1 Change log file for Exim from version 4.21
2 ------------------------------------------
3 This document describes *changes* to previous versions, that might
4 affect Exim's operation, with an unchanged configuration file. For new
5 options, and new features, see the NewStuff file next to this ChangeLog.
8 Since Exim version 4.90
9 -----------------------
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 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
87 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
88 metadata, resulting in a crash in free().
90 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
91 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
92 Broken also in d185889f4, with init system revamp.
94 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
95 Previously we abruptly closed the connection after reading a malware-
96 found indication; now we go on to read the "scan ok" response line,
99 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
100 ACL. Previously, a crash would result.
102 JH/19 Speed up macro lookups during configuration file read, by skipping non-
103 macro text after a replacement (previously it was only once per line) and
104 by skipping builtin macros when searching for an uppercase lead character.
106 JH/20 DANE support moved from Experimental to mainline. The Makefile control
107 for the build is renamed.
109 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
110 was allocated for every new TLS startup, meaning one per message. Fix
111 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
113 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
114 reported the original. Fix to report (as far as possible) the ACL
115 result replacing the original.
117 JH/23 Fix memory leak during multi-message connections using STARTTLS under
118 OpenSSL. Certificate information is loaded for every new TLS startup,
119 and the resources needed to be freed.
125 JH/01 Rework error string handling in TLS interface so that the caller in
126 more cases is responsible for logging. This permits library-sourced
127 string to be attached to addresses during delivery, and collapses
128 pairs of long lines into single ones.
130 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
131 during configuration. Wildcards are allowed and expanded.
133 JH/02 Rework error string handling in DKIM to pass more info back to callers.
134 This permits better logging.
136 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
137 do not close it down and have the child transport start it up again on
138 the passed-on TCP connection. Instead, proxy the child (and any
139 subsequent ones) for TLS via a unix-domain socket channel. Logging is
140 affected: the continued delivery log lines do not have any DNSSEC, TLS
141 Certificate or OCSP information. TLS cipher information is still logged.
143 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
144 identical IP addresses on different listening ports. Will also affect
147 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
148 add noisy ifdef guards to special-case this sillyness.
149 Patch from Bernd Kuhls.
151 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
152 than 255 are no longer allowed.
154 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
155 Disable the session-cache too, which might reduce our load. Since we
156 currrectly use a new context for every connection, both as server and
157 client, there is no benefit for these.
158 GnuTLS appears to not support tickets server-side by default (we don't
159 call gnutls_session_ticket_enable_server()) but client side is enabled
160 by default on recent versions (3.1.3 +) unless the PFS priority string
163 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
164 <https://reproducible-builds.org/specs/source-date-epoch/>.
166 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
167 the check for any unsuccessful recipients did not notice the limit, and
168 erroneously found still-pending ones.
170 JH/08 Pipeline CHUNKING command and data together, on kernels that support
171 MSG_MORE. Only in-clear (not on TLS connections).
173 JH/09 Avoid using a temporary file during transport using dkim. Unless a
174 transport-filter is involved we can buffer the headers in memory for
175 creating the signature, and read the spool data file once for the
176 signature and again for transmission.
178 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
179 4.77 as the kernel support then wasn't solid, having issues in 64bit
180 mode. Now, it's been long enough. Add support for FreeBSD also.
182 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
183 case where the routing stage had gathered several addresses to send to
184 a host before calling the transport for the first, we previously failed
185 to close down TLS in the old transport process before passing the TCP
186 connection to the new process. The new one sent a STARTTLS command
187 which naturally failed, giving a failed delivery and bloating the retry
188 database. Investigation and fix prototype from Wolfgang Breyha.
190 JH/12 Fix check on SMTP command input synchronisation. Previously there were
191 false-negatives in the check that the sender had not preempted a response
192 or prompt from Exim (running as a server), due to that code's lack of
193 awareness of the SMTP input buffering.
195 PP/04 Add commandline_checks_require_admin option.
196 Exim drops privileges sanely, various checks such as -be aren't a
197 security problem, as long as you trust local users with access to their
198 own account. When invoked by services which pass untrusted data to
199 Exim, this might be an issue. Set this option in main configuration
200 AND make fixes to the calling application, such as using `--` to stop
203 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
204 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
205 responses to those, into a single TLS record each way (this usually means
206 a single packet). As a side issue, smtp_enforce_sync now works on TLS
209 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
210 affects you only if you're dancing at the edge of the param size limits.
211 If you are, and this message makes sense to you, then: raise the
212 configured limit or use OpenSSL 1.1. Nothing we can do for older
215 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
216 cmdline element to get the documented default one. Previously it was
219 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
220 are unsuable for non-message-specific reasons - eg. connection timeout,
221 banner-time rejection.
223 JH/16 Fix logging of delivery remote port, when specified by router, under
226 PP/06 Repair manualroute's ability to take options in any order, even if one
227 is the name of a transport.
230 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
232 JH/17 Change the list-building routines interface to use the expanding-string
233 triplet model, for better allocation and copying behaviour.
235 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
236 Previously it was constructed the first time a possibly-matching string
237 was met in the configuration file input during startup; now it is done
240 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
241 compatible one, to avoid the (poorly documented) possibility of a config
242 file in the working directory redirecting the DB files, possibly correpting
243 some existing file. CVE-2017-10140 assigned for BDB.
245 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
246 cache-hot. Previously, although the result was properly cached, the
247 initial verify call returned a defer.
249 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
250 the main verify for receipient in uncached-mode.
252 JH/22 Retire historical build files to an "unsupported" subdir. These are
253 defined as "ones for which we have no current evidence of testing".
255 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
256 if present. Previously it was ignored.
258 JH/24 Start using specified-initialisers in C structure init coding. This is
259 a C99 feature (it's 2017, so now considered safe).
261 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
262 if was a fixed-sized field and bitmask ops via macros; it is now more
265 PP/07 GitHub PR 56: Apply MariaDB build fix.
266 Patch provided by Jaroslav Škarvada.
268 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
269 during Coverity cleanups [4.87 JH/47]
270 Diagnosis and fix provided by Michael Fischer v. Mollard.
272 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
273 the right size to place the terminating semicolon on its own folded
274 line, the header hash was calculated to an incorrect value thanks to
275 the (relaxed) space the fold became.
277 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
278 and confused the parent.
280 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
281 which could crash as a result. This could lead to undeliverable messages.
283 JH/28 Logging: "next input sent too soon" now shows where input was truncated
286 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
287 matters on fast-turnover and PID-randomising systems, which were getting
288 out-of-order delivery.
290 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
291 a possibly-overlapping copy. The symptom was that "Remote host closed
292 connection in response to HELO" was logged instead of the actual 4xx
295 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
296 Previously only that bufferd was discarded, resulting in SYMTP command
299 JH/32 DKIM: when a message has multiple signatures matching an identity given
300 in dkim_verify_signers, run the dkim acl once for each. Previously only
301 one run was done. Bug 2189.
303 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
304 "panic the current process" to "deliberately defer". The panic log is
305 still written with the problem list name; the mail and reject logs now
306 get a temp-reject line for the message that was being handled, saying
307 something like "domains check lookup or other defer". The SMTP 451
308 message is still "Temporary local problem".
310 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
311 A crafted sequence of BDAT commands could result in in-use memory beeing
312 freed. CVE-2017-16943.
314 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
315 from SMTP input. Previously it was always done; now only done for DATA
316 and not BDAT commands. CVE-2017-16944.
318 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
319 to the message (such as an overlong header line). Previously this was
320 not done and we did not exit BDAT mode. Followon from the previous item
321 though a different problem.
327 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
328 than -2003 did; needs libidn2 in addition to libidn.
330 JH/02 The path option on a pipe transport is now expanded before use.
332 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
333 Patch provided by "Björn", documentation fix added too.
335 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
336 missing a wire-to-host endian conversion.
338 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
339 close after a BDAT command line could be taken as a following command,
340 giving a synch failure. Fix by only checking for synch immediately
341 before acknowledging the chunk.
343 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
344 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
345 macro. Patches provided by Josh Soref.
347 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
348 Previously we did not; the RFC seems ambiguous and VRFY is not listed
349 by IANA as a service extension. However, John Klensin suggests that we
352 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
353 the dkim code may be unix-mode line endings rather than smtp wire-format
354 CRLF, so prepend a CR to any bare LF.
356 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
357 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
359 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
360 the first were themselves being wrongly included in the feed into dkim
361 processing; with most chunk sizes in use this resulted in an incorrect
362 body hash calculated value.
364 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
365 DKIM signature block, for verification. Although advised against by
366 standards it is specifically not ruled illegal.
368 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
370 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
371 missing a body hash (the bh= tag).
373 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
374 It seems that HAProxy sends the Proxy Protocol information in clear and
375 only then does a TLS startup, so do the same.
377 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
378 TCP connections (such as for Spamd) unless the daemon successfully set
379 Fast Open mode on its listening sockets. This fixes breakage seen on
380 too-old kernels or those not configured for Fast Open, at the cost of
381 requiring both directions being enabled for TFO, and TFO never being used
382 by non-daemon-related Exim processes.
384 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
385 endings, at least on the first header line. Try to canonify any that get
386 past that check, despite the cost.
388 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
389 now limited to an arbitrary five deep, while parsing addresses with the
390 strip_excess_angle_brackets option enabled.
392 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
393 instead leave the unprompted TLS handshake in socket buffer for the
394 TLS library to consume.
396 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
398 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
400 JH/16 Drop variables when they go out of scope. Memory management drops a whole
401 region in one operation, for speed, and this leaves assigned pointers
402 dangling. Add checks run only under the testsuite which checks all
403 variables at a store-reset and panics on a dangling pointer; add code
404 explicitly nulling out all the variables discovered. Fixes one known
405 bug: a transport crash, where a dangling pointer for $sending_ip_address
406 originally assigned in a verify callout, is re-used.
408 PP/06 Drop '.' from @INC in various Perl scripts.
410 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
412 PP/08 Reduce a number of compilation warnings under clang; building with
413 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
414 should be warning-free.
416 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
418 HS/01 Fix portability problems introduced by PP/08 for platforms where
419 realloc(NULL) is not equivalent to malloc() [SunOS et al].
421 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
422 chunk. This allows us to accept broken chunked messages. We need a more
423 general solution here.
425 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
426 already-broken messages in the queue.
428 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
430 JH/19 Fix reference counting bug in routing-generated-address tracking.
436 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
437 supports it and a size is available (ie. the sending peer gave us one).
439 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
440 years of being deprecated). The replacements are the ACLs
441 acl_smtp_mime and acl_not_smtp_mime.
443 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
444 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
445 or even an in-clear connection were permitted. Now, if the host lookup
446 was dnssec and dane was requested then the host is only used if the
447 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
448 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
449 if one fails this test.
450 This means that a poorly-configured remote DNS will make it incommunicado;
451 but it protects against a DNS-interception attack on it.
453 JH/04 Bug 1810: make continued-use of an open smtp transport connection
454 non-noisy when a race steals the message being considered.
456 JH/05 If main configuration option tls_certificate is unset, generate a
457 self-signed certificate for inbound TLS connections.
459 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
460 in rewrites and routers.
462 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
463 and logged a warning sing 4.83; now they are a configuration file error.
465 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
466 (lacking @domain). Apply the same qualification processing as RCPT.
468 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
470 JH/10 Support ${sha256:} applied to a string (as well as the previous
473 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
474 a cutthrough deliver is pending, as we always want to make a connection.
475 This also avoids re-routing the message when later placing the cutthrough
476 connection after a verify cache hit.
477 Do not update it with the verify result either.
479 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
480 when routing results in more than one destination address.
482 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
483 signing (which inhibits the cutthrough capability). Previously only
484 the presence of an option was tested; now an expansion evaluating as
485 empty is permissible (obviously it should depend only on data available
486 when the cutthrough connection is made).
488 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
489 the relevant preceding SMTP command did not note the pipelining mode.
491 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
492 Previously they were not counted.
494 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
495 as one having no matching records. Previously we deferred the message
496 that needed the lookup.
498 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
499 distinguished as "(=".
501 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
502 for missing MX records. Previously it only worked for missing A records.
504 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
506 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
507 after the data-go-ahead and data-ack. Patch from Jason Betts.
509 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
510 even for a "none" policy. Patch from Tony Meyer.
512 JH/22 Fix continued use of a connection for further deliveries. If a port was
513 specified by a router, it must also match for the delivery to be
516 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
517 When one of the recipients of a message was unsuitable for the connection
518 (has no matching addresses), we lost track of needing to mark it
519 deferred. As a result mail would be lost.
521 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
523 JH/25 Decoding ACL controls is now done using a binary search; the source code
524 takes up less space and should be simpler to maintain. Merge the ACL
525 condition decode tables also, with similar effect.
527 JH/26 Fix problem with one_time used on a redirect router which returned the
528 parent address unchanged. A retry would see the parent address marked as
529 delivered, so not attempt the (identical) child. As a result mail would
532 JH/27 Fix a possible security hole, wherein a process operating with the Exim
533 UID can gain a root shell. Credit to http://www.halfdog.net/ for
534 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
537 JH/28 Enable {spool,log} filesystem space and inode checks as default.
538 Main config options check_{log,spool}_{inodes,space} are now
539 100 inodes, 10MB unless set otherwise in the configuration.
541 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
542 Previously it only applied to the main-section connection policy
545 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
547 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
548 by me. Added RFC7919 DH primes as an alternative.
550 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
551 are not in the system include path.
553 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
554 GnuTLS, when a session startup failed (eg because the client disconnected)
555 Exim did stdio operations after fclose. This was exposed by a recent
556 change which nulled out the file handle after the fclose.
558 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
559 signed directly by the cert-signing cert, rather than an intermediate
560 OCSP-signing cert. This is the model used by LetsEncrypt.
562 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
564 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
565 an incoming connection.
567 HS/02 Bug 1802: Do not half-close the connection after sending a request
570 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
571 fallback to "prime256v1".
573 JH/34 SECURITY: Use proper copy of DATA command in error message.
574 Could leak key material. Remotely exploitable. CVE-2016-9963.
580 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
581 and 3.4.4 - once the server is enabled to respond to an OCSP request
582 it does even when not requested, resulting in a stapling non-aware
583 client dropping the TLS connection.
585 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
586 support variable-length bit vectors. No functional change.
588 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
589 The I= interface field on outgoing lines is now after the H= remote
590 host field, same as incoming lines. There is a separate
591 outgoing_interface log selector which allows you to disable the
594 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
595 If not running log_selector +smtp_connection the mainlog would be held
596 open indefinitely after a "too many connections" event, including to a
597 deleted file after a log rotate. Leave the per net connection logging
598 leaving it open for efficiency as that will be quickly detected by the
599 check on the next write.
601 HS/01 Bug 1671: Fix post transport crash.
602 Processing the wait-<transport> messages could crash the delivery
603 process if the message IDs didn't exist for some reason. When
604 using 'split_spool_directory=yes' the construction of the spool
605 file name failed already, exposing the same netto behaviour.
607 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
608 mime_regex ACL conditions.
610 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
611 to DSN fail messages (bounces): remote IP, remote greeting, remote response
612 to HELO, local diagnostic string.
614 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
615 log line to debug. Even when configured with a tls authenticator many
616 client connections are expected to not authenticate in this way, so
617 an authenticate fail is not an error.
619 HS/02 Add the Exim version string to the process info. This way exiwhat
620 gives some more detail about the running daemon.
622 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
623 matter for fast-change records such as DNSBLs.
625 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
626 retry record, even if constant. There may be multiple transports with
627 different interface settings and the retry behaviour needs to be kept
630 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
632 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
634 JH/10 Bug 840: fix log_defer_output option of pipe transport
636 JH/11 Bug 830: use same host for all RCPTS of a message, even under
637 hosts_randomize. This matters a lot when combined with mua_wrapper.
639 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
640 ${quote_pgsql:<string>} operator.
642 JH/13 Bug 1708: avoid misaligned access in cached lookup.
644 JH/14 Change header file name for freeradius-client. Relevant if compiling
645 with Radius support; from the Gentoo tree and checked under Fedora.
647 JH/15 Bug 1712: Introduce $prdr_requested flag variable
649 JH/16 Bug 1714: Permit an empty string as expansion result for transport
650 option transport_filter, meaning no filtering.
652 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
654 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
655 defaults to "*" (all hosts). The variable is now available when not built
656 with TLS, default unset, mainly to enable keeping the testsuite sane.
657 If a server certificate is not supplied (via tls_certificate) an error is
658 logged, and clients will find TLS connections fail on startup. Presumably
659 they will retry in-clear.
