${if match_address" was not empty, but did not
3614 contain an "@" character, Exim crashed. Now it writes a panic log message
3615 and treats the condition as false.
3617 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3619 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3620 extension "envelope-auth". The code is finished and in agreement with
3621 other implementations, but there is no documentation so far and in fact,
3622 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3623 not changing the active code.
3625 Print executed "if" and "elsif" statements when debugging is used. This
3626 helps a great deal to understand what a filter does.
3628 Document more things not specified clearly in RFC3028. I had all this
3629 sorted out, when out of a sudden new issues came to my mind. Oops."
3631 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3634 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3635 canonical form (as documented). However, after a host name lookup from
3636 the IP address, check_host() was doing a simple string comparison with
3637 addresses acquired from the DNS when checking that the found name did
3638 have the original IP as one of its addresses. Since any found IPv6
3639 addresses are likely to be in abbreviated form, the comparison could
3640 fail. Luckily, there already exists a function for doing the comparison
3641 by converting both addresses to binary, so now that is used instead of
3642 the text comparison.
3644 PH/20 There was another similar case to PH/19, when a complete host name was
3645 given in a host list; looking up its IP address could give an abbreviated
3646 form, whereas the current host's name might or might not be abbreviated.
3647 The same fix has been applied.
3653 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3654 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3657 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3658 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3660 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3662 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3663 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3664 Also, for "Auto-Submitted", treat anything other than "no" as
3665 non-personal, in accordance with RFC 3834. (Previously it treated
3666 anything starting "auto-" as non-personal.)
3668 TF/01 The control=submission/name=... option had a problem with syntax
3669 errors if the name included a slash character. The /name= option
3670 now slurps the rest of the string, so it can include any characters
3671 but it must come last in the list of options (after /sender_retain
3674 PH/05 Some modifications to the interface to the fake nameserver for the new
3682 TK/01 Added the "success_on_redirect" address verification option. See
3683 NewStuff for rationale and an example.
3685 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3687 PH/02 Patch to exigrep to allow it to work on syslog lines.
3689 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3690 fread() to skip over the body file's header line, because in Cygwin the
3691 header line is locked and is inaccessible.
3693 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3694 co-exist for some time) to make it clear that it is the Exim ID that is
3695 referenced, not the Message-ID: header line.
3697 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3698 string_format() function, because snprintf() does not exist on all
3701 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3702 ${stat: expansion item.
3704 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3705 protocol synchronization error", to keep the pedants happy.
3707 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3708 well as for IRIX systems, when gcc is being used. See the host.c source
3711 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3713 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3716 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3717 transport and to the smtp transport in LMTP mode.
3719 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3721 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3722 run a filter in a subprocess. This could lead to confusion in subsequent
3723 lookups in the parent process. There should also be a search_tidyup() at
3724 the end of the subprocess.
3726 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3727 only if the host matched helo_try_verify_hosts, which caused the
3728 verification to occur when the EHLO/HELO command was issued. The ACL just
3729 tested the remembered result. Now, if a previous verification attempt has
3730 not happened, "verify = helo" does it there and then.
3732 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3734 TK/03 Fix log output including CR from clamd.
3736 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3737 exist provoked a memory error which could cause a segfault.
3739 PH/15 Installed PCRE 6.2
3741 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3743 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3744 of the problem. Specifically, suggested +O2 rather than +O1 for the
3747 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3749 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3750 record, the port was not correctly logged when the outgoing_port log
3751 selector was set (it logged the transort's default port).
3753 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3754 fallback_hosts, and "hosts" in the smtp transport.
3756 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3757 host errors such as "Connection refused".
3759 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3760 authentication with radiusclient 0.4.9:
3762 - Error returned from rc_read_config was caught wrongly
3763 - Username/password not passed on to radius server due to wrong length.
3765 The presumption is that some radiusclient API changes for 4.51/PH/17
3766 were not taken care of correctly. The code is still untested by me (my
3767 Linux distribution still has 0.3.2 of radiusclient), but it was
3768 contributed by a Radius user.
3770 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3771 expanding the "port" option of the smtp transport.
3773 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3774 while reading a MIME header. Thanks to Tom Hughes for a patch.
3776 PH/24 Include config.h inside local_scan.h so that configuration settings are
3779 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3780 that in an ACL for RCPT (for example), you can examine exactly what was
3783 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3784 commands, but it was not correctly comparing the address with the actual
3785 client host address. Thus, it would show the EHLO address in Received:
3786 header lines when this was not necessary.
3788 PH/27 Added the % operator to ${eval:}.
3790 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3791 it should be ignoring failures (because with -C, for example, it has lost
3792 privilege). It wasn't ignoring creation failures other than "already
3795 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3798 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3799 because an input line was too long, either on its own, or by virtue of
3800 too many continuations, the temporary file was not being removed, and the
3801 return code was incorrect.
3803 PH/31 Missing "BOOL" in function definition in filtertest.c.
3805 PH/32 Applied Sieve patches from the maintainer.
3807 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3809 PH/33 Added "verify = not_blind".
3811 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3812 Local/Makefile (with some defaults set). These are used in built scripts
3813 such as exicyclog, but they have never been used in the exim_install
3814 script (though there are many overriding facilities there). I have
3815 arranged that the exim_install script now takes note of these two
3818 PH/35 Installed configuration files for Dragonfly.
3820 PH/36 When a locally submitted message by a trusted user did not contain a
3821 From: header, and the sender address was obtained from -f or from an SMTP
3822 MAIL command, and the trusted user did not use -F to supply a sender
3823 name, $originator_name was incorrectly used when constructing a From:
3824 header. Furthermore, $originator_name was used for submission mode
3825 messages from external hosts without From: headers in a similar way,
3826 which is clearly wrong.
3828 PH/37 Added control=suppress_local_fixups.
3830 PH/38 When log_selector = +received_sender was set, and the addition of the
3831 sender made the log line's construction buffer exactly full, or one byte
3832 less than full, an overflow happened when the terminating "\n" was
3835 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3836 when the result of a list match is failure because a DNS lookup failed.
3838 PH/40 RM_COMMAND is now used in the building process.
3840 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3841 the "build-* directories that it finds.
3843 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3844 address in a domain literal was a prefix of an interface address.
3846 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3847 when verifying a sender address, unless rewrite_headers is false.
3849 PH/44 Wrote a long comment about why errors_to addresses are verified as
3850 recipients, not senders.
3852 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3853 the ratelimit ACL was added.
3855 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3857 PH/46 Added extra information about PostgreSQL errors to the error string.
3859 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3860 avoiding the need to install special zones in a real server. This is
3861 backwards compatible; if it can't find the fake resolver, it drops back.
3862 Thus, both old and new test suites can be run.
3864 TF/02 Added util/ratelimit.pl
3866 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3867 clock is set back in time.
3869 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3870 Candler <B.Candler@pobox.com>.
3872 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3873 OK for addresses that are the result of redirections.
3875 PH/48 A number of further additions for the benefit of the new test suite,
3876 including a fake gethostbyname() that interfaces to the fake DNS resolver
3879 TF/06 The fix for widen_domains has also been applied to qualify_single and
3880 search_parents which are the other dnslookup options that can cause
3883 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3886 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3888 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3889 service" if it encountered an MX record with an empty target hostname.
3890 The message is now "an MX or SRV record indicated no SMTP service".
3892 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3893 if the DNS of the sending site is misconfigured. This is quite a
3894 common situation. This change restores the behaviour of treating a
3895 helo verification defer as a failure.
3897 PH/51 If self=fail was set on a router, the bounce message did not include the
3898 actual error message.
3904 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3906 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3907 command itself ended in error, the underlying message about the transport
3908 filter timeout was being overwritten with the pipe command error. Now the
3909 underlying error message should be appended to the second error message.
3911 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3913 PH/02 Reduce the amount of output that "make" produces by default. Full output
3914 can still be requested.
3916 PH/03 The warning log line about a condition test deferring for a "warn" verb
3917 was being output only once per connection, rather than after each
3918 occurrence (because it was using the same function as for successful
3919 "warn" verbs). This seems wrong, so I have changed it.
3921 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3922 it should not have, which might have caused a crash in the right
3923 circumstances, but probably never did.
3925 PH/04 Installed a modified version of Tony Finch's patch to make submission
3926 mode fix the return path as well as the Sender: header line, and to
3927 add a /name= option so that you can make the user's friendly name appear
3930 TF/03 Added the control = fakedefer ACL modifier.
3932 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3933 Mark Lowes for thorough testing.
3935 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3937 TK/03 Merged latest SRS patch from Miles Wilton.
3939 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3940 with the definition in sysexits.h (which is #included earlier).
3941 Fortunately, Exim does not actually use EX_OK. The code used to try to
3942 preserve the sysexits.h value, by assuming that macro definitions were
3943 scanned for macro replacements. I have been disabused of this notion,
3944 so now the code just undefines EX_OK before #including unistd.h.
3946 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3947 in the smtp transport. When a block could not be written in a single
3948 write() function, the timeout was being re-applied to each part-write.
3949 This seems wrong - if the receiver was accepting one byte at a time it
3950 would take for ever. The timeout is now adjusted when this happens. It
3951 doesn't have to be particularly precise.
3953 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3954 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3956 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3957 without a domain if the check to <postmaster@domain> fails.
3959 SC/01 Eximstats: added -xls and the ability to specify output files
3960 (patch written by Frank Heydlauf).
3962 SC/02 Eximstats: use FileHandles for outputting results.
3964 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3966 SC/04 Eximstats: fixed display of large numbers with -nvr option
3968 SC/05 Eximstats: fixed merging of reports with empty tables.
3970 SC/06 Eximstats: added the -include_original_destination flag
3972 SC/07 Eximstats: removed tabs and trailing whitespace.
3974 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3976 TK/06 MBOX spool code: Add real "From " MBOX separator line
3977 so the .eml file is really in mbox format (even though
3978 most programs do not really care). Patch from Alex Miller.
3980 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3981 The latter is generated from $received_to and is only set if the
3982 message has one envelope recipient. SA can use these headers,
3983 obviously out-of-the-box. Patch from Alex Miller.
3985 PH/08 The ${def test on a variable was returning false if the variable's
3986 value was "0", contrary to what the specification has always said!
3987 The result should be true unless the variable is empty.
3989 PH/09 The syntax error of a character other than { following "${if
3990 def:variable_name" (after optional whitespace) was not being diagnosed.
3991 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3992 accidental colon was present, for example, could give incorrect results.
3994 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3995 result is used (not including appendfile, where other changes are about
3998 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3999 This involved changing a lot of size variables from int to off_t. It
4000 should work with maildirs and everything.
4002 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4003 spamd dying while we are connected to it.
4005 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4008 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4009 function for BDB 4.3.
4011 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4013 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4014 the sources for PCRE have been re-arranged and more files are now
4017 PH/15 The code I had for printing potentially long long variables in PH/11
4018 above was not the best (it lost precision). The length of off_t variables
4019 is now inspected at build time, and an appropriate printing format (%ld
4020 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4021 to be "long long int" or "long int". This is needed for the internal
4022 formatting function string_vformat().
4024 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4025 the configuration file to be ":syslog", then the script "guesses" where
4026 the logs files are, rather than using the compiled in default. In our
4027 case the guess is not the same as the compiled default, so the script
4028 suddenly stopped working when I started to use syslog. The patch checks
4029 to see if log_file_path is "". If so, it attempts to read it from exim
4030 with no configuration file to get the compiled in version, before it
4031 falls back to the previous guessing code."
4033 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4034 implementing BATV in an Exim configuration. See NewStuff for the gory
4037 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4038 Makefile that are specific to HP-UX.
4040 PH/18 If the "use_postmaster" option was set for a recipient callout together
4041 with the "random" option, the postmaster address was used as the MAIL
4042 FROM address for the random test, but not for the subsequent recipient
4043 test. It is now used for both.
4045 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4046 patch removes a few documentation additions to RFC 3028, because the
4047 latest draft now contains them. It adds the new en;ascii-case comparator
4048 and a new error check for 8bit text in MIME parts. Comparator and
4049 require names are now matched exactly. I enabled the subaddress
4050 extension, but it is not well tested yet (read: it works for me)."
4052 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4053 rework some of the code of TK/09 above to avoid the hardwired use of
4054 "%lld" and "long long". Replaced the call to snprintf() with a call to
4057 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4058 records point to non-existent hosts", "retry timeout exceeded", and
4059 "retry time not reached for any host after a long failure period".
4061 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4062 experimental DomainKeys support:
4064 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4065 (2) On an error such as an illegally used "control", the wrong name for
4066 the control was given.
4068 These problems did NOT occur unless DomainKeys support was compiled.
4070 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4072 PH/24 Added ${if match_ip condition.
4074 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4075 databases so that it will be absolutely obvious if a crash occurs in the
4076 DB library. This is a regular occurrence (often caused by mis-matched
4079 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4080 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4081 picked up on a user's system that detects such things. There doesn't seem
4082 to be a gcc warning option for this - only an attribute that has to be
4083 put on the function's prototype. It seems that in Fedora Core 4 they have
4084 set this on a number of new functions. No doubt there will be more in due
4087 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4088 specify a transport. However, if an address that was verified by such a
4089 router was the subject of a callout, Exim crashed because it tried to
4090 read the rcpt_include_affixes from the non-existent transport. Now it
4091 just assumes that the setting of that option is false. This bug was
4092 introduced by 4.51/PH/31.
4094 PH/28 Changed -d+all to exclude +memory, because that information is very
4095 rarely of interest, but it makes the output a lot bigger. People tend to
4096 do -d+all out of habit.
4098 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4099 code in os-type was giving problems when libc.so lives in lib64, like on
4102 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4103 aren't the modern standard, and it seems that some systems' include files
4104 don't always have them. Exim was already checking for some of the newer
4105 ones like T_AAAA, and defining it itself. I've added checks for all the
4106 record types that Exim uses.
4108 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4109 not automatically generating a new one, as it is supposed to. This
4110 prevented TLS from working. If the file did exist, but contained invalid
4111 data, a new version was generated, as expected. It was only the case of a
4112 non-existent file that was broken.
4114 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4115 with a change in libdomainkeys > 0.64.
4117 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4118 from DNS. If the selector record carries the flag, it now has
4119 precedence over the domain-wide flag.
4121 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4123 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4124 the use of a port name that isn't defined in /etc/services) occurred, the
4125 message was deferred as in a normal delivery, and thus remained on the
4126 spool, instead of being failed because of the mua_wrapper setting. This
4127 is now fixed, and I tidied up some of the mua_wrapper messages at the
4130 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4131 the messages in a hash of arrays rather than using individual hashes.
4132 This is a bit cleaner and results in dramatic memory savings, albeit
4133 at a slight CPU cost.
4135 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4136 as requested by Marc Sherman.
4138 SC/10 Eximstats: added histograms for user specified patterns as requested
4141 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4143 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4144 fopen() in the content-scanning modules that did not already have it.
4150 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4151 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4153 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4155 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4157 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4158 to test Sieve filters that use "vacation".
4160 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4161 that changes the way the GnuTLS parameters are stored in the cache file.
4162 The new format can be generated externally. For backward compatibility,
4163 if the data in the cache doesn't make sense, Exim assumes it has read an
4164 old-format file, and it generates new data and writes a new file. This
4165 means that you can't go back to an older release without removing the
4168 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4169 work if there are any delivery delays because "one_time" forces the
4170 parent to be marked "delivered", so its unseen clone is never tried
4171 again. For this reason, Exim now forbids the simultaneous setting of
4174 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4175 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4176 entry if you want to know the details. The fix, however, neglected to
4177 consider the case when local delivery batching is involved. The test for
4178 "previously delivered" was not happening when checking to see if an
4179 address could be batched with a previous (undelivered) one; under
4180 certain circumstances this could lead to multiple deliveries to the same
4183 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4184 in its include files, and this causes problems building Exim.
4186 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4187 header_syntax) but Exim was just ignoring anything given after a slash.
4188 In particular, this caused confusion with an attempt to use "verify =
4189 reverse_host_lookup/defer_ok". An error is now given when options are
4190 supplied for verify items that do not have them. (Maybe reverse_host_
4191 lookup should have a defer_ok option, but that's a different point.)
4193 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4194 defined by RFC 821) to 2048, because there were problems with some AUTH
4195 commands, and RFC 1869 says the size should be increased for extended
4196 SMTP commands that take arguments.
4198 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4201 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4202 "unknown" error; now it says that the functionality isn't in the binary.
4204 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4205 an address' error message when a string expansion fails (syntax or
4206 whatever). Otherwise the password may appear in the log. Following change
4207 PH/42 below, there is no longer a chance of it appearing in a bounce
4210 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4212 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4213 of its addresses was ever tried. (Bugzilla bug #2.)
4215 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4216 the result incorrectly in the debug output. (It correctly added a newline
4217 to what was transported.)
4219 TF/01 Added $received_time.
4221 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4222 commented out examples of how to interface to a virus scanner and to
4223 SpamAssassin. Also added commented examples of av_scanner and
4224 spamd_address settings.
4226 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4227 and controls are allowed in which ACLs. There were a couple of minor
4228 errors. Some of the entries in the conditions table (which is a table of
4229 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4230 negation of where the condition IS allowed.
4232 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4234 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4235 header file does not have a version number, so I've had to invent a new
4236 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4237 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4238 radiusclient), but it was contributed by a Radius user.
4240 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4241 files or directories, for OpenSSL.
4243 PH/19 When an Exim process that is running as root has to create an Exim log
4244 file, it does so in a subprocess that runs as exim:exim so as to get the
4245 ownership right at creation (otherwise, other Exim processes might see
4246 the file with the wrong ownership). There was no test for failure of this
4247 fork() call, which would lead to the process getting stuck as it waited
4248 for a non-existent subprocess. Forks do occasionally fail when resources
4249 run out. I reviewed all the other calls to fork(); they all seem to check
4252 PH/20 When checking for unexpected SMTP input at connect time (before writing
4253 the banner), Exim was not dealing correctly with a non-positive return
4254 from the read() function. If the client had disconnected by this time,
4255 the result was a log entry for a synchronization error with an empty
4256 string after "input=" when read() returned zero. If read() returned -1
4257 (an event I could not check), uninitialized data bytes were printed.
4258 There were reports of junk text (parts of files, etc) appearing after
4261 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4263 PH/22 Added support for macro redefinition, and (re)definition in between
4264 driver and ACL definitions.
4266 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4267 forgetting to use the resulting value; it was using the unexpanded value.
4269 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4270 hadn't been configured. The fix is from Juergen Kreileder, who
4271 understands it better than I do:
4273 "Here's what I see happening with three configured cyrus_sasl
4274 authenticators configured (plain, login, cram-md5):
4276 On startup auth_cyrus_sasl_init() gets called for each of these.
4277 This means three calls to sasl_listmech() without a specified mech_list.
4278 => SASL tests which mechs of all available mechs actually work
4279 => three warnings about OTP not working
4280 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4282 With the patch, sasl_listmech() also gets called three times. But now
4283 SASL's mech_list option is set to the server_mech specified in the the
4284 authenticator. Or in other words, the answer from sasl_listmech()
4285 gets limited to just the mech you're testing for (which is different
4287 => the return list contains just 'plain' or 'login', 'cram-md5' or
4288 nothing depending on the value of ob->server_mech.
4290 I've just tested the patch: Authentication still works fine,
4291 unavailable mechs specified in the exim configuration are still
4292 caught, and the auth.log warnings about OTP are gone."
4294 PH/25 When debugging is enabled, the contents of the command line are added
4295 to the debugging output, even when log_selector=+arguments is not
4298 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4299 answer is "GNU", and only if the return is "GNU/something" is the answer
4302 PH/27 $acl_verify_message is now set immediately after the failure of a
4303 verification in an ACL, and so is available in subsequent modifiers. In
4304 particular, the message can be preserved by coding like this:
4306 warn !verify = sender
4307 set acl_m0 = $acl_verify_message
4309 Previously, $acl_verify_message was set only while expanding "message"
4310 and "log_message" when a very denied access.
4312 PH/28 Modified OS/os.c-Linux with
4314 -#ifndef OS_LOAD_AVERAGE
4315 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4317 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4318 nomenclature these days.)
4320 PH/29 Installed patch from the Sieve maintainer that adds the options
4321 sieve_useraddress and sieve_subaddress to the redirect router.
4323 PH/30 In these circumstances:
4324 . Two addresses routed to the same list of hosts;
4325 . First host does not offer TLS;
4326 . First host accepts first address;
4327 . First host gives temporary error to second address;
4328 . Second host offers TLS and a TLS session is established;
4329 . Second host accepts second address.
4330 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4331 and peerdn, if requested) that were in fact used only for the second
4334 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4335 attention to any local part prefix or suffix that was matched by the
4336 router that accepted the address. It now behaves in the same way as it
4337 does for delivery: the affixes are removed from the local part unless
4338 rcpt_include_affixes is set on the transport.
4340 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4341 timeout during the DATA phase of an incoming message.
4343 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4344 applied a patch sanctioned by the Sieve maintainer.
4346 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4347 the uid or gid is negative. A case of a negative gid caused this to be
4348 noticed. The fix allows for either to be negative.
4350 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4351 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4354 PH/36 The change PH/12 above was broken. Fixed it.
4356 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4357 the grounds that routing the same address twice would always produce the
4358 same answer. This might have been true once, but it is certainly no
4359 longer true now. Routing a child address may depend on the previous
4360 routing that produced that child. Some complicated redirection strategies
4361 went wrong when messages had multiple recipients, and made Exim's
4362 behaviour dependent on the order in which the addresses were given.
4364 I have moved the duplicate checking until after the routing is complete.
4365 Exim scans the addresses that are assigned to local and remote
4366 transports, and removes any duplicates. This means that more work will be
4367 done, as duplicates will always all be routed, but duplicates are
4368 presumably rare, so I don't expect this is of any significance.
4370 For deliveries to pipes, files, and autoreplies, the duplicate checking
4371 still happens during the routing process, since they are not going to be
4374 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4375 It corrects a timeout issue with spamd. This is Ian's comment: "The
4376 background is that sometimes spamd either never reads data from a
4377 connection it has accepted, or it never writes response data. The exiscan
4378 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4379 blindly assumes that writes won't block so it may never time out."
4381 PH/39 Allow G after quota size as well as K and M.
4383 PH/40 The value set for $authenticated_id in an authenticator may not contain
4384 binary zeroes or newlines because the value is written to log lines and
4385 to spool files. There was no check on this. Now the value is run through
4386 the string_printing() function so that such characters are converted to
4387 printable escape sequences.
4389 PH/41 $message_linecount is a new variable that contains the total number of
4390 lines in the message. Compare $body_linecount, which is the count for the
4393 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4394 bounce and delay warning messages, except in certain special cases, which
4397 (a) An SMTP error message from a remote host;
4398 (b) A message specified in a :fail: redirection;
4399 (c) A message specified in a "fail" command in a system filter;
4400 (d) A message specified in a FAIL return from the queryprogram router;
4401 (e) A message specified by the cannot_route_message router option.
4403 In these cases only, Exim does include the error details in bounce and
4404 warning messages. There are also a few cases where bland messages such
4405 as "unrouteable address" or "local delivery error" are given.
4407 PH/43 $value is now also set for the "else" part of a ${run expansion.
4409 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4410 being worked on, but at least Exim now implements the latest version to
4413 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4414 process to complete was treated as a delivery failure, a timeout while
4415 writing the message to the pipe was logged, but erroneously treated as a
4416 successful delivery. Such timeouts include transport filter timeouts. For
4417 consistency with the overall process timeout, these timeouts are now
4418 treated as errors, giving rise to delivery failures by default. However,
4419 there is now a new Boolean option for the pipe transport called
4420 timeout_defer, which, if set TRUE, converts the failures into defers for
4421 both kinds of timeout. A transport filter timeout is now identified in
4424 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4425 systems where "make" and "gmake" are different, calling "gmake" at top
4426 level broke things. I've arranged for the value of $(MAKE) to be passed
4427 from the Makefile to this script so that it can call the same version of
4431 A note about Exim versions 4.44 and 4.50
4432 ----------------------------------------
4434 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4435 changes of various kinds. As a consequence, a big documentation update was
4436 needed. This delayed the release for rather longer than seemed good, especially
4437 in the light of a couple of (minor) security issues. Therefore, the changes
4438 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4439 release. So 4.44 and 4.50 are in effect two different branches that both start
4442 I have left the 4.50 change log unchanged; it contains all the changes since
4443 4.43. The change log for 4.44 is below; many of its items are identical to
4444 those for 4.50. This seems to be the most sensible way to preserve the
4445 historical information.
4451 1. Minor wording change to the doc/README.SIEVE file.
4453 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4454 computation of the current number of files was incorrect.
4456 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4457 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4460 4. Give more explanation in the error message when the command for a transport
4461 filter fails to execute.
4463 5. There are several places where Exim runs a non-Exim command in a
4464 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4465 was being done only for the command run by the queryprogram router. It is
4466 now done for all such subprocesses. The other cases are: ${run, transport
4467 filters, and the commands run by the lmtp and pipe transports.
4469 6. Added CONFIGURE_GROUP build-time option.
4471 7. Some older OS have a limit of 256 on the maximum number of file
4472 descriptors. Exim was using setrlimit() to set 1000 as a large value
4473 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4474 systems. I've change it so that if it can't get 1000, it tries for 256.
4476 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4477 was an oversight, and furthermore, ever since the addition of extra
4478 controls (e.g. 4.43/32), the checks on when to allow different forms of
4479 "control" were broken. There should now be diagnostics for all cases when a
4480 control that does not make sense is encountered.
4482 9. Added the /retain_sender option to "control=submission".
4484 10. $recipients is now available in the predata ACL (oversight).
4486 11. Tidy the search cache before the fork to do a delivery from a message
4487 received from the command line. Otherwise the child will trigger a lookup
4488 failure and thereby defer the delivery if it tries to use (for example) a
4489 cached ldap connection that the parent has called unbind on.
4491 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4492 of $address_data from the recipient verification was clobbered by the
4493 sender verification.
4495 13. The value of address_data from a sender verification is now available in
4496 $sender_address_data in subsequent conditions in the ACL statement.
4498 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4500 15. Added a new option "connect=<time>" to callout options, to set a different
4503 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4504 was its contents. (It was OK if the option was not defined at all.)
4506 17. A "Completed" log line is now written for messages that are removed from
4507 the spool by the -Mrm option.
4509 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4510 information about exactly what failed.
4512 19. Added -dd to debug only the daemon process.
4514 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4515 handles timeouts, both on the server side and network timeouts. Renamed the
4516 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4518 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4519 It is now set to "smtps".
4521 22. $host_address is now set to the target address during the checking of
4522 ignore_target_hosts.
4524 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4525 being passed; this would have caused $sender_host_name to have been used if
4526 matching the list had actually called for a host name (not very likely,
4527 since this list is usually IP addresses). A host name is now passed as
4530 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4531 code that specifies a non-restarting handler (typically sigaction() in
4532 modern systems) in an attempt to fix a rare and obscure crash bug.
4534 25. Narrowed the window for a race in the daemon that could cause it to ignore
4535 SIGCHLD signals. This is not a major problem, because they are used only to
4536 wake it up if nothing else does.
4538 26. A malformed maildirsize file could cause Exim to calculate negative values
4539 for the mailbox size or file count. Odd effects could occur as a result.
4540 The maildirsize information is now recalculated if the size or filecount
4543 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4544 support for a long time. Removed HAVE_SYS_VFS_H.
4546 28. Installed the latest version of exipick from John Jetmore.
4548 29. In an address list, if the pattern was not a regular expression, an empty
4549 subject address (from a bounce message) matched only if the pattern was an
4550 empty string. Non-empty patterns were not even tested. This was the wrong
4551 because it is perfectly reasonable to use an empty address as part of a
4552 database query. An empty address is now tested by patterns that are
4553 lookups. However, all the other forms of pattern expect the subject to
4554 contain a local part and a domain, and therefore, for them, an empty
4555 address still always fails if the pattern is not itself empty.
4557 30. Exim went into a mad DNS loop when attempting to do a callout where the
4558 host was specified on an smtp transport, and looking it up yielded more
4559 than one IP address.
4561 31. Re-factored the code for checking spool and log partition space into a
4562 function that finds that data and another that does the check. The former
4563 is then used to implement four new variables: $spool_space, $log_space,
4564 $spool_inodes, and $log_inodes.