660 Packagers of Exim are strongly encouraged to create a server certificate
661 at installation time.
663 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
664 with the $config_file variable.
666 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
667 in transport context, after the attempt, and per-recipient. The latter type
668 is per host attempted. The event data is the error message, and the errno
669 information encodes the lookup type (A vs. MX) used for the (first) host,
670 and the trailing two digits of the smtp 4xx response.
672 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
673 to write to mainlog (or rejectlog, paniclog) in the window between file
674 creation and permissions/ownership being changed. Particularly affects
675 installations where exicyclog is run as root, rather than exim user;
676 result is that the running daemon panics and dies.
678 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
680 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
681 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
682 "pri" and "weight". Note that the previous implicit priority given by the
683 list order is no longer honoured.
685 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
688 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
689 by defining SUPPORT_SOCKS.
691 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
692 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
693 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
694 variables are renamed to proxy_{local,external}_{address,port}.
696 JH/27 Move Internationalisation support from Experimental to mainline, enabled
697 for a build by defining SUPPORT_I18N
699 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
700 of the query string, and make ${quote_redis:} do that quoting.
702 JH/29 Move Events support from Experimental to mainline, enabled by default
703 and removable for a build by defining DISABLE_EVENT.
705 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
707 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
708 cached by the daemon.
710 JH/32 Move Redis support from Experimental to mainline, enabled for a build
711 by defining LOOKUP_REDIS. The libhiredis library is required.
713 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
714 keys are given for lookup.
716 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
717 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
718 only supported when built with TLS support. The PolarSSL SHA routines
719 are still used when the TLS library is too old for convenient support.
721 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
722 openssl_options), for security. OpenSSL forces this from version 1.1.0
723 server-side so match that on older versions.
725 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
726 allocation for $value could be released as the expansion processing
727 concluded, but leaving the global pointer active for it.
729 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
730 and to use the domains and local_parts ACL conditions.
732 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
733 incorrectly not doubled on cutthrough transmission, hence seen as a
734 body-termination at the receiving system - resulting in truncated mails.
735 Commonly the sender saw a TCP-level error, and retransmitted the message
736 via the normal store-and-forward channel. This could result in duplicates
737 received - but deduplicating mailstores were liable to retain only the
738 initial truncated version.
740 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
742 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
744 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
745 we're in there, support oversigning also; bug 1309.
747 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
749 HS/04 Add support for keep_environment and add_environment options.
751 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
752 either intentional arithmetic overflow during PRNG, or testing config-
755 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
756 delivery resulted in actual delivery. Cancel cutthrough before DATA
759 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
760 rejects a recipient: pass the reject to the originator.
762 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
763 Many were false-positives and ignorable, but it's worth fixing the
766 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
767 for the new environment-manipulation done at startup. Move the routines
768 from being local to tls.c to being global via the os.c file.
770 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
771 an extract embedded as result-arg for a map, the first arg for extract
772 is unavailable so we cannot tell if this is a numbered or keyed
773 extraction. Accept either.
779 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
782 JH/02 The smtp transport option "multi_domain" is now expanded.
784 JH/03 The smtp transport now requests PRDR by default, if the server offers
787 JH/04 Certificate name checking on server certificates, when exim is a client,
788 is now done by default. The transport option tls_verify_cert_hostnames
789 can be used to disable this per-host. The build option
790 EXPERIMENTAL_CERTNAMES is withdrawn.
792 JH/05 The value of the tls_verify_certificates smtp transport and main options
793 default to the word "system" to access the system default CA bundle.
794 For GnuTLS, only version 3.0.20 or later.
796 JH/06 Verification of the server certificate for a TLS connection is now tried
797 (but not required) by default. The verification status is now logged by
798 default, for both outbound TLS and client-certificate supplying inbound
801 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
804 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
805 Status Notification (bounce) messages are now MIME format per RFC 3464.
806 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
807 under the control of the dsn_advertise_hosts option, and routers may
808 have a dsn_lasthop option.
810 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
811 default, modifiable by a malware= option. The list separator for
812 the options can now be changed in the usual way. Bug 68.
814 JH/10 The smtp_receive_timeout main option is now expanded before use.
816 JH/11 The incoming_interface log option now also enables logging of the
817 local interface on delivery outgoing connections.
819 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
820 if the interface and destination host and port all match.
822 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
825 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
826 Patch from Andrew Lewis.
828 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
829 now supports optional time-restrictions, weighting, and priority
830 modifiers per server. Patch originally by <rommer@active.by>.
832 JH/16 The spamd_address main option now supports a mixed list of local
833 and remote servers. Remote servers can be IPv6 addresses, and
834 specify a port-range.
836 JH/17 Bug 68: The spamd_address main option now supports an optional
837 timeout value per server.
839 JH/18 Bug 1581: Router and transport options headers_add/remove can
840 now have the list separator specified.
842 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
845 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
848 JH/21 Support for the A6 type of dns record is withdrawn.
850 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
851 rather than the verbs used.
853 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
854 from 255 to 1024 chars.
856 JH/24 Verification callouts now attempt to use TLS by default.
858 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
859 are generic router options now. The defaults didn't change.
861 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
862 Original patch from Alexander Shikoff, worked over by JH.
864 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
865 files if ZCAT_COMMAND is not executable.
867 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
869 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
871 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
872 Normally benign, it bites when the pair was led to by a CNAME;
873 modern usage is to not canonicalize the domain to a CNAME target
874 (and we were inconsistent anyway for A-only vs AAAA+A).
876 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
878 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
879 when evaluating $sender_host_dnssec.
881 JH/31 Check the HELO verification lookup for DNSSEC, adding new
882 $sender_helo_dnssec variable.
884 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
886 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
888 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
890 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
891 documented as working, but never had. Support all but $spam_report.
893 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
894 added for tls authenticator.
896 HS/03 Add perl_taintmode main config option
902 TL/01 When running the test suite, the README says that variables such as
903 no_msglog_check are global and can be placed anywhere in a specific
904 test's script, however it was observed that placement needed to be near
905 the beginning for it to behave that way. Changed the runtest perl
906 script to read through the entire script once to detect and set these
907 variables, reset to the beginning of the script, and then run through
908 the script parsing/test process like normal.
910 TL/02 The BSD's have an arc4random API. One of the functions to induce
911 adding randomness was arc4random_stir(), but it has been removed in
912 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
913 function when detected.
915 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
916 cause callback expansion.
918 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
919 syntax errors in an expansion can be treated as a string instead of
920 logging or causing an error, due to the internal use of bool_lax
921 instead of bool when processing it.
923 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
924 server certificates when making smtp deliveries.
926 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
928 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
930 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
932 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
933 Merged patch from Sebastian Wiedenroth.
935 JH/05 Fix results-pipe from transport process. Several recipients, combined
936 with certificate use, exposed issues where response data items split
937 over buffer boundaries were not parsed properly. This eventually
938 resulted in duplicates being sent. This issue only became common enough
939 to notice due to the introduction of connection certificate information,
940 the item size being so much larger. Found and fixed by Wolfgang Breyha.
942 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
943 size buffer was used, resulting in syntax errors when an expansion
946 JH/07 Add support for directories of certificates when compiled with a GnuTLS
947 version 3.3.6 or later.
949 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
950 is EXPERIMENTAL_EVENT, the main-configuration and transport options
951 both become "event_action", the variables become $event_name, $event_data
952 and $event_defer_errno. There is a new variable $verify_mode, usable in
953 routers, transports and related events. The tls:cert event is now also
954 raised for inbound connections, if the main configuration event_action
957 TL/06 In test suite, disable OCSP for old versions of openssl which contained
958 early OCSP support, but no stapling (appears to be less than 1.0.0).
960 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
961 server certificate names available under the smtp transport option
962 "tls_verify_cert_hostname" now do not permit multi-component wildcard
965 JH/10 Time-related extraction expansions from certificates now use the main
966 option "timezone" setting for output formatting, and are consistent
967 between OpenSSL and GnuTLS compilations. Bug 1541.
969 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
970 encoded parameter in the incoming message. Bug 1558.
972 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
973 include certificate info, eximon was claiming there were spoolfile
976 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
978 JH/14 Log delivery-related information more consistently, using the sequence
979 "H=<name> [<ip>]" wherever possible.
981 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
982 are problematic for Debian distribution, omit them from the release
985 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
987 JH/16 Fix string representation of time values on 64bit time_t architectures.
990 JH/17 Fix a null-indirection in certextract expansions when a nondefault
991 output list separator was used.
996 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
997 checkers that were complaining about end of non-void function with no
1000 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1001 This was a regression introduced in 4.83 by another bugfix.
1003 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1005 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1006 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1012 TF/01 Correctly close the server side of TLS when forking for delivery.
1014 When a message was received over SMTP with TLS, Exim failed to clear up
1015 the incoming connection properly after forking off the child process to
1016 deliver the message. In some situations the subsequent outgoing
1017 delivery connection happened to have the same fd number as the incoming
1018 connection previously had. Exim would try to use TLS and fail, logging
1019 a "Bad file descriptor" error.
1021 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1022 utilities have not been installed.
1024 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1025 temporary space as the ACL may create new global variables.
1027 TL/01 LDAP support uses per connection or global context settings, depending
1028 upon the detected version of the libraries at build time.
1030 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1031 to extract and use the src ip:port in logging and expansions as if it
1032 were a direct connection from the outside internet. PPv2 support was
1033 updated based on HAProxy spec change in May 2014.
1035 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1037 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1038 Properly escape header and check for NULL return.
1040 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1043 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1045 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1046 characters in header names, implemented as a verify condition.
1047 Contributed by Michael Fischer v. Mollard.
1049 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1050 results permerror and temperror. Previous values are deprecated but
1051 still accepted. In a future release, err_perm and err_temp will be
1052 completely removed, which will be a backward incompatibility if the
1053 ACL tests for either of these two old results. Patch contributed by
1054 user bes-internal on the mailing list.
1056 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1058 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1059 selectors, in both main and reject logs.
1061 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1064 JH/07 Add malware type "sock" for talking to simple daemon.
1066 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1068 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1069 routers/transports under cutthrough routing.
1071 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1072 numbers. Touch up "bool" conditional to keep the same definition.
1074 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1076 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1078 JH/12 Expand items in router/transport headers_add or headers_remove lists
1079 individually rather than the list as a whole. Bug 1452.
1081 Required for reasonable handling of multiple headers_ options when
1082 they may be empty; requires that headers_remove items with embedded
1083 colons must have them doubled (or the list-separator changed).
1085 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1086 view the policy declared in the DMARC record. Currently, $dmarc_status
1087 is a combined value of both the record presence and the result of the
1090 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1092 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1093 dnslookup router and the smtp transport (applying to the forward
1096 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1097 of ldap servers used for a specific lookup. Patch provided by Heiko
1100 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1101 New variable $lookup_dnssec_authenticated for observability.
1103 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1104 Patch submitted by Lars Timman.
1106 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1108 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1109 Requires trusted mode and valid format message id, aborts otherwise.
1110 Patch contributed by Heiko Schlichting.
1112 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1113 certextract with support for various fields. Bug 1358.
1115 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1116 is requested by default, modifiable by smtp transport option
1119 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1120 operate on certificate variables to give certificate fingerprints
1121 Also new ${sha256:cert_variable}.
1123 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1125 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1126 Christian Aistleitner.
1128 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1130 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1131 file. Patch from Wolfgang Breyha.
1133 JH/25 Expand the coverage of the delivery $host and $host_address to
1134 client authenticators run in verify callout. Bug 1476.
1136 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1137 align with daemon_smtp_ports. Bug 72.
1139 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1140 support and error reporting did not work properly.
1142 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1143 and is readable. Patch from Andrew Colin Kissa.
1145 TL/14 Enhance documentation of ${run expansion and how it parses the
1146 commandline after expansion, particularly in the case when an
1147 unquoted variable expansion results in an empty value.
1149 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1151 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1152 by renaming away from it. Addresses GH issue 15, reported by
1155 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1157 TL/15 SECURITY: prevent double expansion in math comparison functions
1158 (can expand unsanitized data). Not remotely exploitable.
1165 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1167 PP/02 Make -n do something, by making it not do something.
1168 When combined with -bP, the name of an option is not output.
1170 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1173 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1174 $sender_host_name and config options to manage this, and basic check
1177 PP/05 DSCP support for outbound connections and control modifier for inbound.
1179 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1180 (Only plugin which currently uses this is kerberos4, which nobody should
1181 be using, but we should make it available and other future plugins might
1182 conceivably use it, even though it would break NAT; stuff *should* be
1183 using channel bindings instead).
1185 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1186 name; added for Sendmail compatibility; requires admin caller.
1187 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1188 just ignore it); requires trusted caller.
1189 Also parse but ignore: -Ac -Am -X<logfile>
1192 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1194 TL/02 Add +smtp_confirmation as a default logging option.
1196 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1197 Patch by Magnus Holmgren from 2007-02-20.
1199 TL/04 Bugzilla 1281 - Spec typo.
1200 Bugzilla 1283 - Spec typo.
1201 Bugzilla 1290 - Spec grammar fixes.
1203 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1205 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1207 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1208 mailing list by by Dmitry Isaikin.
1210 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1212 JH/02 Support "G" suffix to numbers in ${if comparisons.
1214 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1216 NM/01 Bugzilla 1197 - Spec typo
1217 Bugzilla 1196 - Spec examples corrections
1219 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1221 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1222 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1225 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1226 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1228 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1229 "acl {{name}{arg}...}", and optional args on acl condition
1232 JH/05 Permit multiple router/transport headers_add/remove lines.
1234 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1236 JH/07 Avoid using a waiting database for a single-message-only transport.
1237 Performance patch from Paul Fisher. Bugzilla 1262.
1239 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1242 JH/09 Add $headers_added variable, with content from use of ACL modifier
1243 add_header (but not yet added to the message). Bugzilla 199.
1245 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1246 Pulled from Bugzilla 817 by Wolfgang Breyha.
1248 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1250 (nb: this is the same fix as in Exim 4.80.1)
1252 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1255 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1258 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1260 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1261 advertise SMTP AUTH mechanism to us, instead of a generic
1262 protocol violation error. Also, make Exim more robust to bad
1263 data from the Dovecot auth socket.
1265 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1267 When a queue runner is handling a message, Exim first routes the
1268 recipient addresses, during which it prunes them based on the retry
1269 hints database. After that it attempts to deliver the message to
1270 any remaining recipients. It then updates the hints database using
1273 So if a recipient address works intermittently, it can get repeatedly
1274 deferred at routing time. The retry hints record remains fresh so the
1275 address never reaches the final cutoff time.
1277 This is a fairly common occurrence when a user is bumping up against
1278 their storage quota. Exim had some logic in its local delivery code
1279 to deal with this. However it did not apply to per-recipient defers
1280 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1282 This change adds a proper retry rule check during routing so that the
1283 final cutoff time is checked against the message's age. We only do
1284 this check if there is an address retry record and there is not a
1285 domain retry record; this implies that previous attempts to handle
1286 the address had the retry_use_local_parts option turned on. We use
1287 this as an approximation for the destination being like a local
1288 delivery, as in LMTP.
1290 I suspect this new check makes the old local delivery cutoff check
1291 redundant, but I have not verified this so I left the code in place.
1293 TF/02 Correct gecos expansion when From: is a prefix of the username.
1295 Test 0254 submits a message to Exim with the header
1299 When I ran the test suite under the user fanf2, Exim expanded
1300 the header to contain my full name, whereas it should have added
1301 a Resent-Sender: header. It erroneously treats any prefix of the
1302 username as equal to the username.
1304 This change corrects that bug.
1306 GF/01 DCC debug and logging tidyup
1307 Error conditions log to paniclog rather than rejectlog.
1308 Debug lines prefixed by "DCC: " to remove any ambiguity.
1310 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1312 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1313 Bug spotted by Jeremy Harris; was flawed since initial commit.
1314 Would have resulted in OCSP responses post-SNI triggering an Exim
1315 NULL dereference and crash.
1317 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1319 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1320 Bug detection, analysis and fix by Samuel Thibault.
1321 Bugzilla 1331, Debian bug #698092.
1323 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1325 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1326 Server implementation by Todd Lyons, client by JH.
1327 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1328 config variable "prdr_enable" controls whether the server
1329 advertises the facility. If the client requests PRDR a new
1330 acl_data_smtp_prdr ACL is called once for each recipient, after
1331 the body content is received and before the acl_smtp_data ACL.
1332 The client is controlled by both of: a hosts_try_prdr option
1333 on the smtp transport, and the server advertisement.