4566 32. The RFC2047 encoding function was originally intended for short strings
4567 such as real names; it was not keeping to the 75-character limit for
4568 encoded words that the RFC imposes. It now respects the limit, and
4569 generates multiple encoded words if necessary. To be on the safe side, I
4570 have increased the buffer size for the ${rfc2047: expansion operator from
4573 33. It is now permitted to omit both strings after an "if" condition; if the
4574 condition is true, the result is "true". As before, when the second string
4575 is omitted, a false condition yields an empty string. This makes it less
4576 cumbersome to write custom ACL and router conditions.
4578 34. Failure to deliver a bounce message always caused it to be frozen, even if
4579 there was an errors_to setting on the router. The errors_to setting is now
4582 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4583 canonical form (fully expanded) before being placed in
4584 $sender_host_address.
4586 36. The table in the code that translates DNS record types into text (T_A to
4587 "A" for instance) was missing entries for NS and CNAME. It is just possible
4588 that this could have caused confusion if both these types were looked up
4589 for the same domain, because the text type is used as part of Exim's
4590 per-process caching. But the chance of anyone hitting this buglet seems
4593 37. The dnsdb lookup has been extended in a number of ways.
4595 (1) There is a new type, "zns", which walks up the domain tree until it
4596 finds some nameserver records. It should be used with care.
4598 (2) There is a new type, "mxh", which is like "mx" except that it returns
4599 just the host names, not the priorities.
4601 (3) It is now possible to give a list of domains (or IP addresses) to be
4602 looked up. The behaviour when one of the lookups defers can be
4603 controlled by a keyword.
4605 (4) It is now possible to specify the separator character for use when
4606 multiple records are returned.
4608 38. The dnslists ACL condition has been extended: it is now possible to supply
4609 a list of IP addresses and/or domains to be looked up in a particular DNS
4612 39. Added log_selector=+queue_time_overall.
4614 40. When running the queue in the test harness, wait just a tad after forking a
4615 delivery process, to get repeatability of debugging output.
4617 41. Include certificate and key file names in error message when GnuTLS fails
4618 to set them up, because the GnuTLS error message doesn't include the name
4619 of the failing file when there is a problem reading it.
4621 42. Allow both -bf and -bF in the same test run.
4623 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4625 44. The "Exiscan patch" is now merged into the mainline Exim source.
4627 45. Sometimes the final signoff response after QUIT could fail to get
4628 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4629 < 0 before doing a fflush(). This bug looks as though it goes back to the
4630 introduction of TLS in release 3.20, but "sometimes" must have been rare
4631 because the tests only now provoked it.
4633 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4634 (this can affect the format of dates).
4636 47. exim_tidydb, when checking for the continued existence of a message for
4637 which it has found a message-specific retry record, was not finding
4638 messages that were in split spool directories. Consequently, it was
4639 deleting retry records that should have stayed in existence.
4641 48. Steve fixed some bugs in eximstats.
4643 49. The SPA authentication driver was not abandoning authentication and moving
4644 on to the next authenticator when an expansion was forced to fail,
4645 contradicting the general specification for all authenticators. Instead it
4646 was generating a temporary error. It now behaves as specified.
4648 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4649 (the order specifies the preference for clients). The order is now AES256,
4650 AES128, 3DES, ARCFOUR128.
4652 51. Small patch to Sieve code - explicitly set From: when generating an
4655 52. Exim crashed if a remote delivery caused a very long error message to be
4656 recorded - for instance if somebody sent an entire SpamAssassin report back
4657 as a large number of 550 error lines. This bug was coincidentally fixed by
4658 increasing the size of one of Exim's internal buffers (big_buffer) that
4659 happened as part of the Exiscan merge. However, to be on the safe side, I
4660 have made the code more robust (and fixed the comments that describe what
4663 53. Now that there can be additional text after "Completed" in log lines (if
4664 the queue_time_overall log selector is set), a one-byte patch to exigrep
4665 was needed to allow it to recognize "Completed" as not the last thing in
4668 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4669 patch that reportedly fixes this has been added. I am not expert enough to
4670 create a test for it. This is what the patch creator wrote:
4672 "I found a little strange behaviour of ldap code when working with
4673 Windows 2003 AD Domain, where users was placed in more than one
4674 Organization Units. When I tried to give exim partial DN, the exit code
4675 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4676 But simultaneously result of request was absolutely normal ldap result,
4677 so I produce this patch..."
4679 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4680 REFERENCE, so I have modified the code to exclude the patch when that macro
4683 55. Some experimental protocols are using DNS PTR records for new purposes. The
4684 keys for these records are domain names, not reversed IP addresses. The
4685 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4686 leaves it alone. Component reversal etc. now happens only for IP addresses.
4689 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4691 57. Double the size of the debug message buffer (to 2048) so that more of very
4692 long debug lines gets shown.
4694 58. The exicyclog utility now does better if the number of log files to keep
4695 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4697 59. Two changes related to the smtp_active_hostname option:
4699 (1) $smtp_active_hostname is now available as a variable.
4700 (2) The default for smtp_banner uses $smtp_active_hostname instead
4701 of $primary_hostname.
4703 60. The host_aton() function is supposed to be passed a string that is known
4704 to be a valid IP address. However, in the case of IPv6 addresses, it was
4705 not checking this. This is a hostage to fortune. Exim now panics and dies
4706 if the condition is not met. A case was found where this could be provoked
4707 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4708 components; fortuitously, this particular loophole had already been fixed
4709 by change 4.50/55 above.
4711 If there are any other similar loopholes, the new check in host_aton()
4712 itself should stop them being exploited. The report I received stated that
4713 data on the command line could provoke the exploit when Exim was running as
4714 exim, but did not say which command line option was involved. All I could
4715 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4716 running as the user.
4719 61. There was a buffer overflow vulnerability in the SPA authentication code
4720 (which came originally from the Samba project). I have added a test to the
4721 spa_base64_to_bits() function which I hope fixes it.
4724 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4725 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4727 63. The daemon start-up calls getloadavg() while still root for those OS that
4728 need the first call to be done as root, but it missed one case: when
4729 deliver_queue_load_max is set with deliver_drop_privilege. This is
4730 necessary for the benefit of the queue runner, because there is no re-exec
4731 when deliver_drop_privilege is set.
4733 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4734 This has been fixed.
4736 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4737 was in use, was not putting the data itself into the right store pool;
4738 consequently, it could be overwritten for a subsequent message in the same
4739 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4742 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4744 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4745 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4746 the function were treating the return as a boolean value, which happened to
4747 work because 0=false and not-0=true, but is not correct code.
4749 68. The host_aton() function was not handling scoped IPv6 addresses (those
4750 with, for example, "%eth0" on the end) correctly.
4752 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4753 negated items (that is, ~something) in unsigned ints. Some compilers
4754 apparently mutter when there is no cast.
4756 70. If an address verification called from an ACL failed, and did not produce a
4757 user-specific message (i.e. there was only a "system" message), nothing was
4758 put in $acl_verify_message. In this situation, it now puts the system
4761 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4762 session; change 4.31/43 added the unwanted input to the log line - except
4763 that it did not do this in the start of session case. It now does.
4765 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4766 This is wrong and can cause the other end to generate a synchronization
4767 error if it is another Exim or anything else that does the synchronization
4768 check. A QUIT command is no longer sent after a timeout.
4770 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4771 during host lookups.
4773 74. The defer_ok option of callout verification was not working if it was used
4774 when verifying addresses in header lines, that is, for this case:
4776 verify = header_sender/callout=defer_ok
4778 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4779 those file descriptors could be used for SMTP connections. If anything
4780 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4781 could be sent to the SMTP client, causing chaos. The daemon now opens
4782 stdin, stdout, and stderr to /dev/null when it puts itself into the
4785 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4786 log by default. The user can override this with suitable Perl magic.
4788 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4789 the log message when discard triggers, was not working for the DATA ACL or
4790 for the non-SMTP ACL.
4792 78. Error message wording change in sieve.c.
4794 79. If smtp_accept_max_per_host was set, the number of connections could be
4795 restricted to fewer than expected, because the daemon was trying to set up
4796 a new connection before checking whether the processes handling previous
4797 connections had finished. The check for completed processes is now done
4798 earlier. On busy systems, this bug wouldn't be noticed because something
4799 else would have woken the daemon, and it would have reaped the completed
4802 80. If a message was submitted locally by a user whose login name contained one
4803 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4804 It caused a spool format error. I have fixed the spool reading code. A
4805 related problem was that the "from" clause in the Received: line became
4806 illegal because of the space(s). It is now covered by ${quote_local_part.
4808 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4811 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4813 83. Patch from Sieve maintainer to fix unterminated string problem in
4814 "vacation" handling.
4816 84. Some minor changes to the Linux configuration files to help with other
4817 OS variants using glibc.
4819 85. One more patch for Sieve to update vacation handling to latest spec.
4822 ----------------------------------------------------
4823 See the note above about the 4.44 and 4.50 releases.
4824 ----------------------------------------------------
4830 1. Change 4.43/35 introduced a bug that caused file counts to be
4831 incorrectly computed when quota_filecount was set in an appendfile
4834 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4835 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4838 3. Give more explanation in the error message when the command for a transport
4839 filter fails to execute.
4841 4. There are several places where Exim runs a non-Exim command in a
4842 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4843 was being done only for the command run by the queryprogram router. It is
4844 now done for all such subprocesses. The other cases are: ${run, transport
4845 filters, and the commands run by the lmtp and pipe transports.
4847 5. Some older OS have a limit of 256 on the maximum number of file
4848 descriptors. Exim was using setrlimit() to set 1000 as a large value
4849 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4850 systems. I've change it so that if it can't get 1000, it tries for 256.
4852 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4853 was an oversight, and furthermore, ever since the addition of extra
4854 controls (e.g. 4.43/32), the checks on when to allow different forms of
4855 "control" were broken. There should now be diagnostics for all cases when a
4856 control that does not make sense is encountered.
4858 7. $recipients is now available in the predata ACL (oversight).
4860 8. Tidy the search cache before the fork to do a delivery from a message
4861 received from the command line. Otherwise the child will trigger a lookup
4862 failure and thereby defer the delivery if it tries to use (for example) a
4863 cached ldap connection that the parent has called unbind on.
4865 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4866 of $address_data from the recipient verification was clobbered by the
4867 sender verification.
4869 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4870 was its contents. (It was OK if the option was not defined at all.)
4872 11. A "Completed" log line is now written for messages that are removed from
4873 the spool by the -Mrm option.
4875 12. $host_address is now set to the target address during the checking of
4876 ignore_target_hosts.
4878 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4879 being passed; this would have caused $sender_host_name to have been used if
4880 matching the list had actually called for a host name (not very likely,
4881 since this list is usually IP addresses). A host name is now passed as
4884 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4885 code that specifies a non-restarting handler (typically sigaction() in
4886 modern systems) in an attempt to fix a rare and obscure crash bug.
4888 15. Narrowed the window for a race in the daemon that could cause it to ignore
4889 SIGCHLD signals. This is not a major problem, because they are used only to
4890 wake it up if nothing else does.
4892 16. A malformed maildirsize file could cause Exim to calculate negative values
4893 for the mailbox size or file count. Odd effects could occur as a result.
4894 The maildirsize information is now recalculated if the size or filecount
4897 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4898 support for a long time. Removed HAVE_SYS_VFS_H.
4900 18. Updated exipick to current release from John Jetmore.
4902 19. Allow an empty sender to be matched against a lookup in an address list.
4903 Previously the only cases considered were a regular expression, or an
4906 20. Exim went into a mad DNS lookup loop when doing a callout where the
4907 host was specified on the transport, if the DNS lookup yielded more than
4910 21. The RFC2047 encoding function was originally intended for short strings
4911 such as real names; it was not keeping to the 75-character limit for
4912 encoded words that the RFC imposes. It now respects the limit, and
4913 generates multiple encoded words if necessary. To be on the safe side, I
4914 have increased the buffer size for the ${rfc2047: expansion operator from
4917 22. Failure to deliver a bounce message always caused it to be frozen, even if
4918 there was an errors_to setting on the router. The errors_to setting is now
4921 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4922 canonical form (fully expanded) before being placed in
4923 $sender_host_address.
4925 24. Updated eximstats to version 1.33
4927 25. Include certificate and key file names in error message when GnuTLS fails
4928 to set them up, because the GnuTLS error message doesn't include the name
4929 of the failing file when there is a problem reading it.
4931 26. Expand error message when OpenSSL has problems setting up cert/key files.
4934 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4935 (this can affect the format of dates).
4937 28. exim_tidydb, when checking for the continued existence of a message for
4938 which it has found a message-specific retry record, was not finding
4939 messages that were in split spool directories. Consequently, it was
4940 deleting retry records that should have stayed in existence.
4942 29. eximstats updated to version 1.35
4943 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4944 1.35 - bugfix such that pie charts by volume are generated correctly
4946 30. The SPA authentication driver was not abandoning authentication and moving
4947 on to the next authenticator when an expansion was forced to fail,
4948 contradicting the general specification for all authenticators. Instead it
4949 was generating a temporary error. It now behaves as specified.
4951 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4952 (the order specifies the preference for clients). The order is now AES256,
4953 AES128, 3DES, ARCFOUR128.
4955 31. Small patch to Sieve code - explicitly set From: when generating an
4958 32. Exim crashed if a remote delivery caused a very long error message to be
4959 recorded - for instance if somebody sent an entire SpamAssassin report back
4960 as a large number of 550 error lines. This bug was coincidentally fixed by
4961 increasing the size of one of Exim's internal buffers (big_buffer) that
4962 happened as part of the Exiscan merge. However, to be on the safe side, I
4963 have made the code more robust (and fixed the comments that describe what
4966 33. Some experimental protocols are using DNS PTR records for new purposes. The
4967 keys for these records are domain names, not reversed IP addresses. The
4968 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4969 leaves it alone. Component reversal etc. now happens only for IP addresses.
4972 34. The host_aton() function is supposed to be passed a string that is known
4973 to be a valid IP address. However, in the case of IPv6 addresses, it was
4974 not checking this. This is a hostage to fortune. Exim now panics and dies
4975 if the condition is not met. A case was found where this could be provoked
4976 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4977 components; fortuitously, this particular loophole had already been fixed
4978 by change 4.50/55 or 4.44/33 above.
4980 If there are any other similar loopholes, the new check in host_aton()
4981 itself should stop them being exploited. The report I received stated that
4982 data on the command line could provoke the exploit when Exim was running as
4983 exim, but did not say which command line option was involved. All I could
4984 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4985 running as the user.