1334 Default client logging of deliveries and rejections involving
1335 PRDR are flagged with the string "PRDR".
1337 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1338 fclose(). Diagnosis by Todd Lyons.
1340 PP/17 Update configure.default to handle IPv6 localhost better.
1341 Patch by Alain Williams (plus minor tweaks).
1344 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1345 This is now consistent with GnuTLS, and is now documented: the
1346 previous undocumented portable approach to treating the option as
1347 unset was to force an expansion failure. That still works, and
1348 an empty string is now equivalent.
1350 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1351 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1352 not performing validation itself.
1354 PP/20 Added force_command boolean option to pipe transport.
1355 Patch from Nick Koston, of cPanel Inc.
1357 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1360 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1362 PP/21 Fix eximon continuous updating with timestamped log-files.
1363 Broken in a format-string cleanup in 4.80, missed when I repaired the
1364 other false fix of the same issue.
1365 Report and fix from Heiko Schlichting.
1368 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1369 Report from Prashanth Katuri.
1371 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1372 It's SecureTransport, so affects any MacOS clients which use the
1373 system-integrated TLS libraries, including email clients.
1375 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1376 using a MIME ACL for non-SMTP local injection.
1377 Report and assistance in diagnosis by Warren Baker.
1379 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1381 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1383 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1384 last id that failed so it may be referenced in subsequent ACL's.
1386 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1389 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1390 ldap library initialization, allowing self-signed CA's to be
1391 used. Also properly sets require_cert option later in code by
1392 using NULL (global ldap config) instead of ldap handle (per
1393 session). Bug diagnosis and testing by alxgomz.
1395 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1396 the src/util/ subdirectory.
1398 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1399 renamed to Transport Post Delivery Action by Jeremy Harris, as
1402 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1403 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1404 redis_servers = needs to be configured which will be used by the redis
1405 lookup. Patch from Warren Baker, of The Packet Hub.
1407 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1409 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1410 hostname or reverse DNS when processing a host list. Used suggestions
1411 from multiple comments on this bug.
1413 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1415 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1416 Missed a few lines, added it to make the runtest require no keyboard
1419 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1420 contains upper case chars. Make router use caseful_local_part.
1422 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1423 support when GnuTLS has been built with p11-kit.
1429 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1431 This, or similar/improved, will also be change PP/11 of 4.82.
1437 PP/01 Handle short writes when writing local log-files.
1438 In practice, only affects FreeBSD (8 onwards).
1439 Bugzilla 1053, with thanks to Dmitry Isaikin.
1441 NM/01 Bugzilla 949 - Documentation tweak
1443 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1446 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1448 PP/02 Implemented gsasl authenticator.
1450 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1452 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1453 `pkg-config foo` for cflags/libs.
1455 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1456 with rest of GSASL and with heimdal_gssapi.
1458 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1459 `pkg-config foo` for cflags/libs for the TLS implementation.
1461 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1462 properties get this fed in as external SSF. A number of robustness
1463 and debugging improvements to the cyrus_sasl authenticator.
1465 PP/08 cyrus_sasl server now expands the server_realm option.
1467 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1468 Patch by Jeremy Harris.
1470 PP/10 Added dbmjz lookup type.
1472 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1474 PP/12 MAIL args handles TAB as well as SP, for better interop with
1475 non-compliant senders.
1476 Analysis and variant patch by Todd Lyons.
1478 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1479 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1480 Patch from Dirk Mueller <dmueller@suse.com>
1482 PP/13 tls_peerdn now print-escaped for spool files.
1483 Observed some $tls_peerdn in wild which contained \n, which resulted
1484 in spool file corruption.
1486 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1487 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1488 or write after TLS renegotiation, which otherwise led to messages
1491 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1492 as a tracking header (ie: a signed header comes before the signature).
1493 Patch from Wolfgang Breyha.
1495 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1496 comma-sep list; embedded commas doubled.
1498 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1500 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1502 Report and patch from Dmitry Banschikov.
1504 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1505 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1506 failures. We appear to now support TLS1.1+ with Exim.
1508 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1509 lets Exim select keys and certificates based upon TLS SNI from client.
1510 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1511 before an outbound SMTP session. New log_selector, +tls_sni.
1513 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1514 NULL dereference. Report and patch from Alun Jones.
1516 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1517 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1518 Not seeing resolver debug output on NetBSD, but suspect this is a
1519 resolver implementation change.
1521 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1522 Left warnings. Added "eximon gdb" invocation mode.
1524 PP/21 Defaulting "accept_8bitmime" to true, not false.
1526 PP/22 Added -bw for inetd wait mode support.
1528 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1529 locate the relevant includes and libraries. Made this the default.
1531 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1532 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1534 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1535 This may cause build issues on older platforms.
1537 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1538 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1539 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1540 Added SNI support via GnuTLS too.
1541 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1543 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1545 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1546 Applied second patch from Janne, implementing suggestion to default
1547 multiple-strings-in-record handling to match SPF spec.
1549 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1551 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1552 read-only, out of scope).
1553 Patch from Wolfgang Breyha, report from Stuart Northfield.
1555 PP/29 Fix three issues highlighted by clang analyser static analysis.
1556 Only crash-plausible issue would require the Cambridge-specific
1557 iplookup router and a misconfiguration.
1558 Report from Marcin Mirosław.
1560 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1562 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1563 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1564 As part of this, removing so much warning spew let me fix some minor
1565 real issues in debug logging.
1567 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1568 assignment on my part. Fixed.
1570 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1571 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1572 Janne Snabb (who went above and beyond: thank you).
1574 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1575 string otherwise requires a connection and a bunch more work and it's
1576 relatively easy to get wrong. Should also expose TLS library linkage
1579 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1580 64-bit ${eval} (JH/03).
1582 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1583 GNU libc to support some of the 64-bit stuff, should not lead to
1584 conflicts. Defined before os.h is pulled in, so if a given platform
1585 needs to override this, it can.
1587 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1588 protection layer was required, which is not implemented.
1589 Bugzilla 1254, patch from Wolfgang Breyha.
1591 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1592 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1593 tls_dhparam take prime identifiers. Also unbreak combination of
1594 OpenSSL+DH_params+TLSSNI.
1596 PP/39 Disable SSLv2 by default in OpenSSL support.
1602 PP/01 Solaris build fix for Oracle's LDAP libraries.
1603 Bugzilla 1109, patch from Stephen Usher.
1605 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1607 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1610 TF/02 Fix a couple more cases where we did not log the error message
1611 when unlink() failed. See also change 4.74-TF/03.
1613 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1614 lock up or crash if it happened to be inside a call to libc when it
1615 got a SIGUSR1 from exiwhat.
1617 The SIGUSR1 handler appends the current process status to the process
1618 log which is later printed by exiwhat. It used to use the general
1619 purpose logging code to do this, but several functions it calls are
1620 not safe for signals.
1622 The new output code in the SIGUSR1 handler is specific to the process
1623 log, and simple enough that it's easy to inspect for signal safety.
1624 Removing some special cases also simplifies the general logging code.
1625 Removing the spurious timestamps from the process log simplifies
1628 TF/04 Improved ratelimit ACL condition.
1630 The /noupdate option has been deprecated in favour of /readonly which
1631 has clearer semantics. The /leaky, /strict, and /readonly update modes
1632 are mutually exclusive. The update mode is no longer included in the
1633 database key; it just determines when the database is updated. (This
1634 means that when you upgrade Exim will forget old rate measurements.)
1636 Exim now checks that the per_* options are used with an update mode that
1637 makes sense for the current ACL. For example, when Exim is processing a
1638 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1639 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1640 must specify per_mail/readonly. If you omit the update mode it defaults to
1641 /leaky where that makes sense (as before) or /readonly where required.
1643 The /noupdate option is now undocumented but still supported for
1644 backwards compatibility. It is equivalent to /readonly except that in
1645 ACLs where /readonly is required you may specify /leaky/noupdate or
1646 /strict/noupdate which are treated the same as /readonly.
1648 A useful new feature is the /count= option. This is a generalization
1649 of the per_byte option, so that you can measure the throughput of other
1650 aggregate values. For example, the per_byte option is now equivalent
1651 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1653 The per_rcpt option has been generalized using the /count= mechanism
1654 (though it's more complicated than the per_byte equivalence). When it is
1655 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1656 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1657 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1658 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1659 Note that using per_rcpt with a non-readonly update mode in more than
1660 one ACL will cause the recipients to be double-counted. (The per_mail
1661 and per_byte options don't have this problem.)
1663 The handling of very low rates has changed slightly. If the computed rate
1664 is less than the event's count (usually one) then this event is the first
1665 after a long gap. In this case the rate is set to the same as this event's
1666 count, so that the first message of a spam run is counted properly.
1668 The major new feature is a mechanism for counting the rate of unique
1669 events. The new per_addr option counts the number of different
1670 recipients that someone has sent messages to in the last time period. It
1671 behaves like per_rcpt if all the recipient addresses are different, but
1672 duplicate recipient addresses do not increase the measured rate. Like
1673 the /count= option this is a general mechanism, so the per_addr option
1674 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1675 example, measure the rate that a client uses different sender addresses
1676 with the options per_mail/unique=$sender_address. There are further
1677 details in the main documentation.
1679 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1681 TF/06 Removed a few PCRE remnants.
1683 TF/07 Automatically extract Exim's version number from tags in the git
1684 repository when doing development or release builds.
1686 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1687 Bugzilla 879. Patch from Paul Fisher.
1689 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1690 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1693 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1695 PP/05 Variable $av_failed, true if the AV scanner deferred.
1696 Bugzilla 1078. Patch from John Horne.
1698 PP/06 Stop make process more reliably on build failure.
1699 Bugzilla 1087. Patch from Heiko Schlittermann.
1701 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1702 Bugzilla 1089. Patch from Heiko Schlittermann.
1704 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1705 Bugzilla 1131. Patch from Holger Weiß.
1707 PP/09 Handle IPv6 addresses with SPF.
1708 Bugzilla 860. Patch from Wolfgang Breyha.
1710 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1712 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1715 PP/11 match_* no longer expand right-hand-side by default.
1716 New compile-time build option, EXPAND_LISTMATCH_RHS.
1717 New expansion conditions, "inlist", "inlisti".
1719 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1721 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1723 PP/14 fix log_write() format string regression from TF/03.
1724 Bugzilla 1152. Patch from Dmitry Isaikin.
1730 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1732 PP/02 Harmonised TLS library version reporting; only show if debugging.
1733 Layout now matches that introduced for other libraries in 4.74 PP/03.
1735 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1737 PP/04 New "dns_use_edns0" global option.
1739 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1742 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1743 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1745 TK/01 Updated PolarSSL code to 0.14.2.
1746 Bugzilla 1097. Patch from Andreas Metzler.
1748 PP/07 Catch divide-by-zero in ${eval:...}.
1749 Fixes bugzilla 1102.
1751 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1754 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1755 format-string attack -- SECURITY: remote arbitrary code execution.
1757 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1758 time unintentionally subject to list matching rules, letting the header
1759 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1760 arbitrary string expansion). This allowed for information disclosure.
1762 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1763 INT_MIN/-1 -- value coerced to INT_MAX.
1769 NM/01 Workaround for PCRE version dependency in version reporting
1772 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1773 This fixes portability to compilers other than gcc, notably
1774 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1776 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1777 makefiles for portability to HP-UX and POSIX correctness.
1779 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1780 Also via indented variable definition in the Makefile.
1781 (Debugging by Oliver Heesakkers).
1783 PP/02 Restore caching of spamd results with expanded spamd_address.
1784 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1786 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1787 Improves build reliability. Fix from: Frank Elsner
1789 NM/02 Fix wide character breakage in the rfc2047 coding
1790 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1792 NM/03 Allow underscore in dnslist lookups
1793 Fixes bug 1026. Patch from Graeme Fowler
1795 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1796 Code patches from Adam Ciarcinski of NetBSD.
1798 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1801 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1802 is logged, to avoid truncation. Patch from John Horne.
1804 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1805 Patch from Jakob Hirsch.
1807 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1808 SQL string expansion failure details.
1809 Patch from Andrey Oktyabrski.
1811 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1812 Patch from Simon Arlott.
1814 PP/09 New lookups functionality failed to compile on old gcc which rejects
1815 extern declarations in function scope.
1816 Patch from Oliver Fleischmann
1818 PP/10 Use sig_atomic_t for flags set from signal handlers.
1819 Check getgroups() return and improve debugging.
1820 Fixed developed for diagnosis in bug 927 (which turned out to be
1823 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1824 Patch from Mark Zealey.
1826 PP/12 Bugzilla 1056: Improved spamd server selection.
1827 Patch from Mark Zealey.
1829 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1830 Based on patch from Heiko Schlittermann.
1832 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1833 Patch from Uwe Doering, sign-off by Michael Haardt.
1835 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1836 variable declaration deep within a block. Bug and patch from
1839 PP/15 lookups-Makefile IRIX compatibility coercion.
1841 PP/16 Make DISABLE_DKIM build knob functional.
1843 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1844 Patch by Simon Arlott
1846 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1847 variable argument macros. Our copy now differs from upstream.
1853 TF/01 Failure to get a lock on a hints database can have serious
1854 consequences so log it to the panic log.
1856 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1857 controlled using the smtp_confirmation log selector.
1859 TF/03 Include the error message when we fail to unlink a spool file.
1861 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1862 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1863 for maintaining out-of-tree patches for some time.
1865 PP/01 Bugzilla 139: Documentation and portability issues.
1866 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1867 Handle per-OS dynamic-module compilation flags.
1869 PP/02 Let /dev/null have normal permissions.
1870 The 4.73 fixes were a little too stringent and complained about the
1871 permissions on /dev/null. Exempt it from some checks.
1872 Reported by Andreas M. Kirchwitz.
1874 PP/03 Report version information for many libraries, including
1875 Exim version information for dynamically loaded libraries. Created
1876 version.h, now support a version extension string for distributors
1877 who patch heavily. Dynamic module ABI change.
1879 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1880 privilege escalation vulnerability whereby the Exim run-time user
1881 can cause root to append content of the attacker's choosing to
1884 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1887 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1888 If dropping privileges for untrusted macros, we disabled normal logging
1889 on the basis that it would fail; for the Exim run-time user, this is not
1890 the case, and it resulted in successful deliveries going unlogged.
1891 Fixed. Reported by Andreas Metzler.
1897 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1898 only prepend for the Resent-* case. Fixes regression introduced in
1899 Exim 4.70 by NM/22 for Bugzilla 607.
1901 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1902 increasing numbers of administrators be bitten by this.
1904 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1906 PP/03 Bugzilla 994: added openssl_options main configuration option.
1908 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1910 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1912 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1914 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1915 without a peer certificate, leading to a segfault because of an
1916 assumption that peers always have certificates. Be a little more
1917 paranoid. Problem reported by Martin Tscholak.
1919 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1920 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1921 NB: ClamAV planning to remove STREAM in "middle of 2010".
1922 CL also introduces -bmalware, various -d+acl logging additions and
1923 more caution in buffer sizes.
1925 PP/09 Implemented reverse_ip expansion operator.
1927 PP/10 Bugzilla 937: provide a "debug" ACL control.
1929 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1931 PP/12 Bugzilla 973: Implement --version.
1933 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1935 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1937 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1939 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1940 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1941 ignore trailing whitespace.
1943 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1945 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1948 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1949 Notification from Dr Andrew Aitchison.
1951 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1952 ExtendedDetectionInfo response format.
1953 Notification from John Horne.
1955 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1958 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1959 XSL and documented dependency on system catalogs, with examples of how
1962 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1965 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1966 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1967 configuration file which is writeable by the Exim user or group.
1969 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1970 of configuration files to cover files specified with the -C option if
1971 they are going to be used with root privileges, not just the default
1974 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1975 option (effectively making it always true).
1977 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1978 files to be used while preserving root privileges.
1980 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1981 that rogue child processes cannot use them.
1983 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1984 run-time user, instead of root.
1986 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1987 Exim run-time user without dropping privileges.
1989 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1990 result string, instead of calling string_vformat() twice with the same
1993 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1994 for other users. Others should always drop root privileges if they use
1995 -C on the command line, even for a whitelisted configure file.
1997 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1999 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2005 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2006 $data_path, and $header_path variables; fixed documentation bugs and
2009 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2010 exipick to access non-standard spools, including the "frozen" queue
2013 NM/01 Bugzilla 965: Support mysql stored procedures.
2014 Patch from Alain Williams
2016 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2018 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2019 Patch from Andreas Metzler
2021 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2022 Patch from Kirill Miazine
2024 NM/05 Bugzilla 671: Added umask to procmail example.
2026 JJ/03 installed exipick 20100323.0, fixing doc bug
2028 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2029 directory. Notification and patch from Dan Rosenberg.
2031 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2033 TK/02 Improve log output when DKIM signing operation fails.
2035 MH/01 Treat the transport option dkim_domain as a colon separated
2036 list, not as a single string, and sign the message with each element,
2037 omitting multiple occurences of the same signer.
2039 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2040 Bugzilla 985, 986. Patch by Simon Arlott
2042 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2043 Patch by Simon Arlott
2045 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2046 MBX locking. Notification from Dan Rosenberg.
2052 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2054 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2056 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2058 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2060 NM/04 Bugzilla 915: Flag broken perl installation during build.
2066 TK/01 Added patch by Johannes Berg that expands the main option
2067 "spamd_address" if it starts with a dollar sign.
2069 TK/02 Write list of recipients to X-Envelope-Sender header when building
2070 the mbox-format spool file for content scanning (suggested by Jakob
2073 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2074 (http://www.dcc-servers.net/) support via dccifd. Activated by
2075 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2077 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2078 by Mark Daniel Reidel <mr@df.eu>.
2080 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2081 When building exim an external PCRE library is now needed -
2082 PCRE is a system library on the majority of modern systems.
2083 See entry on PCRE_LIBS in EDITME file.
2085 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2086 conversation. Added nologin parameter to request.
2087 Patch contributed by Kirill Miazine.
2089 TF/01 Do not log submission mode rewrites if they do not change the address.
2091 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2093 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2094 log files in place. Contributed by Roberto Lima.
2096 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2098 TF/03 Bugzilla 615: When checking the local_parts router precondition
2099 after a local_part_suffix or local_part_prefix option, Exim now
2100 does not use the address's named list lookup cache, since this
2101 contains cached lookups for the whole local part.
2103 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2104 Robert Millan. Documentation is in experimental-spec.txt.
2106 TF/04 Bugzilla 668: Fix parallel build (make -j).
2108 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2110 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2111 Patch provided by Jan Srzednicki.
2113 TF/05 Leading white space used to be stripped from $spam_report which
2114 wrecked the formatting. Now it is preserved.
2116 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2117 that they are available at delivery time.
2119 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2121 TF/08 TLS error reporting now respects the incoming_interface and
2122 incoming_port log selectors.
2124 TF/09 Produce a more useful error message if an SMTP transport's hosts
2125 setting expands to an empty string.
2127 NM/06 Bugzilla 744: EXPN did not work under TLS.
2128 Patch provided by Phil Pennock.
2130 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2131 Patch provided by Richard Godbee.
2133 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2134 acl_smtp_notquit, added index entry.
2136 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2137 Patch provided by Eugene Bujak.
2139 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2140 accept(). Patch provided by Maxim Dounin.
2142 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2143 Patch provided by Phil Pennock.
2145 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2147 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2148 Patch provided by Brad "anomie" Jorsch.
2150 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2151 Patch provided by Dean Brooks.
2153 TK/05 Add native DKIM support (does not depend on external libraries).
2155 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2156 Patch provided by Graeme Fowler.
2158 NM/16 Bugzilla 851: Documentation example syntax fix.
2160 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2162 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2165 NM/19 Bugzilla 745: TLS version reporting.
2166 Patch provided by Phil Pennock.
2168 NM/20 Bugzilla 167: bool: condition support.
2169 Patch provided by Phil Pennock.
2171 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2172 clients. Patch provided by Phil Pennock.
2174 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2175 Patch provided by Brad "anomie" Jorsch.
2177 NM/23 Bugzilla 687: Fix misparses in eximstats.
2178 Patch provided by Heiko Schlittermann.
2180 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2181 Patch provided by Heiko Schlittermann.
2183 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2184 plus update to original patch.
2186 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2188 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2189 Patch provided by David Brownlee.
2191 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2193 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2195 NM/30 Bugzilla 888: TLS documentation bugfixes.
2197 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2199 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2200 Unlike the original bugzilla I have changed all shell scripts in src tree.
2202 NM/33 Bugzilla 898: Transport filter timeout fix.
2203 Patch by Todd Rinaldo.
2205 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2206 Patch by Serge Demonchaux.
2208 NM/35 Bugzilla 39: Base64 decode bug fixes.
2209 Patch by Jakob Hirsch.
2211 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2213 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2215 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2217 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2223 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2224 ALT-N's libdkim that I have put here:
2225 http://duncanthrax.net/exim-experimental/
2227 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2228 were called 'true' and it seems that C99 defines that as a reserved
2229 keyword to be used with 'bool' variable types. That means you could
2230 not include C99-style headers which use bools without triggering
2231 build errors in sieve.c.
2233 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2234 as mailq or other aliases. Changed the --help handling significantly
2235 to do whats expected. exim_usage() emits usage/help information.
2237 SC/01 Added the -bylocaldomain option to eximstats.
2239 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2241 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2243 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2249 PH/01 Another patch from the Sieve maintainer.
2251 PH/02 When an IPv6 address is converted to a string for single-key lookup
2252 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2253 dots are used instead of colons so that keys in lsearch files need not
2254 contain colons. This was done some time before quoting was made available
2255 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2256 (notated using the quote facility) so as to distinguish them from IPv4
2257 keys. This meant that lookups for IP addresses in host lists did not work
2258 for iplsearch lookups.
2260 This has been fixed by arranging for IPv6 addresses to be expressed with
2261 colons if the lookup type is iplsearch. This is not incompatible, because
2262 previously such lookups could never work.
2264 The situation is now rather anomalous, since one *can* have colons in
2265 ordinary lsearch keys. However, making the change in all cases is
2266 incompatible and would probably break a number of configurations.
2268 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2271 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2272 conversion specification without a maximum field width, thereby enabling
2273 a rogue spamd server to cause a buffer overflow. While nobody in their
2274 right mind would setup Exim to query an untrusted spamd server, an
2275 attacker that gains access to a server running spamd could potentially
2276 exploit this vulnerability to run arbitrary code as the Exim user.
2278 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2279 $primary_hostname instead of what libspf2 thinks the hosts name is.
2281 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2282 a directory entry by the name of the lookup key. Previously, if a
2283 symlink pointed to a non-existing file or a file in a directory that
2284 Exim lacked permissions to read, a lookup for a key matching that
2285 symlink would fail. Now it is enough that a matching directory entry
2286 exists, symlink or not. (Bugzilla 503.)
2288 PH/03 The body_linecount and body_zerocount variables are now exported in the
2291 PH/04 Added the $dnslist_matched variable.
2293 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2294 This means they are set thereafter only if the connection becomes
2297 PH/06 Added the client_condition to authenticators so that some can be skipped
2298 by clients under certain conditions.
2300 PH/07 The error message for a badly-placed control=no_multiline_responses left
2301 "_responses" off the end of the name.
2303 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2305 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2306 (without spaces) instead of just copying the configuration text.
2308 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2310 PH/11 Added $max_received_linelength.
2312 PH/12 Added +ignore_defer and +include_defer to host lists.
2314 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2315 way in which PCRE > 7.0 is built.
2317 PH/14 Implemented queue_only_load_latch.
2319 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2320 MAIL command. The effect was to mangle the value on 64-bit systems.
2322 PH/16 Another patch from the Sieve maintainer.
2324 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2326 PH/18 If a system quota error occurred while trying to create the file for
2327 a maildir delivery, the message "Mailbox is full" was not appended to the
2328 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2329 only to a quota excession during the actual writing of the file.
2331 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2332 characters?) which causes problems in log lines. The DN values are now
2333 passed through string_printing() before being added to log lines.
2335 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2336 and InterBase are left for another time.)
2338 PH/21 Added message_body_newlines option.
2340 PH/22 Guard against possible overflow in moan_check_errorcopy().
2342 PH/23 POSIX allows open() to be a macro; guard against that.
2344 PH/24 If the recipient of an error message contained an @ in the local part
2345 (suitably quoted, of course), incorrect values were put in $domain and
2346 $local_part during the evaluation of errors_copy.
2352 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2353 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2356 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2357 issue a MAIL command.
2359 PH/02 In an ACL statement such as
2361 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2363 if a client was not listed at all, or was listed with a value other than
2364 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2365 the condition was not true (as it should be), so access was not denied.
2366 The bug was that the ! inversion was incorrectly passed on to the second
2367 item. This has been fixed.
2369 PH/03 Added additional dnslists conditions == and =& which are different from
2370 = and & when the dns lookup returns more than one IP address.
2372 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2373 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2375 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2376 FSYNC, which compiles an option called disable_fsync that allows for
2377 bypassing fsync(). The documentation is heavily laced with warnings.
2379 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2381 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2382 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2383 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2384 including adding "make clean"; (3) Added -fPIC when compiling the test
2385 dynamically loaded module, to get rid of a warning.
2387 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2388 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2389 = 0s. The bug is otherwise harmless.
2391 PH/07 There was a bug in the dovecot authenticator such that the value of
2392 $auth1 could be overwritten, and so not correctly preserved, after a
2393 successful authentication. This usually meant that the value preserved by
2394 the server_setid option was incorrect.
2396 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2398 PH/09 Installed PCRE release 7.0.
2400 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2401 run for batched SMTP input. It is now run at the start of every message
2402 in the batch. While fixing this I discovered that the process information
2403 (output by running exiwhat) was not always getting set for -bs and -bS
2404 input. This is fixed, and it now also says "batched" for BSMTP.
2406 PH/11 Added control=no_pipelining.
2408 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2409 patch, slightly modified), and move the expansion of helo_data till after
2410 the connection is made in the smtp transport (so it can use these
2413 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2415 PH/14 Added log_selector = +pid.
2417 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2419 PH/16 Add ${if forany and ${if forall.
2421 PH/17 Added dsn_from option to vary the From: line in DSNs.
2423 PH/18 Flush SMTP output before performing a callout, unless control =
2424 no_callout_flush is set.
2426 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2427 was true (the default) a successful delivery failed to delete the retry
2428 item, thus causing premature timeout of the address. The bug is now
2431 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2433 PH/21 Long custom messages for fakedefer and fakereject are now split up
2434 into multiline responses in the same way that messages for "deny" and
2435 other ACL rejections are.
2437 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2438 with slight modification.
2440 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2441 draft, changing the syntax and factoring some duplicate code".
2443 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2444 for deliveries of the second and subsequent messages over the same SMTP
2447 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2448 ${reduce, with only minor "tidies".
2450 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2452 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2453 expansion side effects.
2455 PH/27 When a message times out after an over-quota error from an Exim-imposed
2456 quota, the bounce message says "mailbox is full". This message was not
2457 being given when it was a system quota that was exceeded. It now should
2460 MH/03 Made $recipients available in local_scan(). local_scan() already has
2461 better access to the recipient list through recipients_list[], but
2462 $recipients can be useful in postmaster-provided expansion strings.
2464 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2465 in the case of a MAIL command with additional options following the
2466 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2467 were accidentally chopped off.
2469 PH/29 SMTP synchronization checks are implemented when a command is read -
2470 there is a check that no more input is waiting when there shouldn't be
2471 any. However, for some commands, a delay in an ACL can mean that it is
2472 some time before the response is written. In this time, more input might
2473 arrive, invalidly. So now there are extra checks after an ACL has run for
2474 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2475 pipelining has not been advertised.
2477 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2479 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2480 returned in the SMTP response when smtp_return_error_details was false.
2481 This has been fixed.
2483 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2484 instead of the C I/O that was originally supplied, because problems were
2485 reported on Solaris.
2487 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2488 Exim which did not show up earlier: it was assuming that a call to
2489 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2490 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2491 was a macro that became an assignment, so it seemed to work. This has
2492 changed to a proper function call with a void return, hence the compile
2493 error. Exim's code has been fixed.
2495 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2498 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2500 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2502 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2503 because it turns out to be too big or there is a timeout) there is no
2504 "Completed" line in the log. When some messages of this type were
2505 selected by exigrep, they were listed as "not completed". Others were
2506 picked up by some special patterns. I have improved the selection
2507 criteria to be more general.
2509 PH/38 The host_find_failed option in the manualroute router can now be set
2510 to "ignore", to completely ignore a host whose IP address cannot be
2511 found. If all hosts are ignored, the behaviour is controlled by the new
2512 host_all_ignored option.
2514 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2515 homing or because of multiple MX records with /mx) generated more than
2516 one IP address, and the following item turned out to be the local host,
2517 all the secondary addresses of the first item were incorrectly removed
2518 from the list, along with the local host and any following hosts (which
2519 is what is supposed to happen).
2521 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2522 whoever called Exim into the -H file. In the case of the daemon it was
2523 behaving confusingly. When first started, it used values for whoever
2524 started the daemon, but after a SIGHUP it used the Exim user (because it
2525 calls itself on a restart). I have changed the code so that it now always
2528 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2529 message are rejected with the same error (e.g. no authentication or bad
2530 sender address), and a DATA command is nevertheless sent (as can happen
2531 with PIPELINING or a stupid MUA), the error message that was given to the
2532 RCPT commands is included in the rejection of the DATA command. This is
2533 intended to be helpful for MUAs that show only the final error to their
2536 PH/42 Another patch from the Sieve maintainer.
2538 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2539 Eximstats - Fixed some broken HTML links and added missing column headers
2541 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2542 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2544 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2546 PH/43 Yet another patch from the Sieve maintainer.
2548 PH/44 I found a way to check for a TCP/IP connection going away before sending
2549 the response to the final '.' that terminates a message, but only in the
2550 case where the client has not sent further data following the '.'
2551 (unfortunately, this is allowed). However, in many cases there won't be
2552 any further data because there won't be any more messages to send. A call
2553 to select() can be used: if it shows that the input is "ready", there is
2554 either input waiting, or the socket has been closed. An attempt to read
2555 the next input character can distinguish the two cases. Previously, Exim
2556 would have sent an OK response which the client would never have see.
2557 This could lead to message repetition. This fix should cure that, at
2558 least in a lot of common cases.
2560 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2561 advertised in response to EHLO.
2567 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2568 fixed by 4.65/MH/01 (is this a record?) are fixed:
2570 (i) An empty string was always treated as zero by the numeric comparison
2571 operators. This behaviour has been restored.
2573 (ii) It is documented that the numeric comparison operators always treat
2574 their arguments as decimal numbers. This was broken in that numbers
2575 starting with 0 were being interpreted as octal.
2577 While fixing these problems I realized that there was another issue that
2578 hadn't been noticed. Values of message_size_limit (both the global option
2579 and the transport option) were treated as octal if they started with 0.
2580 The documentation was vague. These values are now always treated as
2581 decimal, and I will make that clear in the documentation.
2587 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2588 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2591 MH/01 Don't check that the operands of numeric comparison operators are
2592 integers when their expansion is in "skipping" mode (fixes bug
2593 introduced by 4.64-PH/07).
2595 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2596 child addresses, Exim now panics and dies. Previously, because the count
2597 is held in a short int, deliveries were likely to be lost. As such a
2598 large number of recipients for a single message is ridiculous
2599 (performance will be very, very poor), I have chosen to impose a limit
2600 rather than extend the field.
2606 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2607 leftover -K file (the existence of which was triggered by #402).
2608 While we were at it, introduced process PID as part of the -K
2609 filename. This should rule out race conditions when creating
2612 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2613 processing considerably. Previous code took too long for large mails,
2614 triggering a timeout which in turn triggers #401.
2616 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2617 in the DK code in transports.c. sendfile() is not really portable,
2618 hence the _LINUX specificness.
2620 TF/01 In the add_headers option to the mail command in an Exim filter,
2621 there was a bug that Exim would claim a syntax error in any
2622 header after the first one which had an odd number of characters
2625 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2626 callout verification, Exim cached a "reject" for the entire domain. This
2627 is correct for most verifications, but it is not correct for a recipient
2628 verification with use_sender or use_postmaster set, because in that case
2629 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2630 case of MAIL FROM:<> rejection from other early rejections (e.g.
2631 rejection of HELO). When verifying a recipient using a non-null MAIL
2632 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2633 Whatever the result of the callout, the value of the domain cache is
2634 left unchanged (for any other kind of callout, getting as far as trying
2635 RCPT means that the domain itself is ok).
2637 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2640 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2641 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2642 session, the daemon ignores ECONNECT errors and logs others; it now
2643 ignores EPIPE as well.
2645 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2646 (quoted-printable decoding).
2648 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2649 later the small subsequent patch to fix an introduced bug.
2651 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2653 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2655 PH/08 An error is now given if message_size_limit is specified negative.