4988 35. There was a buffer overflow vulnerability in the SPA authentication code
4989 (which came originally from the Samba project). I have added a test to the
4990 spa_base64_to_bits() function which I hope fixes it.
4993 36. The daemon start-up calls getloadavg() while still root for those OS that
4994 need the first call to be done as root, but it missed one case: when
4995 deliver_queue_load_max is set with deliver_drop_privilege. This is
4996 necessary for the benefit of the queue runner, because there is no re-exec
4997 when deliver_drop_privilege is set.
4999 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5000 was in use, was not putting the data itself into the right store pool;
5001 consequently, it could be overwritten for a subsequent message in the same
5002 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5005 38. Sometimes the final signoff response after QUIT could fail to get
5006 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5007 < 0 before doing a fflush(). This bug looks as though it goes back to the
5008 introduction of TLS in release 3.20, but "sometimes" must have been rare
5009 because the tests only now provoked it.
5015 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5016 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5017 Now it returns an integer. A number of calls were still expecting a T/F
5018 return. Fortuitously, in all cases, the tests worked in OK situations,
5019 which is the norm. However, things would have gone wrong on any write
5020 failures on the smtp file descriptor. This function is used when sending
5021 messages over SMTP and also when doing verify callouts.
5023 2. When Exim is called to do synchronous delivery of a locally submitted
5024 message (the -odf or -odi options), it no longer closes stderr before doing
5027 3. Implemented the mua_wrapper option.
5029 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5031 5. Implemented the functions header_remove(), header_testname(),
5032 header_add_at_position(), and receive_remove_recipient(), and exported them
5035 6. If an ACL "warn" statement specified the addition of headers, Exim already
5036 inserted X-ACL-Warn: at the start if there was no header name. However, it
5037 was not making this test for the second and subsequent header lines if
5038 there were newlines in the string. This meant that an invalid header could
5039 be inserted if Exim was badly configured.
5041 7. Allow an ACL "warn" statement to add header lines at the start or after all
5042 the Received: headers, as well as at the end.
5044 8. Added the rcpt_4xx retry error code.
5046 9. Added postmaster_mailfrom=xxx to callout verification option.
5048 10. Added mailfrom=xxxx to the callout verification option, for verify=
5051 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5052 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5054 12. Inserted some casts to stop certain compilers warning when using pointer
5055 differences as field lengths or precisions in printf-type calls (mostly
5056 affecting debugging statements).
5058 13. Added optional readline() support for -be (dynamically loaded).
5060 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5061 same clock tick as a message's arrival, so that its received time was the
5062 same as the "first fail" time on the retry record, and that message
5063 remained on the queue past the ultimate address timeout, every queue runner
5064 would try a delivery (because it was past the ultimate address timeout) but
5065 after another failure, the ultimate address timeout, which should have then
5066 bounced the address, did not kick in. This was a "< instead of <=" error;
5067 in most cases the first failure would have been in the next clock tick
5068 after the received time, and all would be well.
5070 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5071 being recognized when the domain list was tested by the match_domain
5072 condition in an expansion string.
5074 16. Added the ${str2b64: operator.
5076 17. Exim was always calling setrlimit() to set a large limit for the number of
5077 processes, without checking whether the existing limit was already
5078 adequate. (It did check for the limit on file descriptors.) Furthermore,
5079 errors from getrlimit() and setrlimit() were being ignored. Now they are
5080 logged to the main and panic logs, but Exim does carry on, to try to do its
5081 job under whatever limits there are.
5083 18. Imported PCRE 5.0.
5085 19. Trivial typo in log message " temporarily refused connection" (the leading
5088 20. If the log selector return_path_on_delivery was set and an address was
5089 redirected to /dev/null, the delivery process crashed because it assumed
5090 that a return path would always be set for a "successful" delivery. In this
5091 case, the whole delivery is bypassed as an optimization, and therefore no
5094 21. Internal re-arrangement: the function for sending a challenge and reading
5095 a response while authentication was assuming a zero-terminated challenge
5096 string. It's now changed to take a pointer and a length, to allow for
5097 binary data in such strings.
5099 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5101 23. Exim was not respecting finduser_retries when seeking the login of the
5102 uid under which it was called; it was always trying 10 times. (The default
5103 setting of finduser_retries is zero.) Also, it was sleeping after the final
5104 failure, which is pointless.
5106 24. Implemented tls_on_connect_ports.
5108 25. Implemented acl_smtp_predata.
5110 26. If the domain in control=submission is set empty, Exim assumes that the
5111 authenticated id is a complete email address when it generates From: or
5112 Sender: header lines.
5114 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5115 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5116 chown and chgrp in /bin and hostname in /usr/bin.
5118 28. Exim was keeping the "process log" file open after each use, just as it
5119 does for the main log. This opens the possibility of it remaining open for
5120 long periods when the USR1 signal hits a daemon. Occasional processlog
5121 errors were reported, that could have been caused by this. Anyway, it seems
5122 much more sensible not to leave this file open at all, so that is what now
5125 29. The long-running daemon process does not normally write to the log once it
5126 has entered its main loop, and it closes the log before doing so. This is
5127 so that log files can straightforwardly be renamed and moved. However,
5128 there are a couple of unusual error situations where the daemon does write
5129 log entries, and I had neglected to close the log afterwards.
5131 30. The text of an SMTP error response that was received during a remote
5132 delivery was being truncated at 512 bytes. This is too short for some of
5133 the long messages that one sometimes sees. I've increased the limit to
5136 31. It is now possible to make retry rules that apply only when a message has a
5137 specific sender, in particular, an empty sender.
5139 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5140 it possible to be selective about when SMTP synchronization is enforced.
5142 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5144 32. Implemented hosts_connection_nolog.
5146 33. Added an ACL for QUIT.
5148 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5151 35. Added mailbox_size and mailbox_filecount to appendfile.
5153 36. Added control = no_multiline_responses to ACLs.
5155 37. There was a bug in the logic of the code that waits for the clock to tick
5156 in the case where the clock went backwards by a substantial amount such
5157 that the microsecond fraction of "now" was more than the microsecond
5158 fraction of "then" (but the whole seconds number was less).
5160 38. Added support for the libradius Radius client library this is found on
5161 FreeBSD (previously only the radiusclient library was supported).
5167 1. When certain lookups returned multiple values in the form name=value, the
5168 quoting of the values was not always being done properly. Specifically:
5169 (a) If the value started with a double quote, but contained no whitespace,
5171 (b) If the value contained whitespace other than a space character (i.e.
5172 tabs or newlines or carriage returns) it was not quoted.
5173 This fix has been applied to the mysql and pgsql lookups by writing a
5174 separate quoting function and calling it from the lookup code. The fix
5175 should probably also be applied to nisplus, ibase and oracle lookups, but
5176 since I cannot test any of those, I have not disturbed their existing code.
5178 2. A hit in the callout cache for a specific address caused a log line with no
5179 reason for rejecting RCPT. Now it says "Previous (cached) callout
5180 verification failure".
5182 3. There was an off-by-one bug in the queryprogram router. An over-long
5183 return line was truncated at 256 instead of 255 characters, thereby
5184 overflowing its buffer with the terminating zero. As well as fixing this, I
5185 have increased the buffer size to 1024 (and made a note to document this).
5187 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5188 when Exim is waiting for an SMTP response from a remote server, Exim
5189 restarts its select() call on the socket, thereby resetting its timeout.
5190 This is not a problem when such interrupts are rare. Somebody set up a cron
5191 job to run exiwhat every 2 minutes, which is less than the normal select()
5192 timeout (5 or 10 minutes). This meant that the select() timeout never
5193 kicked in because it was always reset. I have fixed this by comparing the
5194 time when an interrupt arrives with the time at the start of the first call
5195 to select(). If more time than the timeout has elapsed, the interrupt is
5196 treated as a timeout.
5198 5. Some internal re-factoring in preparation for the addition of Sieve
5199 extensions (by MH). In particular, the "personal" test is moved to a
5200 separate function, and given an option for scanning Cc: and Bcc: (which is
5201 not set for Exim filters).
5203 6. When Exim created an email address using the login of the caller as the
5204 local part (e.g. when creating a From: or Sender: header line), it was not
5205 quoting the local part when it contained special characters such as @.
5207 7. Installed new OpenBSD configuration files.
5209 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5210 try to make them clearer.
5212 9. Callout options, other than the timeout value, were being ignored when
5213 verifying sender addresses in header lines. For example, when using
5215 verify = header_sender/callout=no_cache
5217 the cache was (incorrectly) being used.
5219 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5220 only the Cygwin environment.
5222 11. When return_path_on_delivery was set as a log selector, if different remote
5223 addresses in the same message used different return paths and parallel
5224 remote delivery occurred, the wrong values would sometimes be logged.
5225 (Whenever a remote delivery process finished, the return path value from
5226 the most recently started remote delivery process was logged.)
5228 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5229 lines when AUTH and/or TLS are in use. This is the "received protocol"
5230 field. Exim used to use "asmtp" for authenticated SMTP, without any
5231 indication (in the protocol name) for TLS use. Now it follows the RFC and
5232 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5233 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5234 names appear in log lines as well as in Received: header lines.
5236 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5237 extensions, and comparison tests, and to fix some bugs.
5239 14. Changes to the "personal" filter test:
5241 (1) The test was buggy in that it was just doing the equivalent of
5242 "contains" tests on header lines. For example, if a user's address was
5243 anne@some.where, the "personal" test would incorrectly be true for
5245 To: susanne@some.where
5247 This test is now done by extracting each address from the header in turn,
5248 and checking the entire address. Other tests that are part of "personal"
5249 are now done using regular expressions (for example, to check local parts
5250 of addresses in From: header lines).
5252 (2) The list of non-personal local parts in From: addresses has been
5253 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5254 taken from the Sieve specification recommendations.
5256 (3) If the message contains any header line starting with "List-" it is
5257 treated as non-personal.
5259 (4) The test for "circular" in the Subject: header line has been removed
5260 because it now seems ill-conceived.
5262 15. Minor typos in src/EDITME comments corrected.
5264 16. Installed latest exipick from John Jetmore.
5266 17. If headers_add on a router specified a text string that was too long for
5267 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5268 of string_sprintf() is now avoided.
5270 18. $message_body_size was not set (it was always zero) when running the DATA
5271 ACL and the local_scan() function.
5273 19. For the "mail" command in an Exim filter, no default was being set for
5274 the once_repeat time, causing a random time value to be used if "once" was
5275 specified. (If the value happened to be <= 0, no repeat happened.) The
5276 default is now 0s, meaning "never repeat". The "vacation" command was OK
5277 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5278 (I found it when inspecting the code).
5280 20. There is now an overall timeout for performing a callout verification. It
5281 defaults to 4 times the callout timeout, which applies to individual SMTP
5282 commands during the callout. The overall timeout applies when there is more
5283 than one host that can be tried. The timeout is checked before trying the
5284 next host. This prevents very long delays if there are a large number of
5285 hosts and all are timing out (e.g. when the network connections are timing
5286 out). The value of the overall timeout can be changed by specifying an
5287 additional sub-option for "callout", called "maxwait". For example:
5289 verify = sender/callout=5s,maxwait=20s
5291 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5292 to the end before writing, but this should make it even safer).
5294 22. Exim was forgetting that it had advertised PIPELINING for the second and
5295 subsequent messages on an SMTP connection. It was also not resetting its
5296 memory on STARTTLS and an internal HELO.
5298 23. When Exim logs an SMTP synchronization error within a session, it now
5299 records whether PIPELINING has been advertised or not.
5301 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5302 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5303 rather than long int.
5305 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5307 26. Added the never_mail option to autoreply.
5313 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5314 crash if the getsockname() call failed; this can happen if a connection is
5315 closed very soon after it is established. The problem was simply in the
5316 order in which certain operations were done, causing Exim to try to write
5317 to the SMTP stream before it had set up the file descriptor. The bug has
5318 been fixed by making things happen in the correct order.
5324 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5325 before the connection was closed, thus losing the rejection response.
5327 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5328 some early Solaris releases, but causes trouble in current releases where
5329 socklen_t is defined.
5331 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5334 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5337 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5338 that releasing the top of it at the end releases what was used for sub-
5339 expansions (unless the block got too big). However, discard this block if
5340 the first thing is a variable or header, so that we can use its block when
5341 it is dynamic (useful for very large $message_headers, for example).
5343 6. Lookups now cache *every* query, not just the most recent. A new, separate
5344 store pool is used for this. It can be recovered when all lookup caches are
5345 flushed. Lookups now release memory at the end of their result strings.
5346 This has involved some general refactoring of the lookup sources.
5348 7. Some code has been added to the store_xxx() functions to reduce the amount
5349 of flapping under certain conditions.
5351 8. log_incoming_interface used to affect only the <= reception log lines. Now
5352 it causes the local interface and port to be added to several more SMTP log
5353 lines, for example "SMTP connection from", and rejection lines.
5355 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5357 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5359 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5360 could be overwritten at the end of the current message (or the start of a
5361 new message if it was set in a HELO ACL). The value is now preserved for
5362 the duration of the SMTP connection.
5364 12. If a transport had a headers_rewrite setting, and a matching header line
5365 contained an unqualified address, that address was qualified, even if it
5366 did not match any rewriting rules. The underlying bug was that the values
5367 of the flags that permit the existence of unqualified sender and recipient
5368 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5369 non-local messages, and by -bnq for local messages) were not being
5370 preserved with the message after it was received.
5372 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5373 "next input=" as part of the text comprising the host identity instead of
5374 the correct text. The code was using the same buffer for two different
5375 strings. However, depending on which order the printing function evaluated
5376 its arguments, the bug did not always show up. Under Linux, for example, my
5377 test suite worked just fine.
5379 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5380 doesn't work with some older versions of Perl. It has been changed to "my",
5381 which in any case is probably the better facility to use.
5383 15. A really picky compiler found some instances of statements for creating
5384 error messages that either had too many or two few arguments for the format
5387 16. The size of the buffer for calls to the DNS resolver has been increased
5388 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5389 for addresses that have a lot of PTR records. This alleviates a problem; it
5390 does not fully solve it.
5392 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5393 buffer now truncates the list and logs the incident, which is the same
5394 action as happens when Exim is looking up a host name and its aliases.
5395 Previously in this situation something unpredictable would happen;
5396 sometimes it was "internal error: store_reset failed".