2657 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2658 to be given (somewhat) arbitrary names.
2660 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2663 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2664 miscellaneous code fixes
2666 PH/10 Added the log_reject_target ACL modifier to specify where to log
2669 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2670 hostname. This is wrong, because it relates to the incoming message (and
2671 probably the interface on which it is arriving) and not to the outgoing
2672 callout (which could be using a different interface). This has been
2673 changed to use the value of the helo_data option from the smtp transport
2674 instead - this is what is used when a message is actually being sent. If
2675 there is no remote transport (possible with a router that sets up host
2676 addresses), $smtp_active_hostname is used.
2678 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2679 tweaks were necessary in order to get it to work (see also 21 below):
2680 (a) The code assumed that strncpy() returns a negative number on buffer
2681 overflow, which isn't the case. Replaced with Exim's string_format()
2683 (b) There were several signed/unsigned issues. I just did the minimum
2684 hacking in of casts. There is scope for a larger refactoring.
2685 (c) The code used strcasecmp() which is not a standard C function.
2686 Replaced with Exim's strcmpic() function.
2687 (d) The code set only $1; it now sets $auth1 as well.
2688 (e) A simple test gave the error "authentication client didn't specify
2689 service in request". It would seem that Dovecot has changed its
2690 interface. Fortunately there's a specification; I followed it and
2691 changed what the client sends and it appears to be working now.
2693 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2696 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2697 address A is aliased to B and C, where B exists and C does not. Without
2698 -v the output is "A verified" because verification stops after a
2699 successful redirection if more than one address is generated. However,
2700 with -v the child addresses are also verified. Exim was outputting "A
2701 failed to verify" and then showing the successful verification for C,
2702 with its parentage. It now outputs "B failed to verify", showing B's
2703 parentage before showing the successful verification of C.
2705 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2706 look up a TXT record in a specific list after matching in a combined
2709 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2710 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2711 they consult the DNS. I had assumed they would set it the way they
2712 wanted; and indeed my experiments on Linux seem to show that in some
2713 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2714 To be on the safe side, however, I have now made the interface to
2715 host_find_byname() similar to host_find_bydns(), with an argument
2716 containing the DNS resolver options. The host_find_byname() function now
2717 sets these options at its start, just as host_find_bydns() does. The smtp
2718 transport options dns_qualify_single and dns_search_parents are passed to
2719 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2720 of host_find_byname() use the default settings of RES_DEFNAMES
2721 (qualify_single) but not RES_DNSRCH (search_parents).
2723 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2724 spool_read_header() do less string testing, by means of a preliminary
2725 switch on the second character of optional "-foo" lines. (This is
2726 overdue, caused by the large number of possibilities that now exist.
2727 Originally there were few.) While I was there, I also converted the
2728 str(n)cmp tests so they don't re-test the leading "-" and the first
2729 character, in the hope this might squeeze out yet more improvement.
2731 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2732 flag allowing group syntax was set by the header_syntax check but not
2733 turned off, possible causing trouble later; (2) The flag was not being
2734 set at all for the header_verify test, causing "group"-style headers to
2735 be rejected. I have now set it in this case, and also caused header_
2736 verify to ignore an empty address taken from a group. While doing this, I
2737 came across some other cases where the code for allowing group syntax
2738 while scanning a header line wasn't quite right (mostly, not resetting
2739 the flag correctly in the right place). These bugs could have caused
2740 trouble for malformed header lines. I hope it is now all correct.
2742 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2743 with the "reply" argument non-NULL. The code, however (which originally
2744 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2745 but it didn't always do it. This confused somebody who was copying the
2746 code for some other use. I have removed all the tests.
2748 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2749 feature that was used to support insecure browsers during the U.S. crypto
2750 embargo. It requires special client support, and Exim is probably the
2751 only MTA that supported it -- and would never use it because real RSA is
2752 always available. This code has been removed, because it had the bad
2753 effect of slowing Exim down by computing (never used) parameters for the
2754 RSA_EXPORT functionality.
2756 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2757 authenticator to fail if there's a tab character in the incoming data
2758 (there should never be unless someone is messing about, as it's supposed
2759 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2760 if the connection is using TLS or if the remote IP is the same as the
2761 local IP, and the "valid-client-cert option" if a client certificate has
2764 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2765 authenticators. This can be used for authorization after authentication
2766 succeeds. (In the case of plaintext, it servers for both authentication
2769 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2770 if any retry times were supplied.
2772 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2773 connection, where there is no client IP address to check. In this
2774 situation, the verify now always succeeds.
2776 PH/25 Applied John Jetmore's -Mset patch.
2778 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2780 PH/27 In a string expansion for a processed (not raw) header when multiple
2781 headers of the same name were present, leading whitespace was being
2782 removed from all of them, but trailing whitespace was being removed only
2783 from the last one. Now trailing whitespace is removed from each header
2784 before concatenation. Completely empty headers in a concatenation (as
2785 before) are ignored.
2787 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2788 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2790 PH/29 [Removed. This was a change that I later backed out, and forgot to
2791 correct the ChangeLog entry (that I had efficiently created) before
2792 committing the later change.]
2794 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2795 address errors (4xx response to RCPT) over the same connection as other
2796 messages routed to the same hosts. Such deliveries are always "forced",
2797 so retry times are not inspected. This resulted in far too many retries
2798 for the affected addresses. The effect occurred only when there were more
2799 hosts than the hosts_max_try setting in the smtp transport when it had
2800 the 4xx errors. Those hosts that it had tried were not added to the list
2801 of hosts for which the message was waiting, so if all were tried, there
2802 was no problem. Two fixes have been applied:
2804 (i) If there are any address or message errors in an SMTP delivery, none
2805 of the hosts (tried or untried) are now added to the list of hosts
2806 for which the message is waiting, so the message should not be a
2807 candidate for sending over the same connection that was used for a
2808 successful delivery of some other message. This seems entirely
2809 reasonable: after all the message is NOT "waiting for some host".
2810 This is so "obvious" that I'm not sure why it wasn't done
2811 previously. Hope I haven't missed anything, but it can't do any
2812 harm, as the worst effect is to miss an optimization.
2814 (ii) If, despite (i), such a delivery is accidentally attempted, the
2815 routing retry time is respected, so at least it doesn't keep
2816 hammering the server.
2818 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2819 in ${readsocket because some servers need this prod.
2821 PH/32 Added some extra debug output when updating a wait-xxx database.
2823 PH/33 The hint "could be header name not terminated by colon", which has been
2824 given for certain expansion errors for a long time, was not being given
2825 for the ${if def:h_colon_omitted{... case.
2827 PH/34 The spec says: "With one important exception, whenever a domain list is
2828 being scanned, $domain contains the subject domain." There was at least
2829 one case where this was not true.
2831 PH/35 The error "getsockname() failed: connection reset by peer" was being
2832 written to the panic log as well as the main log, but it isn't really
2833 panic-worthy as it just means the connection died rather early on. I have
2834 removed the panic log writing for the ECONNRESET error when getsockname()
2837 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2838 runs only) independently of the message's sender address. This meant
2839 that, if the 4xx error was in fact related to the sender, a different
2840 message to the same recipient with a different sender could confuse
2841 things. In particular, this can happen when sending to a greylisting
2842 server, but other circumstances could also provoke similar problems.
2843 I have changed the default so that the retry time for these errors is now
2844 based a combination of the sender and recipient addresses. This change
2845 can be overridden by setting address_retry_include_sender=false in the
2848 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2849 remote server are returned as part of bounce messages. This was not
2850 happening for LMTP over a pipe (the lmtp transport), but now it is the
2851 same for both kinds of LMTP.
2853 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2854 in header lines that were in fact CNAMEs. This is no longer the case.
2856 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2857 and queue runs started by the daemon processed all messages. This has
2858 been fixed so that -R and -S can now usefully be given with -q<time>.
2860 PH/40 Import PCRE release 6.7 (fixes some bugs).
2862 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2864 PH/42 Give an error if -q is specified more than once.
2866 PH/43 Renamed the variables $interface_address and $interface_port as
2867 $received_ip_address and $received_port, to make it clear that these
2868 values apply to message reception, and not to the outgoing interface when
2869 a message is delivered. (The old names remain recognized, of course.)
2871 PH/44 There was no timeout on the connect() call when using a Unix domain
2872 socket in the ${readsocket expansion. There now is.
2874 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2875 be meaningful with "accept".
2877 SC/01 Eximstats V1.43
2878 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2880 SC/02 Eximstats V1.44
2881 Use a glob alias rather than an array ref in the generated
2882 parser. This improves both readability and performance.
2884 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2885 Collect SpamAssassin and rejection statistics.
2886 Don't display local sender or destination tables unless
2887 there is data to show.
2888 Added average volumes into the top table text output.
2890 SC/04 Eximstats V1.46
2891 Collect data on the number of addresses (recipients)
2892 as well as the number of messages.
2894 SC/05 Eximstats V1.47
2895 Added 'Message too big' to the list of mail rejection
2896 reasons (thanks to Marco Gaiarin).
2898 SC/06 Eximstats V1.48
2899 Mainlog lines which have GMT offsets and are too short to
2900 have a flag are now skipped.
2902 SC/07 Eximstats V1.49 (Alain Williams)
2903 Added the -emptyok flag.
2905 SC/08 Eximstats V1.50
2906 Fixes for obtaining the IP address from reject messages.
2908 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2909 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2910 whitespace changes from 4.64-PH/27
2912 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2915 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2916 are found, allow negative numbers in numeric criteria)
2918 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2920 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2921 to match changes made in 4.64-PH/43
2923 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2925 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2926 "domain" is actually an IP address, and force a failure. This locks out
2927 those revolvers/nameservers that support "A-for-A" lookups, in
2928 contravention of the specifications.
2930 PH/48 When a host name was looked up from an IP address, and the subsequent
2931 forward lookup of the name timed out, the host name was left in
2932 $sender_host_name, contrary to the specification.
2934 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2935 restricted to single-key lookups, Exim was not diagnosing an error if
2936 * or *@ was used with a query-style lookup.
2938 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2940 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2941 long ago, but noone interested enough thought of it. Let's just say that
2942 the "1.1" means that there are some new functions that weren't there at
2943 some point in the past.
2945 PH/51 Error processing for expansion failure of helo_data from an smtp
2946 transport during callout processing was broken.
2948 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2949 tested/used via the -bh/-bhc/-bs options.
2951 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2952 bug, fixed in subsequent PCRE releases).
2954 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2955 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2957 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2963 SC/01 Use a glob alias rather than an array ref in eximstats generated
2964 parser. This improves both readability and performance.
2966 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2967 Don't display local sender or destination tables in eximstats unless
2968 there is data to show.
2969 Added average volumes into the eximstats top table text output.
2971 SC/03 Collect data on the number of addresses (recipients) as well
2972 as the number of messages in eximstats.
2974 TF/01 Correct an error in the documentation for the redirect router. Exim
2975 does (usually) call initgroups() when daemonizing.
2977 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2978 with consistent privilege compared to when running as a daemon.
2980 TF/03 Note in the spec that $authenticated_id is not set for local
2981 submissions from trusted users.
2983 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2984 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2986 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2987 by adding some example configuration directives to the default
2988 configuration file. A little bit of work is required to uncomment the
2989 directives and define how usernames and passwords are checked, but
2990 there is now a framework to start from.
2992 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2993 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2994 without this. I don't know how relevant this is to other LDAP libraries.
2996 PH/02 Add the verb name to the "unknown ACL verb" error.
2998 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3000 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3002 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3003 directory not expanded when it should be if an expanded home directory
3004 was set for the address (which is overridden by the transport).
3006 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3009 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3010 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3011 because it is too late at that time, and has no effect.
3013 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3014 security issue with \' (bugzilla #107). I could not use the
3015 PQescapeStringConn() function, because it needs a PGconn value as one of
3018 PH/08 When testing addresses using -bt, indicate those final addresses that
3019 are duplicates that would not cause an additional delivery. At least one
3020 person was confused, thinking that -bt output corresponded to deliveries.
3021 (Suppressing duplicates isn't a good idea as you lose the information
3022 about possibly different redirections that led to the duplicates.)
3024 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3025 systems where poll() doesn't work, in particular OS X.
3027 PH/10 Added more information to debugging output for retry time not reached.
3029 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3030 operations in malware.c.
3032 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3035 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3036 syslog_duplication set false, log lines that would normally be written
3037 both the the main log and to the reject log were not written to syslog at
3040 PH/14 In the default configuration, change the use of "message" in ACL warn
3041 statements to "add_header".
3043 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3044 not followed by a command (e.g. "seen endif").
3046 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3047 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3050 PH/17 Added extra conditions to the default value of delay_warning_condition
3054 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3055 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3056 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3059 The Auto-Submitted: and various List- headers are standardised, whereas I
3060 don't think Precedence: ever was.
3062 PH/18 Refactored debugging code in route_finduser() to show more information,
3063 in particular, the error code if getpwnam() issues one.
3065 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3066 This is apparently needed in addition to the PH/07 change above to avoid
3067 any possible encoding problems.
3069 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3070 but not after initializing Perl.
3072 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3073 output them only if debugging. By default they are written stderr,
3074 apparently, which is not desirable.
3076 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3079 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3082 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3084 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3085 authenticated or an ident call has been made. Suppress the default
3086 values for $authenticated_id and $authenticated_sender (but permit -oMai
3087 and -oMas) when testing with -bh.
3089 PH/24 Re-jigged the order of the tests in the default configuration so that the
3090 tests for valid domains and recipients precede the DNS black list and CSA
3091 tests, on the grounds that those ones are more expensive.
3093 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3094 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3095 This bug exists in every version of Exim that I still have, right back to
3098 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3099 However, an attempt to turn on case-sensitivity in a regex key by
3100 including (?-i) didn't work because the subject string was already
3101 lowercased, and the effects were non-intuitive. It turns out that a
3102 one-line patch can be used to allow (?-i) to work as expected.
3108 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3109 other effects) broke the use of negated acl sub-conditions.
3111 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3114 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3115 "Deny" causes Exim to reject the incoming connection with a 554 error.
3116 Unfortunately, if there is a major crisis, such as a disk failure,
3117 tcp-wrappers gives "deny", whereas what one would like would be some
3118 kind of temporary error. A kludge has been added to help with this.
3119 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3120 554 error is used if errno is still zero or contains ENOENT (which occurs
3121 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3124 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3126 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3127 errors. Otherwise a message that provokes a temporary error (when other
3128 messages do not) can cause a whole host to time out.
3130 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3131 addresses were routed directly to files or pipes from a redirect router.
3132 File deliveries just didn't batch; pipe deliveries might have suffered
3135 PH/06 A failure to get a lock for a hints database would erroneously always say
3136 "Failed to get write lock", even when it was really a read lock.
3138 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3139 of 0600. This has been changed to use the value of the lockfile_mode
3140 option (which defaults to 0600).
3142 PH/08 Applied small patch from the Sieve maintainer.
3144 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3145 folder from quota calculations, a direct delivery into this folder messed
3146 up the contents of the maildirsize file. This was because the regex was
3147 used only to exclude .Trash (or whatever) when the size of the mailbox
3148 was calculated. There was no check that a delivery was happening into an
3149 excluded directory. This bug has been fixed by ignoring all quota
3150 processing for deliveries into excluded directories.
3152 PH/10 Added the maildirfolder_create_regex option to appendfile.
3158 PH/01 The code for finding all the local interface addresses on a FreeBSD
3159 system running IPv6 was broken. This may well have applied to all BSD
3160 systems, as well as to others that have similar system calls. The broken
3161 code found IPv4 interfaces correctly, but gave incorrect values for the
3162 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3163 that it would not match correctly against @[] and not recognize the IPv6
3166 PH/02 The ipliteral router was not recognizing addresses of the form user@
3167 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3169 PH/03 Added disable_ipv6.
3171 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3172 decoded form, because it is most often used to construct To: headers
3173 lines in autoreplies, and the decoded form may well be syntactically
3174 invalid. However, $reply_address has leading white space removed, and all
3175 newlines turned into spaces so that the autoreply transport does not
3178 PH/05 If group was specified without a user on a router, and no group or user
3179 was specified on a transport, the group from the router was ignored.
3181 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3182 for visible compile-time settings that can be used to change these
3183 numbers, for those that want even more. Backwards compatibility with old
3184 spool files has been maintained. However, going back to a previous Exim
3185 release will lost any variables that are in spool files.
3187 PH/07 Two small changes when running in the test harness: increase delay when
3188 passing a TCP/IP connection to a new process, in case the original
3189 process has to generate a bounce, and remove special handling of
3190 127.0.0.2 (sic), which is no longer necessary.