5398 18. If a server dropped the connection unexpectedly when an Exim client was
5399 using GnuTLS and trying to read a response, the client delivery process
5400 crashed while trying to generate an error log message.
5402 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5403 string, for example:
5405 warn message = H1: something\nH2: something
5407 the text was added as a single header line from Exim's point of view
5408 though it ended up OK in the delivered message. However, searching for the
5409 second and subsequent header lines using $h_h2: did not work. This has been
5410 fixed. Similarly, if a system filter added multiple headers in this way,
5411 the routers could not see them.
5413 20. Expanded the error message when iplsearch is called with an invalid key to
5414 suggest using net-iplsearch in a host list.
5416 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5417 ACLs are no longer actually imposed (and a message to that effect is
5420 22. If a "gecos" field in a passwd entry contained escaped characters, in
5421 particular, if it contained a \" sequence, Exim got it wrong when building
5422 a From: or a Sender: header from that name. A second bug also caused
5423 incorrect handling when an unquoted " was present following a character
5424 that needed quoting.
5426 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5427 was not being matched caselessly.
5429 24. Arranged for all hyphens in the exim.8 source to be escaped with
5432 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5433 properly. Recipient callout cache records were still being keyed to include
5434 the sender, even when use_sender was set false. This led to far more
5435 callouts that were necessary. The sender is no longer included in the key
5436 when use_sender is false.
5438 26. Added "control = submission" modifier to ACLs.
5440 27. Added the ${base62d: operator to decode base 62 numbers.
5442 28. dnsdb lookups can now access SRV records.
5444 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5445 the configuration file.
5447 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5448 (-v) mode. This makes the output for a verbose queue run more intelligible.
5450 31. Added a use_postmaster feature to recipient callouts.
5452 32. Added the $body_zerocount variable, containing the number of binary zero
5453 bytes in the message body.
5455 33. The time of last modification of the "new" subdirectory is now used as the
5456 "mailbox time last read" when there is a quota error for a maildir
5459 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5461 35. Added +ignore_unknown as a special item in host lists.
5463 36. Code for decoding IPv6 addresses in host lists is now included, even if
5464 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5465 address was recognized as an IP address, but was then not correctly decoded
5466 into binary, causing unexpected and incorrect effects when compared with
5473 1. Very minor rewording of debugging text in manualroute to say "list of
5474 hosts" instead of "hostlist".
5476 2. If verify=header_syntax was set, and a header line with an unqualified
5477 address (no domain) and a large number of spaces between the end of the
5478 name and the colon was received, the reception process suffered a buffer
5479 overflow, and (when I tested it) crashed. This was caused by some obsolete
5480 code that should have been removed. The fix is to remove it!
5482 3. When running in the test harness, delay a bit after writing a bounce
5483 message to get a bit more predictability in the log output.
5485 4. Added a call to search_tidyup() just before forking a reception process. In
5486 theory, someone could use a lookup in the expansion of smtp_accept_max_
5487 per_host which, without the tidyup, could leave open a database connection.
5489 5. Added the variables $recipient_data and $sender_data which get set from a
5490 lookup success in an ACL "recipients" or "senders" condition, or a router
5491 "senders" option, similar to $domain_data and $local_part_data.
5493 6. Moved the writing of debug_print from before to after the "senders" test
5496 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5497 problems for message scanning, either using a data ACL, or using
5498 local_scan() because the Received: header was not generated till after they
5499 were called (in order to set the time as the time of reception completion).
5500 I have revised the way this works. The header is now generated after the
5501 body is received, but before the ACL or local_scan() are called. After they
5502 are run, the timestamp in the header is updated.
5508 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5509 before starting a queue runner without re-exec. This happened only when
5510 deliver_drop_privilege was set or when the Exim user was set to root. The
5511 effect of the bug was that timeouts during subsequent deliveries caused
5512 crashes instead of being properly handled. The handler is now left at its
5513 default (and expected) setting.
5515 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5516 message, again when deliver_drop_privilege is set or Exim is run as root.
5517 The bug described in (1) was not present in this case, but the tidying up
5518 of the other signals was missing. I have made the two cases consistent.
5520 3. The ignore_target_hosts setting on a manualroute router was being ignored
5521 for hosts that were looked up using the /MX notation.
5523 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5526 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5527 operated on the sender address. After changing the $sender_address to <>
5528 for the sender address verify, Exim was re-instated it as the original
5529 (before rewriting) address, but remembering that it had rewritten it, so it
5530 wasn't rewriting it again. This bug also had the effect of breaking the
5531 sender address verification caching when the sender address was rewritten.
5533 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5534 This has been changed so that if the ip literal address matches
5535 ignore_target_hosts, the router declines.
5537 7. Added expansion conditions match_domain, match_address, and match_local_
5538 part (NOT match_host).
5540 8. The placeholder for the Received: header didn't have a length field set.
5542 9. Added code to Exim itself and to exim_lock to test for a specific race
5543 condition that could lead to file corruption when using MBX delivery. The
5544 issue is with the lockfile that is created in /tmp. If this file is removed
5545 after a process has opened it but before that process has acquired a lock,
5546 there is the potential for a second process to recreate the file and also
5547 acquire a lock. This could lead to two Exim processes writing to the file
5548 at the same time. The added code performs the same test as UW imapd; it
5549 checks after acquiring the lock that its file descriptor still refers to
5550 the same named file.
5552 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5553 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5556 11. Added the smtp_active_hostname option. If used, this will typically be made
5557 to depend on the incoming interface address. Because $interface_address is
5558 not set up until the daemon has forked a reception process, error responses
5559 that can happen earlier (such as "too many connections") no longer contain
5562 12. If an expansion in a condition on a "warn" statement fails because a lookup
5563 defers, the "warn" statement is abandoned, and the next ACL statement is
5564 processed. Previously this caused the whole ACL to be aborted.
5566 13. Added the iplsearch lookup type.
5568 14. Added ident_timeout as a log selector.
5570 15. Added tls_certificate_verified as a log selector.
5572 16. Added a global option tls_require_ciphers (compare the smtp transport
5573 option of the same name). This controls incoming TLS connections.
5575 17. I finally figured out how to make tls_require_ciphers do a similar thing
5576 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5577 before starting the TLS session.
5579 18. Tabs are now shown as \t in -bP output.
5581 19. If the log selector return_path_on_delivery was set, Exim crashed when
5582 bouncing a message because it had too many Received: header lines.
5584 20. If two routers both had headers_remove settings, and the first one included
5585 a superfluous trailing colon, the final name in the first list and the
5586 first name in the second list were incorrectly joined into one item (with a
5587 colon in the middle).
5593 1. Added -C and -D options to the exinext utility, mainly to make it easier
5594 to include in the automated testing, but these could be helpful when
5595 multiple configurations are in use.
5597 2. The exinext utility was not formatting the output nicely when there was
5598 an alternate port involved in the retry record key, nor when there was a
5599 message id as well (for retries that were specific to a specific message
5600 and a specific host). It was also confused by IPv6 addresses, because of
5601 the additional colons they contain. I have fixed the IPv4 problem, and
5602 patched it up to do a reasonable job for IPv6.
5604 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5605 delivery, the log line now contains "pipelined" if PIPELINING was used.
5607 4. An SMTP transport process used to panic and die if the bind() call to set
5608 an explicit outgoing interface failed. This has been changed; it is now
5609 treated in the same way as a connect() failure.
5611 5. A reference to $sender_host_name in the part of a conditional expansion
5612 that was being skipped was still causing a DNS lookup. This no longer
5615 6. The def: expansion condition was not recognizing references to header lines
5616 that used bh_ and bheader_.
5618 7. Added the _cache feature to named lists.
5620 8. The code for checking quota_filecount in the appendfile transport was
5621 allowing one more file than it should have been.
5623 9. For compatibility with Sendmail, the command line option
5631 and sets the incoming protocol and host name (for trusted callers). The
5632 host name and its colon can be omitted when only the protocol is to be set.
5633 Note the Exim already has two private options, -pd and -ps, that refer to
5634 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5635 "s", but I don't think that's a major issue.
5637 10. A number of refactoring changes to the code, none of which should affect
5640 (a) The number of logging options was getting close to filling up the
5641 32-bit word that was used as a bit map. I have split them into two classes:
5642 those that are passed in the argument to log_write(), and those that are
5643 only ever tested independently outside of that function. These are now in
5644 separate 32-bit words, so there is plenty of room for expansion again.
5645 There is no change in the user interface or the logging behaviour.
5647 (b) When building, for example, log lines, the code previously used a
5648 macro that called string_cat() twice, in order to add two strings. This is
5649 not really sufficiently general. Furthermore, there was one instance where
5650 it was actually wrong because one of the argument was used twice, and in
5651 one call a function was used. (As it happened, calling the function twice
5652 did not affect the overall behaviour.) The macro has been replaced by a
5653 function that can join an arbitrary number of extra strings onto a growing
5656 (c) The code for expansion conditions now uses a table and a binary chop
5657 instead of a serial search (which was left over from when there were very
5658 few conditions). Also, it now recognizes conditions like "pam" even when
5659 the relevant support is not compiled in: a suitably worded error message is
5660 given if an attempt is made to use such a condition.
5662 11. Added ${time_interval:xxxxx}.
5664 12. A bug was causing one of the ddress fields not to be passed back correctly
5665 from remote delivery subprocesses. The field in question was not being
5666 subsequently used, so this caused to problems in practice.
5668 13. Added new log selectors queue_time and deliver_time.
5670 14. Might have fixed a bug in maildirsizefile handling that threw up
5671 "unexpected character" debug warnings, and recalculated the data
5672 unnecessarily. In any case, I expanded the warning message to give more
5675 15. Added the message "Restricted characters in address" to the statements in
5676 the default ACL that block characters like @ and % in local parts.
5678 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5679 Three changes have been made:
5681 (a) There was a serious bug; a negative response to MAIL caused the whole
5682 recipient domain to be cached as invalid, thereby blocking all messages
5683 to all local parts at the same domain, from all senders. This bug has
5684 been fixed. The domain is no longer cached after a negative response to
5685 MAIL if the sender used is not empty.
5687 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5690 (c) A new callout option, "use_sender" has been added for people who want
5691 the modified behaviour.
5697 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5700 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5701 indeed breaks things for older releases.
5703 3. Added additional logging to the case where there is a problem reading data
5704 from a filter that is running in a subprocess using a pipe, in order to
5705 try to track down a specific problem.
5707 4. Testing facility fudge: when running in the test harness and attempting
5708 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5709 getting "No route to host". Convert this to a timeout.
5711 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5714 6. Some OS don't have socklen_t but use size_t instead. This affects the
5715 fifth argument of getsockopt() amongst other things. This is now
5716 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5717 can be set for individual OS. I have set it for SunOS5, OSF1, and
5718 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5719 some earlier ones do not.
5721 7. Change 4.30/15 was not doing the test caselessly.
5723 8. The standard form for an IPv6 address literal was being rejected by address
5724 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5725 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5726 this, as well as the form without the "IPv6" on the front (but only when
5727 address literals are enabled, of course).
5729 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5731 10. Exim crashed if a message with an empty sender address specified by -f
5732 encountered a router with an errors_to setting. This could be provoked only
5733 by a command such as
5737 where an empty string was supplied; "<>" did not hit this bug.
5739 11. Installed PCRE release 4.5.
5741 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5742 remained set. It is now erased.
5744 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5745 times from message ids (which are base 36 rather than the normal 62).
5747 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5748 were being counted as actual protocol errors, and logged if the log
5749 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5750 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5751 and DATA following a set of rejected RCPTs do not count as protocol errors.
5752 In other words, Exim assumes they were pipelined, though this may not
5753 actually be the case. Of course, in all cases the client gets an
5754 appropriate error code.
5756 15. If a lookup fails in an ACL condition, a message about the failure may
5757 be available; it is used if testing the ACL cannot continue, because most
5758 such messages specify what the cause of the deferral is. However, some
5759 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5760 that caused an old message to be retained and used if a later statement
5761 caused a defer, replacing the real cause of the deferral.
5763 16. If an IP address had so many PTR records that the DNS lookup buffer
5764 was not large enough to hold them, Exim could crash while trying to process
5765 the truncated data. It now detects and logs this case.
5767 17. Further to 4.21/58, another change has been made: if (and only if) the
5768 first line of a message (the first header line) ends with CRLF, a bare LF
5769 in a subsequent header line has a space inserted after it, so as not to
5770 terminate the header.
5772 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5773 unnecessarily, used atoi() instead of strtol(), and didn't check the
5774 termination when getting file sizes from file names by regex.
5776 19. Completely re-implemented the support for maildirsize files, in the light
5777 of a number of problems with the previous contributed implementation
5778 (4.30/29). In particular:
5780 . If the quota is zero, the maildirsize file is maintained, but no quota is
5783 . If the maildir directory does not exist, it is created before any attempt
5784 to write a maildirsize file.
5786 . The quota value in the file is just a cache; if the quota is changed in
5787 the transport, the new value overrides.
5789 . A regular expression is available for excluding directories from the
5792 20. The autoreply transport checks the characters in options that define the
5793 message's headers; it allows continued headers, but it was checking with
5794 isspace() after an embedded newline instead of explicitly looking for a
5797 21. If all the "regular" hosts to which an address was routed had passed their
5798 expiry times, and had not reached their retry times, the address was
5799 bounced, even if fallback hosts were defined. Now Exim should go on to try
5802 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5803 equivalent code in the SMTP transport. Some hosts send humungous responses
5804 to HELO/EHLO, more than 1024 it seems.
5806 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5807 gives compiler warnings in some environments. I've now done it "properly",
5810 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5811 (because of problems with the built-in one) was declared to return uschar *
5812 instead of char *, causing compiler failure.
5814 25. Fixed a file descriptor leak when processing alias/forward files.
5816 26. Fixed a minor format string issue in dbfn.c.
5818 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5820 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5821 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5824 29. When a sender address is verified, it is cached, to save repeating the test
5825 when there is more than one recipient in a message. However, when the
5826 verification involves a callout, it is possible for different callout
5827 options to be set for different recipients. It is too complicated to keep
5828 track of this in the cache, so now Exim always runs a verification when a
5829 callout is required, relying on the callout cache for the optimization.