3192 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3193 be the same on different OS.
3195 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3198 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3199 whether --show-vars was specified or not
3201 JJ/02 exipick: Added support for new ACL variable spool format introduced
3204 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3205 syntactically invalid From: or Reply-to: line, and a filter used this to
3206 generate an autoreply, and therefore failed to obtain an address for the
3207 autoreply, Exim could try to deliver to a non-existent relative file
3208 name, causing unrelated and misleading errors. What now happens is that
3209 it logs this as a hard delivery error, but does not attempt to create a
3212 PH/11 The exinext utility has a -C option for testing purposes, but although
3213 the given file was scanned by exinext itself; it wasn't being passed on
3214 when Exim was called.
3216 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3217 an end-of-file indication when reading a command response.
3219 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3220 compiled. In many other places in Exim, IPv6 addresses are always
3221 recognized, so I have changed this. It also means that IPv4 domain
3222 literals of the form [IPV4:n.n.n.n] are now always recognized.
3224 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3225 used if the router is not running as root, for example, when verifying at
3226 ACL time, or when using -bh. The debugging output from this situation was
3227 non-existent - all you got was a failure to exec. I have made two
3230 (a) Failures to set uid/gid, the current directory, or a process leader
3231 in a subprocess such as that created by queryprogram now generate
3232 suitable debugging output when -d is set.
3234 (b) The queryprogram router detects when it is not running as root,
3235 outputs suitable debugging information if -d is set, and then runs
3236 the subprocess without attempting to change uid/gid.
3238 PH/15 Minor change to Makefile for building test_host (undocumented testing
3241 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3242 additional section of a DNS packet that returns MX or SRV records.
3243 Instead, it always explicitly searches for A/AAAA records. This avoids
3244 major problems that occur when a DNS server includes only records of one
3245 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3246 fixed another bug: if SRV records were looked up and the corresponding
3247 address records were *not* found in the additional section, the port
3248 values from the SRV records were lost.
3250 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3251 using the correct key (the original address) when searching the retry
3252 rules in order to find which one to use for generating the retry hint.
3254 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3255 adding the default one. Similarly, if it contains a Reply-To: header, the
3256 errors_reply_to option, if set, is not used.
3258 PH/19 When calculating a retry time, Exim used to measure the "time since
3259 failure" by looking at the "first failed" field in the retry record. Now
3260 it does not use this if it is later than than the arrival time of the
3261 message. Instead it uses the arrival time. This makes for better
3262 behaviour in cases where some deliveries succeed, thus re-setting the
3263 "first failed" field. An example is a quota failure for a huge message
3264 when small messages continue to be delivered. Without this change, the
3265 "time since failure" will always be short, possible causing more frequent
3266 delivery attempts for the huge message than are intended.
3267 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3269 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3270 $1, $2, $3) because the numerical variables can be reset during some
3271 expansion items (e.g. "match"), thereby losing the authentication data.
3273 PH/21 Make -bV show the size of off_t variables so that the test suite can
3274 decide whether to run tests for quotas > 2G.
3276 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3277 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3278 filecount value is greater than 2G or if a quota value is greater than 2G
3279 on a system where the size of off_t is not greater than 4, a panic error
3282 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3283 never match. The debug and -bh output now contains an explicit error
3284 message indicating a malformed IPv4 address or mask.
3286 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3287 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3288 PH/23 above applies.
3290 PH/25 Do not write to syslog when running in the test harness. The only
3291 occasion when this arises is a failure to open the main or panic logs
3292 (for which there is an explicit test).
3294 PH/26 Added the /no_tell option to "control=freeze".
3296 PH/27 If a host name lookup failed very early in a connection, for example, if
3297 the IP address matched host_lookup and the reverse lookup yielded a name
3298 that did not have a forward lookup, an error message of the form "no IP
3299 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3300 could be logged. Now it outputs the IP address instead of "NULL".
3302 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3303 allows the sender and the authenticated sender to be set when
3304 submitting a message from within Exim. Since child_open_exim() is
3305 documented for local_scan(), the new function should be too.
3307 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3308 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3309 results in an empty string is now treated as unset.
3311 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3313 PH/31 Added sender_verify_fail logging option.
3315 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3316 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3317 tidied the source and removed it altogether.
3319 PH/33 When a queue run was abandoned because the load average was too high, a
3320 log line was always written; now it is written only if the queue_run log
3321 selector is set. In addition, the log line for abandonment now contains
3322 information about the queue run such as the pid. This is always present
3323 in "start" and "stop" lines but was omitted from the "abandon" line.
3325 PH/34 Omit spaces between a header name and the colon in the error message that
3326 is given when verify = headers_syntax fails (if there are lots of them,
3327 the message gets confusing).
3329 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3330 names, as there are now some PTR records that contain slashes. This check
3331 is only to protect against broken name servers that fall over on strange
3332 characters, so the fact that it applies to all lookups doesn't matter.
3334 PH/36 Now that the new test suite is complete, we can remove some of the
3335 special code in Exim that was needed for the old test suite. For example,
3336 sorting DNS records because real resolvers return them in an arbitrary
3337 order. The new test suite's fake resolver always returns records in the
3340 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3341 bounces) except when queue_only is set, to avoid logging races between
3342 the different processes.
3344 PH/38 Panic-die if .include specifies a non-absolute path.
3346 PH/39 A tweak to the "H" retry rule from its user.
3348 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3349 a label. They prevented compilation on older perls.
3351 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3352 a warning to be raised on newish perls.
3354 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3355 on queue. Changes to match documented behaviour of showing count of
3356 messages matching specified criteria.
3358 PH/40 Changed the default ident timeout from 30s to 5s.
3360 PH/41 Added support for the use of login_cap features, on those BSD systems
3361 that have them, for controlling the resources used by pipe deliveries.
3363 PH/42 The content-scanning code uses fopen() to create files in which to put
3364 message data. Previously it was not paying any attention to the mode of
3365 the files. Exim runs with umask(0) because the rest of the code creates
3366 files with open(), and sets the required mode explicitly. Thus, these
3367 files were ending up world-writeable. This was not a big issue, because,
3368 being within the spool directory, they were not world-accessible. I have
3369 created a function called modefopen, which takes an additional mode
3370 argument. It sets umask(777), creates the file, chmods it to the required
3371 mode, then resets the umask. All the relevant calls to fopen() in the
3372 content scanning code have been changed to use this function.
3374 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3375 to 24 hours. This avoids potential overflow problems when processing G
3376 and H retry rules. I suspect nobody ever tinkers with this value.
3378 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3380 PH/45 When the plaintext authenticator is running as a client, the server's
3381 challenges are checked to ensure they are valid base64 strings. By
3382 default, the authentication attempt is cancelled if an invalid string is
3383 received. Setting client_ignore_invalid_base64 true ignores these errors.
3384 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3385 they are received. Thus, the responses can be made to depend on the
3386 challenges. If an invalid string is ignored, an empty string is placed in
3389 PH/46 Messages that are created by the autoreply transport now contains a
3390 References: header, in accordance with RFCs 2822 and 3834.
3392 PH/47 Added authenticated_sender_force to the smtp transport.
3394 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3396 PH/49 Installed latest patch from the Sieve maintainer.
3398 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3399 was also set, the appendfile transport was unnecessarily scanning a
3400 directory of message files (e.g. for maildir delivery) to find the count
3401 of files (along with the size), even though it did not need this
3402 information. It now does the scan only if it needs to find either the
3403 size of the count of files.
3405 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3407 PH/52 Two bugs concerned with error handling when the smtp transport is
3410 (i) Exim was not creating retry information for temporary errors given
3411 for individual recipients after the DATA command when the smtp transport
3412 was used in LMTP mode. This meant that they could be retried too
3413 frequently, and not timed out correctly.
3415 (ii) Exim was setting the flag that allows error details to be returned
3416 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3417 recipients that were returned after the DATA command.
3419 PH/53 This is related to PH/52, but is more general: for any failing address,
3420 when detailed error information was permitted to be returned to the
3421 sender, but the error was temporary, then after the final timeout, only
3422 "retry timeout exceeded" was returned. Now it returns the full error as
3423 well as "retry timeout exceeded".
3425 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3426 do this, and (what is worse) MTAs that accept it.
3428 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3429 will now be deprecated.
3431 PH/56 New os.c-cygwin from the Cygwin maintainer.
3433 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3434 formats (previously only available in exim formats via -bpr, -bpru,
3435 and -bpra. Now also available in native and exiqgrep formats)
3437 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3438 with very large, slow to parse queues
3440 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3442 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3444 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3445 responses to authentication challenges, though it was showing the
3446 challenges; (ii) I've removed the CR characters from the debug output for
3449 PH/58 Allow for the insertion of a newline as well as a space when a string
3450 is turned into more than one encoded-word during RFC 2047 encoding. The
3451 Sieve code now uses this.
3453 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3454 data_4xx, lost_connection, tls_required.
3456 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3457 message was being sent as an SMTP response.
3459 PH/61 Add -l and -k options to exicyclog.
3461 PH/62 When verifying, if an address was redirected to one new address, so that
3462 verification continued, and the new address failed or deferred after
3463 having set something in $address_data, the value of $address_data was not
3464 passed back to the ACL. This was different to the case when no
3465 redirection occurred. The value is now passed back in both cases.
3467 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3468 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3469 use login_cap.h, so on its own it isn't the distinguishing feature. The
3470 new name refers directly to the setclassresources() function.
3472 PH/65 Added configuration files for NetBSD3.
3474 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3476 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3477 is preferred over IPv4.
3479 PH/68 The bounce_return_message and bounce_return_body options were not being
3480 honoured for bounces generated during the reception of non-SMTP messages.
3481 In particular, this applied to messages rejected by the ACL. This bug has
3482 been fixed. However, if bounce_return_message is true and bounce_return_
3483 body is false, the headers that are returned for a non-SMTP message
3484 include only those that have been read before the error was detected.
3485 (In the case of an ACL rejection, they have all been read.)
3487 PH/69 The HTML version of the specification is now built in a directory called
3488 spec_html instead of spec.html, because the latter looks like a path with
3489 a MIME-type, and this confuses some software.
3491 PH/70 Catch two compiler warnings in sieve.c.
3493 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3494 function verify_get_ident() calls ip_connect() to connect a socket, but
3495 if the "connect()" function timed out, ip_connect() used to close the
3496 socket. However, verify_get_ident() also closes the socket later, and in
3497 between Exim writes to the log, which may get opened at this point. When
3498 the socket was closed in ip_connect(), the log could get the same file
3499 descriptor number as the socket. This naturally causes chaos. The fix is
3500 not to close the socket in ip_connect(); the socket should be closed by
3501 the function that creates it. There was only one place in the code where
3502 this was missing, in the iplookup router, which I don't think anybody now
3503 uses, but I've fixed it anyway.
3505 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3506 well as to direct DNS lookups. Otherwise the handling of names in host
3507 lists is inconsistent and therefore confusing.
3513 PH/01 Two changes to the default runtime configuration:
3515 (1) Move the checks for relay_from_hosts and authenticated clients from
3516 after to before the (commented out) DNS black list checks.
3518 (2) Add control=submission to the relay_from_hosts and authenticated
3519 clients checks, on the grounds that messages accepted by these
3520 statements are most likely to be submissions.
3522 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3524 (1) Generate an error if the third argument for the ${prvs expansion is
3527 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3530 (3) Reset the variables that are obtained from the first argument of
3531 ${prvscheck and used in the second argument before leaving the code,
3532 because their memory is reclaimed, so using them afterwards may do
3535 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3536 one (it's much easier than Tom thought :-).
3538 (5) Because of (4), we can now allow for the use of $prvscheck_result
3539 inside the third argument.
3541 PH/03 For some reason, the default setting of PATH when running a command from
3542 a pipe transport was just "/usr/bin". I have changed it to
3545 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3546 anything to be listed in the output from -bV.
3548 PH/05 When a filter generated an autoreply, the entire To: header line was
3549 quoted in the delivery log line, like this:
3551 => >A.N.Other <ano@some.domain> <original@ddress> ...
3553 This has been changed so that it extracts the operative address. There
3554 may be more than one such address. If so, they are comma-separated, like
3557 => >ano@some.domain,ona@other.domain <original@ddress> ...
3559 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3560 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3561 not being looked up in the rDNS to get a host name, Exim was showing the
3562 IP address twice in Received: lines, even though the IP addresses were
3563 identical. For example:
3565 Received: from [1.2.3.4] (helo=[1.2.3.4])
3567 However, if the real host name was known, it was omitting the HELO data
3568 if it matched the actual IP address. This has been tidied up so that it
3569 doesn't show the same IP address twice.
3571 PH/07 When both +timestamp and +memory debugging was on, the value given by
3572 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3573 called by the debug printing, thereby overwriting the timestamp buffer.
3574 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3576 PH/08 When the original message was included in an autoreply transport, it
3577 always said "this is a copy of the message, including all the headers",
3578 even if body_only or headers_only was set. It now gives an appropriate
3581 PH/09 Applied a patch from the Sieve maintainer which:
3583 o fixes some comments
3584 o adds the (disabled) notify extension core
3585 o adds some debug output for the result of if/elsif tests
3586 o points to the current vacation draft in the documentation
3587 and documents the missing references header update
3591 o fixes a bug in processing the envelope test (when testing
3592 multiple envelope elements, the last element determined the
3595 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3596 Electronic Mail") by including:
3598 Auto-submitted: auto-generated
3600 in the messages that it generates (bounce messages and others, such as
3601 warnings). In the case of bounce messages for non-SMTP messages, there was
3602 also a typo: it was using "Auto_submitted" (underscore instead of
3603 hyphen). Since every message generated by Exim is necessarily in response
3604 to another message, thes have all been changed to:
3606 Auto-Submitted: auto-replied
3608 in accordance with these statements in the RFC:
3610 The auto-replied keyword:
3612 - SHOULD be used on messages sent in direct response to another
3613 message by an automatic process,
3615 - MUST NOT be used on manually-generated messages,
3617 - MAY be used on Delivery Status Notifications (DSNs) and Message
3618 Disposition Notifications (MDNs),
3620 - MUST NOT be used on messages generated by automatic or periodic
3621 processes, except for messages which are automatic responses to
3624 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3625 to the default Received: header definition.
3627 PH/12 Added log selector acl_warn_skipped (default on).
3629 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3630 variables because (a) they are in the wrong storage pool and (b) even if
3631 they were copied, it wouldn't work properly because of the caching.
3633 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3634 checking when decoding. Apparently there are clients that generate
3635 overlong encoded strings. Why am I not surprised?
3637 PH/15 If the first argument of "${if match_address" was not empty, but did not
3638 contain an "@" character, Exim crashed. Now it writes a panic log message
3639 and treats the condition as false.
3641 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3643 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3644 extension "envelope-auth". The code is finished and in agreement with
3645 other implementations, but there is no documentation so far and in fact,
3646 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3647 not changing the active code.
3649 Print executed "if" and "elsif" statements when debugging is used. This
3650 helps a great deal to understand what a filter does.
3652 Document more things not specified clearly in RFC3028. I had all this
3653 sorted out, when out of a sudden new issues came to my mind. Oops."
3655 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3658 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3659 canonical form (as documented). However, after a host name lookup from
3660 the IP address, check_host() was doing a simple string comparison with
3661 addresses acquired from the DNS when checking that the found name did
3662 have the original IP as one of its addresses. Since any found IPv6
3663 addresses are likely to be in abbreviated form, the comparison could
3664 fail. Luckily, there already exists a function for doing the comparison
3665 by converting both addresses to binary, so now that is used instead of
3666 the text comparison.
3668 PH/20 There was another similar case to PH/19, when a complete host name was
3669 given in a host list; looking up its IP address could give an abbreviated
3670 form, whereas the current host's name might or might not be abbreviated.
3671 The same fix has been applied.
3677 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3678 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3681 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3682 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3684 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3686 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3687 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3688 Also, for "Auto-Submitted", treat anything other than "no" as
3689 non-personal, in accordance with RFC 3834. (Previously it treated
3690 anything starting "auto-" as non-personal.)
3692 TF/01 The control=submission/name=... option had a problem with syntax
3693 errors if the name included a slash character. The /name= option
3694 now slurps the rest of the string, so it can include any characters
3695 but it must come last in the list of options (after /sender_retain
3698 PH/05 Some modifications to the interface to the fake nameserver for the new
3706 TK/01 Added the "success_on_redirect" address verification option. See
3707 NewStuff for rationale and an example.
3709 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3711 PH/02 Patch to exigrep to allow it to work on syslog lines.
3713 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3714 fread() to skip over the body file's header line, because in Cygwin the
3715 header line is locked and is inaccessible.
3717 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3718 co-exist for some time) to make it clear that it is the Exim ID that is
3719 referenced, not the Message-ID: header line.
3721 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3722 string_format() function, because snprintf() does not exist on all
3725 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3726 ${stat: expansion item.
3728 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3729 protocol synchronization error", to keep the pedants happy.
3731 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3732 well as for IRIX systems, when gcc is being used. See the host.c source
3735 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3737 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3740 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3741 transport and to the smtp transport in LMTP mode.
3743 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3745 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3746 run a filter in a subprocess. This could lead to confusion in subsequent
3747 lookups in the parent process. There should also be a search_tidyup() at
3748 the end of the subprocess.
3750 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3751 only if the host matched helo_try_verify_hosts, which caused the
3752 verification to occur when the EHLO/HELO command was issued. The ACL just
3753 tested the remembered result. Now, if a previous verification attempt has
3754 not happened, "verify = helo" does it there and then.
3756 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3758 TK/03 Fix log output including CR from clamd.
3760 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3761 exist provoked a memory error which could cause a segfault.
3763 PH/15 Installed PCRE 6.2
3765 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3767 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3768 of the problem. Specifically, suggested +O2 rather than +O1 for the
3771 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3773 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3774 record, the port was not correctly logged when the outgoing_port log
3775 selector was set (it logged the transort's default port).
3777 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3778 fallback_hosts, and "hosts" in the smtp transport.
3780 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3781 host errors such as "Connection refused".
3783 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3784 authentication with radiusclient 0.4.9:
3786 - Error returned from rc_read_config was caught wrongly
3787 - Username/password not passed on to radius server due to wrong length.
3789 The presumption is that some radiusclient API changes for 4.51/PH/17
3790 were not taken care of correctly. The code is still untested by me (my
3791 Linux distribution still has 0.3.2 of radiusclient), but it was
3792 contributed by a Radius user.
3794 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3795 expanding the "port" option of the smtp transport.
3797 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3798 while reading a MIME header. Thanks to Tom Hughes for a patch.
3800 PH/24 Include config.h inside local_scan.h so that configuration settings are
3803 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3804 that in an ACL for RCPT (for example), you can examine exactly what was
3807 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3808 commands, but it was not correctly comparing the address with the actual
3809 client host address. Thus, it would show the EHLO address in Received:
3810 header lines when this was not necessary.
3812 PH/27 Added the % operator to ${eval:}.
3814 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3815 it should be ignoring failures (because with -C, for example, it has lost
3816 privilege). It wasn't ignoring creation failures other than "already
3819 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3822 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3823 because an input line was too long, either on its own, or by virtue of
3824 too many continuations, the temporary file was not being removed, and the
3825 return code was incorrect.
3827 PH/31 Missing "BOOL" in function definition in filtertest.c.
3829 PH/32 Applied Sieve patches from the maintainer.
3831 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3833 PH/33 Added "verify = not_blind".
3835 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3836 Local/Makefile (with some defaults set). These are used in built scripts
3837 such as exicyclog, but they have never been used in the exim_install
3838 script (though there are many overriding facilities there). I have
3839 arranged that the exim_install script now takes note of these two
3842 PH/35 Installed configuration files for Dragonfly.
3844 PH/36 When a locally submitted message by a trusted user did not contain a
3845 From: header, and the sender address was obtained from -f or from an SMTP
3846 MAIL command, and the trusted user did not use -F to supply a sender
3847 name, $originator_name was incorrectly used when constructing a From:
3848 header. Furthermore, $originator_name was used for submission mode
3849 messages from external hosts without From: headers in a similar way,
3850 which is clearly wrong.
3852 PH/37 Added control=suppress_local_fixups.
3854 PH/38 When log_selector = +received_sender was set, and the addition of the
3855 sender made the log line's construction buffer exactly full, or one byte
3856 less than full, an overflow happened when the terminating "\n" was
3859 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3860 when the result of a list match is failure because a DNS lookup failed.
3862 PH/40 RM_COMMAND is now used in the building process.
3864 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3865 the "build-* directories that it finds.
3867 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3868 address in a domain literal was a prefix of an interface address.
3870 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3871 when verifying a sender address, unless rewrite_headers is false.
3873 PH/44 Wrote a long comment about why errors_to addresses are verified as
3874 recipients, not senders.
3876 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3877 the ratelimit ACL was added.
3879 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3881 PH/46 Added extra information about PostgreSQL errors to the error string.
3883 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3884 avoiding the need to install special zones in a real server. This is
3885 backwards compatible; if it can't find the fake resolver, it drops back.
3886 Thus, both old and new test suites can be run.
3888 TF/02 Added util/ratelimit.pl
3890 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3891 clock is set back in time.
3893 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3894 Candler <B.Candler@pobox.com>.
3896 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3897 OK for addresses that are the result of redirections.
3899 PH/48 A number of further additions for the benefit of the new test suite,
3900 including a fake gethostbyname() that interfaces to the fake DNS resolver
3903 TF/06 The fix for widen_domains has also been applied to qualify_single and
3904 search_parents which are the other dnslookup options that can cause
3907 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3910 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3912 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3913 service" if it encountered an MX record with an empty target hostname.
3914 The message is now "an MX or SRV record indicated no SMTP service".
3916 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3917 if the DNS of the sending site is misconfigured. This is quite a
3918 common situation. This change restores the behaviour of treating a
3919 helo verification defer as a failure.
3921 PH/51 If self=fail was set on a router, the bounce message did not include the
3922 actual error message.
3928 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3930 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3931 command itself ended in error, the underlying message about the transport
3932 filter timeout was being overwritten with the pipe command error. Now the
3933 underlying error message should be appended to the second error message.
3935 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3937 PH/02 Reduce the amount of output that "make" produces by default. Full output
3938 can still be requested.
3940 PH/03 The warning log line about a condition test deferring for a "warn" verb
3941 was being output only once per connection, rather than after each
3942 occurrence (because it was using the same function as for successful
3943 "warn" verbs). This seems wrong, so I have changed it.
3945 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3946 it should not have, which might have caused a crash in the right
3947 circumstances, but probably never did.
3949 PH/04 Installed a modified version of Tony Finch's patch to make submission
3950 mode fix the return path as well as the Sender: header line, and to
3951 add a /name= option so that you can make the user's friendly name appear
3954 TF/03 Added the control = fakedefer ACL modifier.
3956 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3957 Mark Lowes for thorough testing.
3959 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3961 TK/03 Merged latest SRS patch from Miles Wilton.
3963 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3964 with the definition in sysexits.h (which is #included earlier).
3965 Fortunately, Exim does not actually use EX_OK. The code used to try to
3966 preserve the sysexits.h value, by assuming that macro definitions were
3967 scanned for macro replacements. I have been disabused of this notion,
3968 so now the code just undefines EX_OK before #including unistd.h.
3970 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3971 in the smtp transport. When a block could not be written in a single
3972 write() function, the timeout was being re-applied to each part-write.
3973 This seems wrong - if the receiver was accepting one byte at a time it
3974 would take for ever. The timeout is now adjusted when this happens. It
3975 doesn't have to be particularly precise.
3977 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3978 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3980 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3981 without a domain if the check to <postmaster@domain> fails.
3983 SC/01 Eximstats: added -xls and the ability to specify output files
3984 (patch written by Frank Heydlauf).
3986 SC/02 Eximstats: use FileHandles for outputting results.
3988 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3990 SC/04 Eximstats: fixed display of large numbers with -nvr option
3992 SC/05 Eximstats: fixed merging of reports with empty tables.
3994 SC/06 Eximstats: added the -include_original_destination flag
3996 SC/07 Eximstats: removed tabs and trailing whitespace.
3998 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4000 TK/06 MBOX spool code: Add real "From " MBOX separator line
4001 so the .eml file is really in mbox format (even though
4002 most programs do not really care). Patch from Alex Miller.
4004 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4005 The latter is generated from $received_to and is only set if the
4006 message has one envelope recipient. SA can use these headers,
4007 obviously out-of-the-box. Patch from Alex Miller.
4009 PH/08 The ${def test on a variable was returning false if the variable's
4010 value was "0", contrary to what the specification has always said!
4011 The result should be true unless the variable is empty.
4013 PH/09 The syntax error of a character other than { following "${if
4014 def:variable_name" (after optional whitespace) was not being diagnosed.
4015 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4016 accidental colon was present, for example, could give incorrect results.
4018 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4019 result is used (not including appendfile, where other changes are about
4022 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4023 This involved changing a lot of size variables from int to off_t. It
4024 should work with maildirs and everything.
4026 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4027 spamd dying while we are connected to it.
4029 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4032 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4033 function for BDB 4.3.
4035 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4037 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4038 the sources for PCRE have been re-arranged and more files are now
4041 PH/15 The code I had for printing potentially long long variables in PH/11
4042 above was not the best (it lost precision). The length of off_t variables
4043 is now inspected at build time, and an appropriate printing format (%ld
4044 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4045 to be "long long int" or "long int". This is needed for the internal
4046 formatting function string_vformat().
4048 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4049 the configuration file to be ":syslog", then the script "guesses" where
4050 the logs files are, rather than using the compiled in default. In our
4051 case the guess is not the same as the compiled default, so the script
4052 suddenly stopped working when I started to use syslog. The patch checks
4053 to see if log_file_path is "". If so, it attempts to read it from exim
4054 with no configuration file to get the compiled in version, before it
4055 falls back to the previous guessing code."
4057 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4058 implementing BATV in an Exim configuration. See NewStuff for the gory
4061 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4062 Makefile that are specific to HP-UX.
4064 PH/18 If the "use_postmaster" option was set for a recipient callout together
4065 with the "random" option, the postmaster address was used as the MAIL
4066 FROM address for the random test, but not for the subsequent recipient
4067 test. It is now used for both.
4069 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4070 patch removes a few documentation additions to RFC 3028, because the
4071 latest draft now contains them. It adds the new en;ascii-case comparator
4072 and a new error check for 8bit text in MIME parts. Comparator and
4073 require names are now matched exactly. I enabled the subaddress
4074 extension, but it is not well tested yet (read: it works for me)."
4076 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4077 rework some of the code of TK/09 above to avoid the hardwired use of
4078 "%lld" and "long long". Replaced the call to snprintf() with a call to
4081 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4082 records point to non-existent hosts", "retry timeout exceeded", and
4083 "retry time not reached for any host after a long failure period".
4085 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4086 experimental DomainKeys support:
4088 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4089 (2) On an error such as an illegally used "control", the wrong name for
4090 the control was given.
4092 These problems did NOT occur unless DomainKeys support was compiled.
4094 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4096 PH/24 Added ${if match_ip condition.
4098 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4099 databases so that it will be absolutely obvious if a crash occurs in the
4100 DB library. This is a regular occurrence (often caused by mis-matched
4103 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4104 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4105 picked up on a user's system that detects such things. There doesn't seem
4106 to be a gcc warning option for this - only an attribute that has to be
4107 put on the function's prototype. It seems that in Fedora Core 4 they have
4108 set this on a number of new functions. No doubt there will be more in due
4111 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4112 specify a transport. However, if an address that was verified by such a
4113 router was the subject of a callout, Exim crashed because it tried to
4114 read the rcpt_include_affixes from the non-existent transport. Now it
4115 just assumes that the setting of that option is false. This bug was
4116 introduced by 4.51/PH/31.
4118 PH/28 Changed -d+all to exclude +memory, because that information is very
4119 rarely of interest, but it makes the output a lot bigger. People tend to
4120 do -d+all out of habit.
4122 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4123 code in os-type was giving problems when libc.so lives in lib64, like on
4126 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4127 aren't the modern standard, and it seems that some systems' include files
4128 don't always have them. Exim was already checking for some of the newer
4129 ones like T_AAAA, and defining it itself. I've added checks for all the
4130 record types that Exim uses.
4132 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4133 not automatically generating a new one, as it is supposed to. This
4134 prevented TLS from working. If the file did exist, but contained invalid
4135 data, a new version was generated, as expected. It was only the case of a
4136 non-existent file that was broken.
4138 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4139 with a change in libdomainkeys > 0.64.
4141 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4142 from DNS. If the selector record carries the flag, it now has
4143 precedence over the domain-wide flag.
4145 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4147 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4148 the use of a port name that isn't defined in /etc/services) occurred, the
4149 message was deferred as in a normal delivery, and thus remained on the
4150 spool, instead of being failed because of the mua_wrapper setting. This
4151 is now fixed, and I tidied up some of the mua_wrapper messages at the
4154 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4155 the messages in a hash of arrays rather than using individual hashes.
4156 This is a bit cleaner and results in dramatic memory savings, albeit
4157 at a slight CPU cost.
4159 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4160 as requested by Marc Sherman.
4162 SC/10 Eximstats: added histograms for user specified patterns as requested
4165 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4167 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4168 fopen() in the content-scanning modules that did not already have it.
4174 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4175 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4177 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4179 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4181 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4182 to test Sieve filters that use "vacation".
4184 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4185 that changes the way the GnuTLS parameters are stored in the cache file.
4186 The new format can be generated externally. For backward compatibility,
4187 if the data in the cache doesn't make sense, Exim assumes it has read an
4188 old-format file, and it generates new data and writes a new file. This
4189 means that you can't go back to an older release without removing the
4192 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4193 work if there are any delivery delays because "one_time" forces the
4194 parent to be marked "delivered", so its unseen clone is never tried
4195 again. For this reason, Exim now forbids the simultaneous setting of
4198 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4199 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4200 entry if you want to know the details. The fix, however, neglected to
4201 consider the case when local delivery batching is involved. The test for
4202 "previously delivered" was not happening when checking to see if an
4203 address could be batched with a previous (undelivered) one; under
4204 certain circumstances this could lead to multiple deliveries to the same
4207 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4208 in its include files, and this causes problems building Exim.
4210 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4211 header_syntax) but Exim was just ignoring anything given after a slash.
4212 In particular, this caused confusion with an attempt to use "verify =
4213 reverse_host_lookup/defer_ok". An error is now given when options are
4214 supplied for verify items that do not have them. (Maybe reverse_host_
4215 lookup should have a defer_ok option, but that's a different point.)
4217 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4218 defined by RFC 821) to 2048, because there were problems with some AUTH
4219 commands, and RFC 1869 says the size should be increased for extended
4220 SMTP commands that take arguments.
4222 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4225 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4226 "unknown" error; now it says that the functionality isn't in the binary.
4228 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4229 an address' error message when a string expansion fails (syntax or
4230 whatever). Otherwise the password may appear in the log. Following change
4231 PH/42 below, there is no longer a chance of it appearing in a bounce
4234 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4236 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4237 of its addresses was ever tried. (Bugzilla bug #2.)
4239 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4240 the result incorrectly in the debug output. (It correctly added a newline
4241 to what was transported.)
4243 TF/01 Added $received_time.
4245 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4246 commented out examples of how to interface to a virus scanner and to
4247 SpamAssassin. Also added commented examples of av_scanner and
4248 spamd_address settings.
4250 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4251 and controls are allowed in which ACLs. There were a couple of minor
4252 errors. Some of the entries in the conditions table (which is a table of
4253 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4254 negation of where the condition IS allowed.
4256 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4258 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4259 header file does not have a version number, so I've had to invent a new
4260 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4261 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4262 radiusclient), but it was contributed by a Radius user.
4264 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4265 files or directories, for OpenSSL.
4267 PH/19 When an Exim process that is running as root has to create an Exim log
4268 file, it does so in a subprocess that runs as exim:exim so as to get the
4269 ownership right at creation (otherwise, other Exim processes might see
4270 the file with the wrong ownership). There was no test for failure of this
4271 fork() call, which would lead to the process getting stuck as it waited
4272 for a non-existent subprocess. Forks do occasionally fail when resources
4273 run out. I reviewed all the other calls to fork(); they all seem to check
4276 PH/20 When checking for unexpected SMTP input at connect time (before writing
4277 the banner), Exim was not dealing correctly with a non-positive return
4278 from the read() function. If the client had disconnected by this time,
4279 the result was a log entry for a synchronization error with an empty
4280 string after "input=" when read() returned zero. If read() returned -1
4281 (an event I could not check), uninitialized data bytes were printed.
4282 There were reports of junk text (parts of files, etc) appearing after
4285 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4287 PH/22 Added support for macro redefinition, and (re)definition in between
4288 driver and ACL definitions.