5830 The overhead is duplication of the address routing, but this should not be
5833 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5834 to be verified with callout=postmaster, and the main callout worked but the
5835 postmaster check failed, the verification correctly failed. However, if a
5836 subsequent RCPT command asked for sender verification *without* the
5837 postmaster check, incorrect caching caused this verification also to fail,
5840 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5841 it was not caching the DNS options (qualify_single, search_parents) that
5842 were used when the lookup failed. A subsequent lookup with different
5843 options therefore always gave the same answer, though there were cases
5844 where it should not have. (Example: a "domains = !$mx_any" option on a
5845 dnslookup router: the "domains" option is always processed without any
5846 widening, but the router might have qualify_single set.) Now Exim uses the
5847 cached value only when the same options are set.
5849 32. Added John Jetmore's "exipick" utility to the distribution.
5851 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5852 than a timeout (e.g. a certificate is required, and is not provided), an
5853 Exim server now closes the connection immediately. Previously it waited for
5854 the client to close - but if the client is SSL, it seems that they each
5855 wait for each other, leading to a delay before one of them times out.
5857 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5858 maintained 0.8.x compatibility because I don't think many are using it, and
5859 it is clearly obsolete.
5861 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5864 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5865 client certificate was expired. A simple patch fixes this, though I don't
5866 understand the full logic of why the verify callback is called multiple
5869 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5870 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5871 which causes problems with some clients (such as the Certicom SSL Plus
5872 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5873 disables the coutermeasure allowing Eudora to connect."
5875 38. Exim was not checking that a write() to a log file succeeded. This could
5876 lead to Bad Things if a log got too big, in particular if it hit a file
5877 size limit. Exim now panics and dies if it cannot write to a log file, just
5878 as it does if it cannot open a log file.
5880 39. Modified OS/Makefile-Linux so that it now contains
5882 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5884 The two -D definitions ensure that Exim is compiled with large file
5885 support, which makes it possible to handle log files that are bigger than
5888 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5889 instance) a domain was checked against a named list that involved a lookup,
5890 causing $domain_data to be set, then another domain was checked against the
5891 same list, then the first domain was re-checked, the value of $domain_data
5892 after the final check could be wrong. In particular, if the second check
5893 failed, it could be set empty. This bug probably also applied to
5896 41. The strip_trailing_dot option was not being applied to the address given
5897 with the -f command-line option.
5899 42. The code for reading a message's header from the spool was incrementing
5900 $received_count, but never initializing it. This meant that the value was
5901 incorrect (doubled) while delivering a message in the same process in which
5902 it was received. In the most common configuration of Exim, this never
5903 happens - a fresh exec is done - but it can happen when
5904 deliver_drop_privilege is set.
5906 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5907 it now includes up to 150 characters of the unexpected data in the log
5910 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5911 and building data strings. The size of both of these buffers was 10 000
5912 bytes - far larger than anybody would *ever* want, thought I. Needless to
5913 say, somebody hit the limit. I have increased the maximum line length to
5914 20 000 and the maximum data length of concatenated lines to 100 000. I have
5915 also fixed two bugs, because there was no checking on these buffers. Tsk,
5916 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5917 buffer is too small.
5919 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5920 lsearch lookups. Now it does.
5922 46. When parsing a route_list item in a manualroute router, a fixed-length
5923 buffer was used for the list of hosts. I made this 1024 bytes long,
5924 thinking that nobody would ever have a list of hosts that long. Wrong.
5925 Somebody had a whole pile of complicated expansion conditions, and the
5926 string was silently truncated, leading to an expansion error. It turns out
5927 that it is easier to change to an unlimited length (owing to other changes
5928 that have happened since this code was originally written) than to build
5929 structure for giving a limitation error. The length of the item that
5930 expands into the list of hosts is now unlimited.
5932 47. The lsearch lookup could not handle data where the length of text line was
5933 more than 4095 characters. Such lines were truncated, leading to shortened
5934 data being returned. It should now handle lines of any length.
5936 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5937 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5940 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5941 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5942 no longer happens in comment lines. A list of replacements is now placed
5943 at the head of all of the source files, except those whose only change is
5944 to replace PERL_COMMAND in the very first #! line.
5946 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5947 messages on the queue, with a bottom-up merge sort, using code contributed
5948 by Michael Haardt. This should make operations like -bp somewhat faster on
5949 large queues. It won't affect queue runners, except when queue_run_in_order
5952 51. Installed eximstats 1.31 in the distribution.
5954 52. Added support for SRV lookups to the dnslookup router.
5956 53. If an ACL referred to $message_body or $message_body_end, the value was not
5957 reset for any messages that followed in the same SMTP session.
5959 54. The store-handling optimization for building very long strings was not
5960 differentiating between the different store pools. I don't think this
5961 actually made any difference in practice, but I've tidied it.
5963 55. While running the routers to verify a sender address, $sender_address
5964 was still set to the sender address. This is wrong, because when routing to
5965 send a bounce to the sender, it would be empty. Therefore, I have changed
5966 it so that, while verifying a sender address, $sender_address is set to <>.
5967 (There is no change to what happens when verifying a recipient address.)
5969 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5970 target A or AAAA records (if not already returned) without resetting the
5971 qualify_single or search_parents options of the DNS resolver. These are
5972 inappropriate in this case because the targets of MX and SRV records must
5973 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5974 target that, when qualified, matched something in the local domain. These
5975 two options are now turned off when doing these lookups.
5977 57. It seems that at least some releases of Reiserfs (which does not have the
5978 concept of a fixed number of inodes) returns zero and not -1 for the
5979 number of available inodes. This interacted badly with check_spool_inodes,
5980 which assumed that -1 was the "no such thing" setting. What I have done is
5981 to check that the total number of inodes is greater than zero before doing
5982 the test of how many are available.
5984 58. When a "warn" ACL statement has a log_message modifier, the message is
5985 remembered, and not repeated. This is to avoid a lot of repetition when a
5986 message has many recipients that cause the same warning to be written.
5987 However, Exim was preserving the list of already written lines for an
5988 entire SMTP session, which doesn't seem right. The memory is now reset if a
5989 new message is started.
5991 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5992 debugging output unless log_rewrite was also set.
5994 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5995 of (int)(handle) when we know that handle contains (void *)(-1).
5997 61. The Exim daemon panic-logs an error return when it closes the incoming
5998 connection. However "connection reset by peer" seems to be common, and
5999 isn't really an error worthy of noting specially, so that particular error
6002 62. When Exim is trying to find all the local interfaces, it used to panic and
6003 die if the ioctl to get the interface flags failed. However, it seems that
6004 on at least one OS (Solaris 9) it is possible to have an interface that is
6005 included in the list of interfaces, but for which you get a failure error
6006 for this call. This happens when the interface is not "plumbed" into a
6007 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6008 failure of the "get flags" call assumes that the interface is down.
6010 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6011 makes life easier for people who are doing arithmetic on fields extracted
6012 from dates, where you often get leading zeros that should not be
6013 interpreted as octal.
6015 64. Added qualify_domain to the redirect router, to override the global
6018 65. If a pathologically long header line contained very many addresses (the
6019 report of this problem mentioned 10 000) and each of them was rewritten,
6020 Exim could use up a very large amount of memory. (It kept on making new
6021 copies of the header line as it rewrote, and never released the old ones.)
6022 At the expense of a bit more processing, the header rewriting function has
6023 been changed so that it no longer eats memory in this way.
6025 66. The generation of the Received: header has been moved from the time that a
6026 message starts to be received, to the time that it finishes. The timestamp
6027 in the Received: header should now be very close to that of the <= log
6028 line. There are two side-effects of this change:
6030 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6031 logged header lines no longer include the local Received: line, because
6032 it has not yet been created. The same applies to a copy of the message
6033 that is returned to a non-SMTP sender when a message is rejected.
6035 (b) When a filter file is tested using -bf, no additional Received: header
6036 is added to the test message. After some thought, I decided that this
6039 This change does not affect the value of $received_for. It is still set
6040 after address rewriting, but before local_scan() is called.
6042 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6044 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6045 gave an unhelpful panic error message, and a defer error. I have managed to
6046 change this behaviour so that it now rejects any supplied certificate,
6047 which seems right, as the list of acceptable certificates is empty.
6049 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6050 gave an unhelpful defer error. I have not managed to make this reject any
6051 supplied certificates, but the error message it gives is "no certificate
6052 supplied", which is not helpful.
6054 70. exigrep's output now also includes lines that are not associated with any
6055 message, but which match the given pattern. Implemented by a patch from
6056 Martin Sluka, which also tidied up the Perl a bit.
6058 71. Recipient callout verification, like sender verification, was using <> in
6059 the MAIL FROM command. This isn't really the right thing, since the actual
6060 sender may affect whether the remote host accepts the recipient or not. I
6061 have changed it to use the actual sender in the callout; this means that
6062 the cache record is now keyed on a recipient/sender pair, not just the
6063 recipient address. There doesn't seem to be a real danger of callout loops,
6064 since a callout by the remote host to check the sender would use <>.
6065 [SEE ABOVE: changed after hitting problems.]
6067 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6068 temporary errors. However, in the case of such a code being given after
6069 the end of a data transmission (i.e. after ".") Exim was failing to write
6070 a retry record for the message. (Yes, there was some broken host that was
6071 actually sending 8xx at this point.)
6073 73. An unknown lookup type in a host list could cause Exim to panic-die when
6074 the list was checked. (An example that provoked this was putting <; in the
6075 middle of a list instead of at the start.) If this happened during a DATA
6076 ACL check, a -D file could be left lying around. This kind of configuration
6077 error no longer causes Exim to die; instead it causes a defer error. The
6078 incident is still logged to the main and panic logs.
6080 74. Buglet left over from Exim 3 conversion. The message "too many messages
6081 in one connection" was written to the rejectlog but not the mainlog, except
6082 when address rewriting (yes!) was being logged.
6084 75. Added write_rejectlog option.
6086 76. When a system filter was run not as root (that is, when system_filter_user
6087 was set), the values of the $n variables were not being returned to the
6088 main process; thus, they were not subsequently available in the $sn
6091 77. Added +return_path_on_delivery log selector.
6093 78. A connection timeout was being treated differently from recipients deferred
6094 when testing hosts_max_try with a message that was older than the host's
6095 retry timeout. (The host should not be counted, thus allowing all hosts to
6096 be tried at least once before bouncing.) This may have been the cause of an
6097 occasionally reported bug whereby a message would remain on the queue
6098 longer than the retry timeout, but would be bounced if a delivery was
6099 forced. I say "may" because I never totally pinned down the problem;
6100 setting up timeout/retry tests is difficult. See also the next item.
6102 79. The ultimate address timeout was not being applied to errors that involved
6103 a combination of host plus message (for example, a timeout on a MAIL
6104 command). When an address resolved to a number of possible hosts, and they
6105 were not all tried for each delivery (e.g. because of hosts_max_try), a
6106 message could remain on the queue longer than the retry timeout.
6108 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6111 81. Fixed an obscure SMTP outgoing bug which required at least the following
6112 conditions: (a) there was another message waiting for the same server;
6113 (b) the server returned 5xx to all RCPT commands in the first message so
6114 that the message was not completed; (c) the server dropped the connection
6115 or gave a negative response to the RSET that Exim sends to abort the
6116 transaction. The observed case was a dropped connection after DATA that had
6117 been sent in pipelining mode. That is, the server had advertised PIPELINING
6118 but was not implementing it correctly. The effect of the bug was incorrect
6119 behaviour, such as trying another host, and this could lead to a crash.
6125 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6126 and daemon.c were passed as pointers to ints; they should have been
6127 pointers to socklen_t variables (which are typically unsigned ints).
6129 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6132 3. Fixed a really odd bug that affected only the testing scheme; patching a
6133 certain fixed string in the binary changed the value of another string that
6134 happened to be identical to the end of the original first string.
6136 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6137 name", it returns that address as the IP address. On some operating
6138 systems (e.g. Solaris), it also passes back the IP address string as the
6139 "host name". However, on others (e.g. Linux), it passes back an empty
6140 string. Exim wasn't checking for this, and was changing the host name to an
6141 empty string, assuming it had been canonicalized.
6143 5. Although rare, it is permitted to have more than one PTR record for a given
6144 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6145 all the names associated with an address, because they do in Solaris.
6146 However, it seems that they do not in Linux for data that comes from the
6147 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6148 I found this out when I moved to a new Linux workstation and tried to run
6149 the Exim test suite.
6151 To get round this problem I have changed the code so that it now does its
6152 own call to the DNS to look up PTR records when searching for a host name.
6153 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6154 addresses that are only in /etc/hosts are still found.
6156 This behaviour is, however, controlled by an option called host_lookup_
6157 order, which defaults to "bydns:byaddr". If people want to use the other
6158 order, or indeed, just use one or the other means of lookup, they can
6159 specify it in this variable.
6161 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6162 some operating systems, this comes back from gethostbyaddr() as an empty
6163 string, and this is what Exim used to test for. However, it seems that in
6164 other systems, "." is yielded. Exim now tests for this case too.
6166 7. The values of check_spool_space and check_log_space are now held internally
6167 as a number of kilobytes instead of an absolute number of bytes. If a
6168 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6169 kilobyte. This means that much larger values can be stored.
6171 8. Exim monitor: an attempt to get the action menu when not actually pointing
6172 at a message produces an empty menu entitled "No message selected". This
6173 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6174 no entries in it ("Shell widget menu has zero width and/or height"). So I
6175 have added a single, blank menu entry in this case.
6177 9. Added ${quote_local_part.
6179 10. MIME decoding is now applied to the contents of Subject: header lines when
6182 11. Now that a reference to $sender_host_address automatically causes a reverse
6183 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6184 host lookup before query-style lookups in lists that might use this
6185 variable. This has therefore been abolished, and the "net-" prefix is no
6186 longer necessary for query-style lookups.
6188 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6189 have been a typo for LFLAGS, so it has been changed.
6191 13. The install script calls Exim with "-C /dev/null" in order to find the
6192 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6193 to be output. However, since Exim outputs its version number before the
6194 error, it didn't break the script. It just looked ugly. I fixed this by
6195 always allowing "-C /dev/null" if the caller is root.
6197 14. Ignore overlarge ACL variable number when reading spool file - insurance
6198 against a later release with more variables having written the file.
6200 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6201 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6202 this, as well as the form without the "IPv6" on the front.
6204 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6205 OS/Makefile-Darwin file.
6207 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6208 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6210 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6211 variable. However, some OS use ints for inodes. I've added cast to long int
6212 to get rid of the compiler warning.
6214 19. I had forgotten to lock out "/../" in configuration file names when
6215 ALT_CONFIG_PREFIX was set.
6217 20. Routers used for verification do not need to specify transports. However,
6218 if such a router generated a host list, and callout was configured, Exim
6219 crashed, because it could not find a port number from the (non-existent)
6220 transport. It now assumes port 25 in this circumstance.
6222 21. Added the -t option to exigrep.
6224 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6225 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6226 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6227 them is set, LOOKUP_LSEARCH is forced.
6229 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6231 24. Added sender and host information to the "rejected by local_scan()" log
6232 line; previously there was no indication of these.
6234 25. Added .include_if_exists.
6236 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6237 Linux. It turns out that not all file systems support this. Apparently some
6238 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6239 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6240 not supported on the file descriptor, is now ignored when Exim is trying to
6241 sync a directory. This applies only to Linux.
6243 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6245 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6246 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6249 29. Added support for maildirsize files from supplied patch (modified a bit).
6251 30. The use of :fail: followed by an empty string could lead Exim to respond to
6252 sender verification failures with (e.g.):
6254 550 Verification failed for <xxx>
6255 550 Sender verify failed
6257 where the first response line was missing the '-' that indicates it is not
6258 the final line of the response.
6260 31. The loop for finding the name of the user that called Exim had a hardwired
6261 limit of 10; it now uses the value of finduser_retries, which is used for
6262 all other user lookups.
6264 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6267 33. Exim was neglecting to zero errno before one call of strtol() when
6268 expanding a string and expecting an integer value. On some systems this
6269 resulted in spurious "integer overflow" errors. Also, it was casting the
6270 result into an int without checking.
6272 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6273 did not work. The code looks as if it has *never* worked, though it appears
6274 to have been documented since at least release 1.62. I have made it work.
6276 35. The "timeout_DNS" error in retry rules, also documented since at least
6277 1.62, also never worked. As it isn't clear exactly what this means, and
6278 clearly it isn't a major issue, I have abolished the feature by treating it
6279 as "timeout", and writing a warning to the main and panic logs.
6281 36. The display of retry rules for -brt wasn't always showing the error code
6284 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6285 timeout_connect_A, timeout_connect_MX.
6287 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6288 to the empty sender.
6290 39. The daemon was not analysing the content of -oX till after it had closed
6291 stderr and disconnected from the controlling terminal. This meant that any
6292 syntax errors were only noted on the panic log, and the return code from
6293 the command was 0. By re-arranging the code a little, I've made the
6294 decoding happen first, so such errors now appear on stderr, and the return
6295 code is 1. However, the actual setting up of the sockets still happens in
6296 the disconnected process, so errors there are still only recorded on the
6299 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6300 connections (as happens on some IP stacks) was logged at start up time as
6301 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6302 it from "IPv6 and IPv4", which means that two separate sockets are being
6305 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6306 says whether AF_INET or AF_INET6 was passed as an argument.
6308 42. Exiwhat output was messed up when time zones were included in log
6311 43. Exiwhat now gives more information about the daemon's listening ports,
6312 and whether -tls-on-connect was used.
6314 44. The "port" option of the smtp transport is now expanded.
6316 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6317 silently ignored. Now an error message is written to the main and panic
6320 46. There's a new ACL modifier called "logwrite" which writes to a log file
6321 as soon as it is encountered.
6323 47. Added $local_user_uid and $local_user_gid at routing time.
6325 48. Exim crashed when trying to verify a sender address that was being
6328 49. Exim was recognizing only a space character after ".include". It now also
6329 recognizes a tab character.
6331 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6332 extracting the relevant information from the specification. The man page no
6333 longer contains scrambled data for the -d option, and I've added a section
6334 at the front about calling Exim under different names.
6336 51. Added "extra_headers" argument to the "mail" command in filter files.
6338 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6341 53. Installed eximstats 1.29.
6343 54. Added transport_filter_timeout as a generic transport option.
6345 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6346 Cc: header lines. This was required by RFC 822, but it not required by RFC
6349 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6350 incoming messages that did not have them. Now it does so only if the
6351 message originates locally, that is, if there is no associated remote host
6352 address. When Resent- header lines are present, this applies to the Resent-
6353 lines rather than the non-Resent- lines.
6355 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6356 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6358 58. Messages for configuration errors now include the name of the main
6359 configuration file - useful now that there may be more than one file in a
6360 list (.included file names were always shown).
6362 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6363 for those rare installations that do not start the daemon as root or run it
6364 setuid root. I've cut out the call to initgroups() if the daemon is not
6367 60. The Exim user and group can now be bound into the binary as text strings
6368 that are looked up at the start of Exim's processing.
6370 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6372 62. Added $mailstore_basename variable.
6374 63. Installed patch to sieve.c from Michael Haardt.
6376 64. When Exim failed to open the panic log after failing to open the main log,
6377 the original message it was trying to log was written to stderr and debug
6378 output, but if they were not available (the usual case in production), it
6379 was lost. Now it is written to syslog before the two lines that record the
6380 failures to open the logs.
6382 65. Users' Exim filters run in subprocesses under the user's uid. It is
6383 possible for a "deliver" command or an alias in a "personal" command to
6384 provoke an address rewrite. If logging of address rewriting is configured,
6385 this fails because the process is not running as root or exim. There may be
6386 a better way of dealing with this, but for the moment (because 4.30 needs
6387 to be released), I have disabled address rewrite logging when running a
6388 filter in a non-root, non-exim process.
6394 1. The buildconfig auxiliary program wasn't quoting the value set for
6395 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6396 not defined. This bug was masked in 4.22 by the effect that was fixed in
6399 2. Some messages that were rejected after a message id was allocated were
6400 shown as "incomplete" by exigrep. It no longer does this for messages that
6401 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6403 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6404 have allow_domain_literals set, the ID did not get logged in the <= line.
6405 Domain literals are now always recognized in Message-ID: header lines.
6407 4. The first argument for a ${extract expansion item is the key name or field
6408 number. Leading and trailing spaces in this item were not being ignored,
6409 causing some misleading effects.
6411 5. When deliver_drop_privilege was set, single queue runner processes started
6412 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6413 same command in the process it spins off) were not dropping privilege.
6415 6. When the daemon running as "exim" started a queue runner, it always
6416 re-executed Exim in the spun-off process. This is a waste of effort when
6417 deliver_drop_privilege is set. The new process now just calls the
6418 queue-runner function directly.
6424 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6427 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6428 ignored. Though the use of -oP was forcing the writing of a pid file, it
6429 was always written to the default place.
6431 3. If the message "no IP address found for host xxxx" is generated during
6432 incoming verification, it is now followed by identification of the incoming
6433 connection (so you can more easily find what provoked it).
6435 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6437 5. Added some features to "harden" Exim a bit more against certain attacks:
6439 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6440 be put in Local/Makefile. This is like the never_users runtime option,
6441 but it cannot be overridden. The default setting is "root".
6443 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6444 prefix string with which any file named in a -C command line option
6447 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6448 is retained for -C and -D only if the caller of Exim is root. Without
6449 it, the exim user may also use -C and -D and retain privilege.
6451 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6452 command line option is disabled.
6454 6. Macro names set by the -D option must start with an upper case letter, just
6455 like macro names defined in the configuration file.
6457 7. Added "dereference=" facility to LDAP.
6459 8. Two instances of the typo "uknown" in the source files are fixed.
6461 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6462 the Configure-Makefile script screwed up while processing it.
6464 10. Incorporated PCRE 4.4.
6466 11. The SMTP synchronization check was not operating right at the start of an
6467 SMTP session. For example, it could not catch a HELO sent before the client
6468 waited for the greeting. There is now a check for outstanding input at the
6469 point when the greeting is written. Because of the duplex, asynchronous
6470 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6471 way, but not yet received, when the check is performed.
6473 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6474 on TCP/IP sockets, because this apparently causes some broken clients to
6477 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6478 unchanged) from the Cygwin maintainer.
6480 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6481 was supported instead of testing for mbx. Effectively a typo.
6483 15. The spa authenticator server code was not checking that the input it
6484 received was valid base64.
6486 16. The debug output line for the "set" modifier in ACLs was not showing the
6487 name of the variable that was being set.
6489 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6491 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6492 lookup on its own. Something else was needed to trigger the lookup. For
6493 example, a match in host_lookup or the need for a host name in a host list.
6494 Now, if $sender_host_name is referenced and the host name has not yet been
6495 looked up, a lookup is performed. If the lookup fails, the variable remains
6496 empty, and $host_lookup_failed is set to "1".
6498 19. Added "eqi" as a case-independent comparison operator.
6500 20. The saslauthd authentication condition could segfault if neither service
6501 nor realm was specified.
6503 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6504 error message that was logged was misleading, and incoming SMTP
6505 connections were dropped. The message is now more accurate, and temporary
6506 errors are given to SMTP connections.
6508 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6509 (except RSET) with a 421 error, until QUIT is received. However, it was
6510 failing to send a response to QUIT.
6512 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6513 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6514 ACL is now run after the helo_try_verify_hosts code.
6516 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6517 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6518 case-independent, so other case variants are also recognized). Apparently
6519 some systems use these upper case variants.
6521 25. If more than two messages were waiting for the same host, and a transport
6522 filter was specified for the transport, Exim sent two messages over the
6523 same TCP/IP connection, and then failed with "socket operation on non-
6524 socket" when it tried to send the third.
6526 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6528 27. The extern definition of crypt16() in expand.c was not being excluded when
6529 the OS had its own crypt16() function.
6531 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6532 as a preferred synonym for return_size_limit, both as an option and as an
6535 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6537 30. Changed the default configuration ACL to relax the local part checking rule
6538 for addresses that are not in any local domains. For these addresses,
6539 slashes and pipe symbols are allowed within local parts, but the sequence
6540 /../ is explicitly forbidden.
6542 31. SPA server authentication was not clearing the challenge buffer before
6545 32. log_message in a "warn" ACL statement was writing to the reject log as
6546 well as to the main log, which contradicts the documentation and doesn't
6547 seem right (because no rejection is happening). So I have stopped it.
6549 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6550 However, I am unable to do any testing of this.
6552 34. Fixed an infelicity in the appendfile transport. When checking directories
6553 for a mailbox, to see if any needed to be created, it was accidentally
6554 using path names with one or more superfluous leading slashes; tracing
6555 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6557 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6558 contents are added to the log line that is written for every discarded
6559 recipient. (Previously a log_message setting was ignored.)
6561 36. The ${quote: operator now quotes the string if it is empty.
6563 37. The install script runs exim in order to find its version number. If for
6564 some reason other than non-existence or emptiness, which it checks, it
6565 could not run './exim', it was installing it with an empty version number,
6566 i.e. as "exim-". This error state is now caught, and the installation is
6569 38. An argument was missing from the function that creates an error message
6570 when Exim fails to connect to the socket for saslauthd authentication.
6571 This could cause Exim to crash, or give a corrupted message.
6573 39. Added isip, isip4, and isip6 to ${if conditions.
6575 40. The ACL variables $acl_xx are now saved with the message, and can be
6576 accessed later in routers, transports, and filters.
6578 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6579 strings in the file are not string-expanded.
6581 42. If a MAIL command specified a SIZE value that was too large to fit into an
6582 int variable, the check against message_size_limit failed. Such values are
6583 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6584 day this will have to be increased, but I don't think I want to be around
6585 when emails are that large.
6592 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6593 iconv() is not standard in FreeBSD.
6595 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6596 IPv6 enabled. The observed symptom was a segmentation fault on return from
6597 the function os_common_find_running_interfaces() in src/os.c.
6599 3. In the check_special_case() function in daemon.c I had used "errno" as an
6600 argument name, which causes warnings on some systems. This was basically a
6601 typo, since it was named "eno" in the comments!
6603 4. The code that waits for the clock to tick (at a resolution of some fraction
6604 of a second) so as to ensure message-id uniqueness was always waiting for
6605 at least one whole tick, when it could have waited for less. [This is
6606 almost certainly not relevant at current processor speeds, where it is
6607 unlikely to ever wait at all. But we try to future-proof.]
6609 5. The function that sleeps for a time interval that includes fractions of a
6610 second contained a race. It did not block SIGALRM between setting the
6611 timer, and suspending (a couple of lines later). If the interval was short
6612 and the sigsuspend() was delayed until after it had expired, the suspension
6613 never ended. On busy systems this could lead to processes getting stuck for
6616 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6617 process, before it forks any delivery processes. The open lookup caching
6618 mechanism meant that the open file or database connection was passed into
6619 the delivery process. The problem was that delivery processes always tidy
6620 up cached lookup data. This could cause a problem for the next delivery
6621 process started by the queue runner, because the external queue runner
6622 process does not know about the closure. So the next delivery process
6623 still has data in the lookup cache. In the case of a file lookup, there was
6624 no problem because closing a file descriptor in a subprocess doesn't affect
6625 the parent. However, if the lookup was caching a connection to a database,
6626 the connection was closed, and the second delivery process was likely to
6627 see errors such as "PGSQL: query failed: server closed the connection
6628 unexpectedly". The problem has been fixed by closing all cached lookups
6629 in a queue runner before running a delivery process.
6631 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6632 seem to have the "const" qualifier which it has on other OS. I've
6635 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6636 *of the same type* (client or server) with the same public name that an
6637 error should be diagnosed.
6639 9. When Exim looked up a host name for an IP address, but failed to find the
6640 original IP address when looking up the host name (a safety check), it
6641 output the message "<ip address> does not match any IP for NULL", which was
6642 confusing, to say the least. The bug was that the host name should have
6643 appeared instead of "NULL".
6645 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6646 user that is built into the binary, and the -C or -D options is used, root
6647 privilege is dropped before the configuration file is read. In addition,
6648 logging is switched to stderr instead of the normal log files. If the
6649 configuration then re-defines the Exim user, the unprivileged environment
6650 is probably not what is expected, so Exim logs a panic warning message (but
6653 However, if deliver_drop_privilege is set, the unprivileged state may well
6654 be exactly what is intended, so the warning has been cut out in that case,
6655 and Exim is allowed to try to write to its normal log files.
6661 1. smtp_return_error_details was not giving details for temporary sender
6662 or receiver verification errors.
6664 2. Diagnose a configuration error if two authenticators have the same public
6667 3. Exim used not to create the message log file for a message until the first
6668 delivery attempt. This could be confusing when incoming messages were held
6669 for policy or load reasons. The message log file is now created at the time
6670 the message is received, and an initial "Received" line is written to it.
6672 4. The automatically generated man page for command line options had a minor
6673 bug that caused no ill effects; however, a more serious problem was that
6674 the procedure for building the man page automatically didn't always
6675 operate. Consequently, release 4.20 contains an out-of-date version. This
6676 shouldn't happen again.
6678 5. When building Exim with embedded Perl support, the script that builds the
6679 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6680 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6682 6. The freeze_tell option was not being used for messages that were frozen on
6683 arrival, either by an ACL or by local_scan().
6685 7. Added the smtp_incomplete_transaction log selector.
6687 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6688 was accepting AUTH without a new EHLO.
6690 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6691 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6694 10. Logging of TCP/IP connections (when configured) now happens in the main
6695 daemon process instead of the child process, so that the TCP/IP connection
6696 count is more accurate (but it can never be perfect).
6698 11. The use of "drop" in a nested ACL was not being handled correctly in the
6699 outer ACL. Now, if condition failure induced by the nested "drop" causes
6700 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6701 or "require"), the connection is dropped.
6703 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6704 that yield "discard" can now be used with an "accept" or a "discard" verb,
6705 but an error is generated for any others (because I can't see a useful way
6706 to define what should happen).
6708 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6709 are now processed in the same way as lines in the Exim configuration file.
6710 In particular, continuation lines are supported.
6712 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6714 15. Added -ti meaning -t -i.
6716 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6717 domains, and warn by logging if others are found.
6719 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6720 returned from GIFCONF when Exim is trying to find the list of interfaces on
6721 a host. The code in os.c has been modified to copy each ifreq to an aligned
6722 structure in all cases.
6724 Also, in some cases, the returned ifreq's were being copied to a 'struct
6725 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6726 means the last couple of bytes of an IPv6 address could be chopped if the
6727 ifreq contained only a normal sockaddr (14 bytes storage).
6729 18. Named domain lists were not supported in the hosts_treat_as_local option.
6730 An entry such as +xxxx was not recognized, and was treated as a literal
6733 19. Ensure that header lines added by a DATA ACL are included in the reject log
6734 if the ACL subsequently rejects the message.
6736 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6737 MD5 (which is deprecated).
6739 21. When testing a filter file using -bf, Exim was writing a message when it
6740 took the sender from a "From " line in the message, but it was not doing so
6741 when it took $return_path from a Return-Path: header line. It now does.
6743 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6744 with a valid header line field name (a series of printing characters
6745 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6747 23. Changed "disc" in the source to "disk" to conform to the documentation and
6748 the book and for uniformity.
6750 24. Ignore Sendmail's -Ooption=value command line item.
6752 25. When execve() failed while trying to run a command in a pipe transport,
6753 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6754 could be confused with a return value of 69 from the command itself. This
6755 has been changed to 127, the value the shell returns if it is asked to run
6756 a non-existent command. The wording for the related log line suggests a
6757 non-existent command as the problem.
6759 26. If received_header_text expands to an empty string, do not add a Received:
6760 header line to the message. (Well, it adds a token one on the spool, but
6761 marks it "old" so that it doesn't get used or transmitted.)
6763 27. Installed eximstats 1.28 (addition of -nt option).
6765 28. There was no check for failure on the call to getsockname() in the daemon
6766 code. This can fail if there is a shortage of resources on the system, with
6767 ENOMEM, for example. A temporary error is now given on failure.
6769 29. Contrary to the C standard, it seems that in some environments, the
6770 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6771 program. Exim now does this explicitly; it affects the formatting of
6772 timestamps using strftime().
6774 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6775 complaints. I've now initialized all the variables, to avoid this.
6777 32. Header lines added by a system filter were not being "seen" during
6778 transport-time rewrites.
6780 33. The info_callback() function passed to OpenSSL is set up with type void
6781 (*)(SSL *, int, int), as described somewhere. However, when calling the
6782 function (actually a macro) that sets it up, the type void(*)() is
6783 expected. I've put in a cast to prevent warnings from picky compilers.
6785 34. If a DNS black list lookup found a CNAME record, but there were no A
6786 records associated with the domain it pointed at, Exim crashed.
6788 35. If a DNS black list lookup returned more than one A record, Exim ignored
6789 all but the first. It now scans all returned addresses if a particular IP
6790 value is being sought. In this situation, the contents of the
6791 $dnslist_value variable are a list of all the addresses, separated by a
6794 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6795 to accept a host name and all its aliases if the forward lookup for any of
6796 them yielded the IP address of the incoming connection. Now it accepts only
6797 those names whose forward lookup yields the correct IP address. Any other
6798 names are discarded. This closes a loophole whereby a rogue DNS
6799 administrator could create reverse DNS records to break through a
6800 wildcarded host restriction in an ACL.
6802 37. If a user filter or a system filter that ran in a subprocess used any of
6803 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6804 the wrong values were passed to the pipe command ($thisaddress had the
6805 value of $0, $0 had the value of $1, etc). This bug was introduced by
6806 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6808 38. Improved the line breaking for long SMTP error messages from ACLs.
6809 Previously, if there was no break point between 40 and 75 characters, Exim
6810 left the rest of the message alone. Two changes have been made: (a) I've
6811 reduced the minimum length to 35 characters; (b) if it can't find a break
6812 point between 35 and 75 characters, it looks ahead and uses the first one
6813 that it finds. This may give the occasional overlong line, but at least the
6814 remaining text gets split now.
6816 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6817 file descriptors might be low, and that setting 1000 would always raise it.
6818 It turns out that in some environments, the limit is already over 1000 and
6819 that lowering it causes trouble. So now Exim takes care not to decrease it.
6821 40. When delivering a message, the value of $return_path is set to $sender_
6822 address at the start of routing (routers may change the value). By an
6823 oversight, this default was not being set up when an address was tested by
6824 -bt or -bv, which affected the outcome if any router or filter referred to
6827 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6828 log only when "message" or "log_message" are set. However, if one of the
6829 conditions was an address verification, or a call to a nested ACL, the
6830 messages generated by the underlying test were being passed through. This
6831 no longer happens. The underlying message is available in $acl_verify_
6832 message for both "message" and "log_message" expansions, so it can be
6833 passed through if needed.
6835 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6836 new expansion $bh_ to give the encoded byte string without charset
6837 translation. Translation happens only if iconv() is available; HAVE_ICONV
6838 indicates this at build time. HEADERS_CHARSET gives the charset to
6839 translate to; headers_charset can change it in the configuration, and
6840 "headers charset" can change it in an individual filter file.
6842 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6843 that creates RFC 2047 encoded "words" labels them as that charset instead
6844 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6845 expansion operator; (ii) when Exim creates a From: line for a local
6846 message; (iii) when a header line is rewritten to include a "phrase" part.
6848 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6849 buggy, causing it to skip the first lines of messages whose message ID
6850 ended in 'D'. This would not have bitten before Exim release 4.14, because
6851 message IDs were unlikely to end in 'D' before then. The effect was to have
6852 incorrect size information for certain domains.
6854 45. #include "config.h" was missing at the start of the crypt16.c module. This
6855 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6858 46. If there was a timeout during a "random" callout check, Exim treated it as
6859 a failure of the random address, and carried on sending RSET and the real
6860 address. If the delay was just some slowness somewhere, the response to the
6861 original RCPT would be taken as a response to RSET and so on, causing
6862 mayhem of various kinds.
6864 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6865 when I implemented it. It didn't allow for the fact that some option values
6866 may legitimately be negative (e.g. size_addition), and it didn't even do
6867 the right test for positive values.
6869 48. Domain names in DNS records are case-independent. Exim always looks them up
6870 in lower case. Some resolvers return domain names in exactly the case they
6871 appear in the zone file, that is, they may contain uppercase letters. Not
6872 all resolvers do this - some return always lower case. Exim was treating a
6873 change of case by a resolver as a change of domain, similar to a widening
6874 of a domain abbreviation. This triggered its re-routing code and so it was
6875 trying to route what was effectively the same domain again. This normally
6876 caused routing to fail (because the router wouldn't handle the domain
6877 twice). Now Exim checks for this case specially, and just changes the
6878 casing of the domain that it ultimately uses when it transmits the message
6881 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6884 50. If a filter generated a file delivery with a non-absolute name (possible if
6885 no home directory exists for the router), the forbid_file option was not
6888 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6889 the existing equality matching.
6891 52. Exim was using ints instead of ino_t variables in some places where it was
6892 dealing with inode numbers.
6894 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6895 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6896 called TMPDIR, and if it finds it is different, it changes its value.
6898 54. The smtp_printf() function is now made available to local_scan() so
6899 additional output lines can be written before returning. There is also an
6900 smtp_fflush() function to enable the detection of a dropped connection.
6901 The variables smtp_input and smtp_batched_input are exported to
6904 55. Changed the default runtime configuration: the message "Unknown user"
6905 has been removed from the ACL, and instead placed on the localuser router,
6906 using the cannot_route_message feature. This means that any verification
6907 failures that generate their own messages won't get overridden. Similarly,
6908 the "Unrouteable address" message that was in the ACL for unverifiable
6909 relay addresses has also been removed.
6911 56. Added hosts_avoid_esmtp to the smtp transport.
6913 57. The exicyclog script was not checking for the esoteric option
6914 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6915 will work only if exicyclog is run under the appropriate euid.
6917 58. Following a discussion on the list, the rules by which Exim recognises line
6918 endings on incoming messages have been changed. The -dropcr and drop_cr
6919 options are now no-ops, retained only for backwards compatibility. The
6920 following line terminators are recognized: LF CRLF CR. However, special
6921 processing applies to CR:
6923 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6924 nor a local message in the state where . is a terminator.
6926 (ii) If a bare CR is encountered in a header line, an extra space is added
6927 after the line terminator so as not to end the header. The reasoning
6928 behind this is that bare CRs in header lines are most likely either
6929 to be mistakes, or people trying to play silly games.
6931 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6932 was being incorrectly given as 18 bytes larger than it should have been.
6933 This is a VOB (very old bug).
6935 60. This may never have affected anything current, but just in case it has:
6936 When the local host is found other than at the start of a list of hosts,
6937 the local host, those with the same MX, and any that follow, are discarded.
6938 When the list in question was part of a longer list of hosts, the following
6939 hosts (not currently being processed) were also being discarded. This no
6940 longer happens. I'm not sure if this situation could ever has previously
6943 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6946 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6947 header. This is something that is recommended in a new Internet Draft, and
6948 is something that is documented as being done by Sendmail. There are two
6949 possible values. For messages generated by the autoreply transport, Exim
6952 Auto-Submitted: auto-replied
6954 whereas for all other generated messages (e.g. bounces) it adds
6956 Auto-Submitted: auto-generated
6958 63. The "personal" condition in filters now includes a test for the
6959 Auto-Submitted: header. If it contains the string "auto-" the message it
6960 not considered personal.
6962 64. Added rcpt_include_affixes as a generic transport option.
6964 65. Added queue_only_override (default true).
6966 66. Added the syslog_duplication option.
6968 67. If what should have been the first header line of a message consisted of
6969 a space followed by a colon, Exim was mis-interpreting it as a header line.
6970 It isn't of course - it is syntactically invalid and should therefore be
6971 treated as the start of the message body. The misbehaviour could have
6972 caused a number of strange effects, including loss of data in subsequent
6973 header lines, and spool format errors.
6975 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6976 client host had authenticated. This control can now be exercised by an ACL
6977 for more flexibility.
6979 69. By default, callouts do not happen when testing with -bh. There is now a
6980 variant, -bhc, which does actually run the callout code, including
6981 consulting and updating the callout cache.
6983 70. Added support for saslauthd authentication, courtesy of Alexander
6986 71. If statvfs() failed on the spool or log directories while checking their
6987 size for availability, Exim confusingly gave the error "space shortage".
6988 Furthermore, in debugging mode it crashed with a floating point exception.
6989 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6990 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6991 serious problem, Exim now writes to the main and panic logs when this
6992 happens, with details of the failure. It then refuses to accept the
6993 incoming message, giving the message "spool directory problem" or "log
6994 directory problem" with a 421 code for SMTP messages.
6996 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6997 0, 1, and 2 exist, because some OS complain for execs without them (see
6998 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6999 descriptors. However, the code omitted to check that the open succeeded,
7000 causing mysterious errors if for some reason the permissions on /dev/null
7001 got screwed. Now Exim writes a message to the main and panic logs, and
7002 bombs out if it can't open /dev/null.
7004 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7005 interact so that it is all more flexible. It is supposed to remain
7006 backwards compatible. Also added extra_local_interfaces.
7008 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7009 to bomb out with an assertion failure - to the client this appears as a
7010 connection drop. This problem occurs in the part of the code that was taken
7011 from the Samba project. Fortunately, the assertion is in a very simple
7012 function, so I have fixed this by reproducing the function inline in the
7013 one place where it is called, and arranging for authentication to fail
7014 instead of killing the process with assert().
7016 75. The SPA client code was not working when the server requested OEM rather
7017 than Unicode encoding.
7019 76. Added code to make require_files with a specific uid setting more usable in
7020 the case where statting the file as root fails - usually a non-root-mounted
7021 NFS file system. When this happens and the failure is EACCES, Exim now
7022 forks a subprocess and does the per-uid checking as the relevant uid.
7024 77. Added process_log_path.
7026 78. If log_file_path was not explicitly set, a setting of check_log_space or
7027 check_log_inodes was ignored.
7029 79. If a space check for the spool or log partitions fails, the incident is now
7030 logged. Of course, in the latter case the data may get lost...
7032 80. Added the %p formatting code to string_format() so that it can be used to
7033 print addresses in debug_print(). Adjusted all the address printing in the
7034 debugging in store.c to use %p rather than %d.
7036 81. There was a concern that a line of code in smtp_in.c could overflow a
7037 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7038 initially expressed, the concern was not well-founded, because trailing
7039 spaces are removed early. However, if the trailing spaces were followed by
7040 a NULL, they did not get removed, so the overflow was possible. Two fixes
7043 (a) I re-wrote the offending code in a cleaner fashion.
7044 (b) If an incoming SMTP command contains a NULL character, it is rejected
7047 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7048 runs initgroups(), so that if the Exim user is in any additional groups,
7049 they will be used during message reception.
7055 The change log for 4.20 and earlier releases has been archived.