4290 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4291 forgetting to use the resulting value; it was using the unexpanded value.
4293 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4294 hadn't been configured. The fix is from Juergen Kreileder, who
4295 understands it better than I do:
4297 "Here's what I see happening with three configured cyrus_sasl
4298 authenticators configured (plain, login, cram-md5):
4300 On startup auth_cyrus_sasl_init() gets called for each of these.
4301 This means three calls to sasl_listmech() without a specified mech_list.
4302 => SASL tests which mechs of all available mechs actually work
4303 => three warnings about OTP not working
4304 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4306 With the patch, sasl_listmech() also gets called three times. But now
4307 SASL's mech_list option is set to the server_mech specified in the the
4308 authenticator. Or in other words, the answer from sasl_listmech()
4309 gets limited to just the mech you're testing for (which is different
4311 => the return list contains just 'plain' or 'login', 'cram-md5' or
4312 nothing depending on the value of ob->server_mech.
4314 I've just tested the patch: Authentication still works fine,
4315 unavailable mechs specified in the exim configuration are still
4316 caught, and the auth.log warnings about OTP are gone."
4318 PH/25 When debugging is enabled, the contents of the command line are added
4319 to the debugging output, even when log_selector=+arguments is not
4322 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4323 answer is "GNU", and only if the return is "GNU/something" is the answer
4326 PH/27 $acl_verify_message is now set immediately after the failure of a
4327 verification in an ACL, and so is available in subsequent modifiers. In
4328 particular, the message can be preserved by coding like this:
4330 warn !verify = sender
4331 set acl_m0 = $acl_verify_message
4333 Previously, $acl_verify_message was set only while expanding "message"
4334 and "log_message" when a very denied access.
4336 PH/28 Modified OS/os.c-Linux with
4338 -#ifndef OS_LOAD_AVERAGE
4339 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4341 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4342 nomenclature these days.)
4344 PH/29 Installed patch from the Sieve maintainer that adds the options
4345 sieve_useraddress and sieve_subaddress to the redirect router.
4347 PH/30 In these circumstances:
4348 . Two addresses routed to the same list of hosts;
4349 . First host does not offer TLS;
4350 . First host accepts first address;
4351 . First host gives temporary error to second address;
4352 . Second host offers TLS and a TLS session is established;
4353 . Second host accepts second address.
4354 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4355 and peerdn, if requested) that were in fact used only for the second
4358 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4359 attention to any local part prefix or suffix that was matched by the
4360 router that accepted the address. It now behaves in the same way as it
4361 does for delivery: the affixes are removed from the local part unless
4362 rcpt_include_affixes is set on the transport.
4364 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4365 timeout during the DATA phase of an incoming message.
4367 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4368 applied a patch sanctioned by the Sieve maintainer.
4370 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4371 the uid or gid is negative. A case of a negative gid caused this to be
4372 noticed. The fix allows for either to be negative.
4374 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4375 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4378 PH/36 The change PH/12 above was broken. Fixed it.
4380 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4381 the grounds that routing the same address twice would always produce the
4382 same answer. This might have been true once, but it is certainly no
4383 longer true now. Routing a child address may depend on the previous
4384 routing that produced that child. Some complicated redirection strategies
4385 went wrong when messages had multiple recipients, and made Exim's
4386 behaviour dependent on the order in which the addresses were given.
4388 I have moved the duplicate checking until after the routing is complete.
4389 Exim scans the addresses that are assigned to local and remote
4390 transports, and removes any duplicates. This means that more work will be
4391 done, as duplicates will always all be routed, but duplicates are
4392 presumably rare, so I don't expect this is of any significance.
4394 For deliveries to pipes, files, and autoreplies, the duplicate checking
4395 still happens during the routing process, since they are not going to be
4398 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4399 It corrects a timeout issue with spamd. This is Ian's comment: "The
4400 background is that sometimes spamd either never reads data from a
4401 connection it has accepted, or it never writes response data. The exiscan
4402 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4403 blindly assumes that writes won't block so it may never time out."
4405 PH/39 Allow G after quota size as well as K and M.
4407 PH/40 The value set for $authenticated_id in an authenticator may not contain
4408 binary zeroes or newlines because the value is written to log lines and
4409 to spool files. There was no check on this. Now the value is run through
4410 the string_printing() function so that such characters are converted to
4411 printable escape sequences.
4413 PH/41 $message_linecount is a new variable that contains the total number of
4414 lines in the message. Compare $body_linecount, which is the count for the
4417 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4418 bounce and delay warning messages, except in certain special cases, which
4421 (a) An SMTP error message from a remote host;
4422 (b) A message specified in a :fail: redirection;
4423 (c) A message specified in a "fail" command in a system filter;
4424 (d) A message specified in a FAIL return from the queryprogram router;
4425 (e) A message specified by the cannot_route_message router option.
4427 In these cases only, Exim does include the error details in bounce and
4428 warning messages. There are also a few cases where bland messages such
4429 as "unrouteable address" or "local delivery error" are given.
4431 PH/43 $value is now also set for the "else" part of a ${run expansion.
4433 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4434 being worked on, but at least Exim now implements the latest version to
4437 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4438 process to complete was treated as a delivery failure, a timeout while
4439 writing the message to the pipe was logged, but erroneously treated as a
4440 successful delivery. Such timeouts include transport filter timeouts. For
4441 consistency with the overall process timeout, these timeouts are now
4442 treated as errors, giving rise to delivery failures by default. However,
4443 there is now a new Boolean option for the pipe transport called
4444 timeout_defer, which, if set TRUE, converts the failures into defers for
4445 both kinds of timeout. A transport filter timeout is now identified in
4448 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4449 systems where "make" and "gmake" are different, calling "gmake" at top
4450 level broke things. I've arranged for the value of $(MAKE) to be passed
4451 from the Makefile to this script so that it can call the same version of
4455 A note about Exim versions 4.44 and 4.50
4456 ----------------------------------------
4458 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4459 changes of various kinds. As a consequence, a big documentation update was
4460 needed. This delayed the release for rather longer than seemed good, especially
4461 in the light of a couple of (minor) security issues. Therefore, the changes
4462 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4463 release. So 4.44 and 4.50 are in effect two different branches that both start
4466 I have left the 4.50 change log unchanged; it contains all the changes since
4467 4.43. The change log for 4.44 is below; many of its items are identical to
4468 those for 4.50. This seems to be the most sensible way to preserve the
4469 historical information.
4475 1. Minor wording change to the doc/README.SIEVE file.
4477 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4478 computation of the current number of files was incorrect.
4480 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4481 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4484 4. Give more explanation in the error message when the command for a transport
4485 filter fails to execute.
4487 5. There are several places where Exim runs a non-Exim command in a
4488 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4489 was being done only for the command run by the queryprogram router. It is
4490 now done for all such subprocesses. The other cases are: ${run, transport
4491 filters, and the commands run by the lmtp and pipe transports.
4493 6. Added CONFIGURE_GROUP build-time option.
4495 7. Some older OS have a limit of 256 on the maximum number of file
4496 descriptors. Exim was using setrlimit() to set 1000 as a large value
4497 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4498 systems. I've change it so that if it can't get 1000, it tries for 256.
4500 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4501 was an oversight, and furthermore, ever since the addition of extra
4502 controls (e.g. 4.43/32), the checks on when to allow different forms of
4503 "control" were broken. There should now be diagnostics for all cases when a
4504 control that does not make sense is encountered.
4506 9. Added the /retain_sender option to "control=submission".
4508 10. $recipients is now available in the predata ACL (oversight).
4510 11. Tidy the search cache before the fork to do a delivery from a message
4511 received from the command line. Otherwise the child will trigger a lookup
4512 failure and thereby defer the delivery if it tries to use (for example) a
4513 cached ldap connection that the parent has called unbind on.
4515 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4516 of $address_data from the recipient verification was clobbered by the
4517 sender verification.
4519 13. The value of address_data from a sender verification is now available in
4520 $sender_address_data in subsequent conditions in the ACL statement.
4522 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4524 15. Added a new option "connect=<time>" to callout options, to set a different
4527 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4528 was its contents. (It was OK if the option was not defined at all.)
4530 17. A "Completed" log line is now written for messages that are removed from
4531 the spool by the -Mrm option.
4533 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4534 information about exactly what failed.
4536 19. Added -dd to debug only the daemon process.
4538 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4539 handles timeouts, both on the server side and network timeouts. Renamed the
4540 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4542 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4543 It is now set to "smtps".
4545 22. $host_address is now set to the target address during the checking of
4546 ignore_target_hosts.
4548 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4549 being passed; this would have caused $sender_host_name to have been used if
4550 matching the list had actually called for a host name (not very likely,
4551 since this list is usually IP addresses). A host name is now passed as
4554 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4555 code that specifies a non-restarting handler (typically sigaction() in
4556 modern systems) in an attempt to fix a rare and obscure crash bug.
4558 25. Narrowed the window for a race in the daemon that could cause it to ignore
4559 SIGCHLD signals. This is not a major problem, because they are used only to
4560 wake it up if nothing else does.
4562 26. A malformed maildirsize file could cause Exim to calculate negative values
4563 for the mailbox size or file count. Odd effects could occur as a result.
4564 The maildirsize information is now recalculated if the size or filecount
4567 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4568 support for a long time. Removed HAVE_SYS_VFS_H.
4570 28. Installed the latest version of exipick from John Jetmore.
4572 29. In an address list, if the pattern was not a regular expression, an empty
4573 subject address (from a bounce message) matched only if the pattern was an
4574 empty string. Non-empty patterns were not even tested. This was the wrong
4575 because it is perfectly reasonable to use an empty address as part of a
4576 database query. An empty address is now tested by patterns that are
4577 lookups. However, all the other forms of pattern expect the subject to
4578 contain a local part and a domain, and therefore, for them, an empty
4579 address still always fails if the pattern is not itself empty.
4581 30. Exim went into a mad DNS loop when attempting to do a callout where the
4582 host was specified on an smtp transport, and looking it up yielded more
4583 than one IP address.
4585 31. Re-factored the code for checking spool and log partition space into a
4586 function that finds that data and another that does the check. The former
4587 is then used to implement four new variables: $spool_space, $log_space,
4588 $spool_inodes, and $log_inodes.
4590 32. The RFC2047 encoding function was originally intended for short strings
4591 such as real names; it was not keeping to the 75-character limit for
4592 encoded words that the RFC imposes. It now respects the limit, and
4593 generates multiple encoded words if necessary. To be on the safe side, I
4594 have increased the buffer size for the ${rfc2047: expansion operator from
4597 33. It is now permitted to omit both strings after an "if" condition; if the
4598 condition is true, the result is "true". As before, when the second string
4599 is omitted, a false condition yields an empty string. This makes it less
4600 cumbersome to write custom ACL and router conditions.
4602 34. Failure to deliver a bounce message always caused it to be frozen, even if
4603 there was an errors_to setting on the router. The errors_to setting is now
4606 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4607 canonical form (fully expanded) before being placed in
4608 $sender_host_address.
4610 36. The table in the code that translates DNS record types into text (T_A to
4611 "A" for instance) was missing entries for NS and CNAME. It is just possible
4612 that this could have caused confusion if both these types were looked up
4613 for the same domain, because the text type is used as part of Exim's
4614 per-process caching. But the chance of anyone hitting this buglet seems
4617 37. The dnsdb lookup has been extended in a number of ways.
4619 (1) There is a new type, "zns", which walks up the domain tree until it
4620 finds some nameserver records. It should be used with care.
4622 (2) There is a new type, "mxh", which is like "mx" except that it returns
4623 just the host names, not the priorities.
4625 (3) It is now possible to give a list of domains (or IP addresses) to be
4626 looked up. The behaviour when one of the lookups defers can be
4627 controlled by a keyword.
4629 (4) It is now possible to specify the separator character for use when
4630 multiple records are returned.
4632 38. The dnslists ACL condition has been extended: it is now possible to supply
4633 a list of IP addresses and/or domains to be looked up in a particular DNS
4636 39. Added log_selector=+queue_time_overall.
4638 40. When running the queue in the test harness, wait just a tad after forking a
4639 delivery process, to get repeatability of debugging output.
4641 41. Include certificate and key file names in error message when GnuTLS fails
4642 to set them up, because the GnuTLS error message doesn't include the name
4643 of the failing file when there is a problem reading it.
4645 42. Allow both -bf and -bF in the same test run.
4647 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4649 44. The "Exiscan patch" is now merged into the mainline Exim source.
4651 45. Sometimes the final signoff response after QUIT could fail to get
4652 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4653 < 0 before doing a fflush(). This bug looks as though it goes back to the
4654 introduction of TLS in release 3.20, but "sometimes" must have been rare
4655 because the tests only now provoked it.
4657 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4658 (this can affect the format of dates).
4660 47. exim_tidydb, when checking for the continued existence of a message for
4661 which it has found a message-specific retry record, was not finding
4662 messages that were in split spool directories. Consequently, it was
4663 deleting retry records that should have stayed in existence.
4665 48. Steve fixed some bugs in eximstats.
4667 49. The SPA authentication driver was not abandoning authentication and moving
4668 on to the next authenticator when an expansion was forced to fail,
4669 contradicting the general specification for all authenticators. Instead it
4670 was generating a temporary error. It now behaves as specified.
4672 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4673 (the order specifies the preference for clients). The order is now AES256,
4674 AES128, 3DES, ARCFOUR128.
4676 51. Small patch to Sieve code - explicitly set From: when generating an
4679 52. Exim crashed if a remote delivery caused a very long error message to be
4680 recorded - for instance if somebody sent an entire SpamAssassin report back
4681 as a large number of 550 error lines. This bug was coincidentally fixed by
4682 increasing the size of one of Exim's internal buffers (big_buffer) that
4683 happened as part of the Exiscan merge. However, to be on the safe side, I
4684 have made the code more robust (and fixed the comments that describe what
4687 53. Now that there can be additional text after "Completed" in log lines (if
4688 the queue_time_overall log selector is set), a one-byte patch to exigrep
4689 was needed to allow it to recognize "Completed" as not the last thing in
4692 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4693 patch that reportedly fixes this has been added. I am not expert enough to
4694 create a test for it. This is what the patch creator wrote:
4696 "I found a little strange behaviour of ldap code when working with
4697 Windows 2003 AD Domain, where users was placed in more than one
4698 Organization Units. When I tried to give exim partial DN, the exit code
4699 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4700 But simultaneously result of request was absolutely normal ldap result,
4701 so I produce this patch..."
4703 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4704 REFERENCE, so I have modified the code to exclude the patch when that macro
4707 55. Some experimental protocols are using DNS PTR records for new purposes. The
4708 keys for these records are domain names, not reversed IP addresses. The
4709 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4710 leaves it alone. Component reversal etc. now happens only for IP addresses.
4713 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4715 57. Double the size of the debug message buffer (to 2048) so that more of very
4716 long debug lines gets shown.
4718 58. The exicyclog utility now does better if the number of log files to keep
4719 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4721 59. Two changes related to the smtp_active_hostname option:
4723 (1) $smtp_active_hostname is now available as a variable.
4724 (2) The default for smtp_banner uses $smtp_active_hostname instead
4725 of $primary_hostname.
4727 60. The host_aton() function is supposed to be passed a string that is known
4728 to be a valid IP address. However, in the case of IPv6 addresses, it was
4729 not checking this. This is a hostage to fortune. Exim now panics and dies
4730 if the condition is not met. A case was found where this could be provoked
4731 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4732 components; fortuitously, this particular loophole had already been fixed
4733 by change 4.50/55 above.
4735 If there are any other similar loopholes, the new check in host_aton()
4736 itself should stop them being exploited. The report I received stated that
4737 data on the command line could provoke the exploit when Exim was running as
4738 exim, but did not say which command line option was involved. All I could
4739 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4740 running as the user.
4743 61. There was a buffer overflow vulnerability in the SPA authentication code
4744 (which came originally from the Samba project). I have added a test to the
4745 spa_base64_to_bits() function which I hope fixes it.
4748 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4749 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4751 63. The daemon start-up calls getloadavg() while still root for those OS that
4752 need the first call to be done as root, but it missed one case: when
4753 deliver_queue_load_max is set with deliver_drop_privilege. This is
4754 necessary for the benefit of the queue runner, because there is no re-exec
4755 when deliver_drop_privilege is set.
4757 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4758 This has been fixed.
4760 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4761 was in use, was not putting the data itself into the right store pool;
4762 consequently, it could be overwritten for a subsequent message in the same
4763 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4766 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4768 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4769 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4770 the function were treating the return as a boolean value, which happened to
4771 work because 0=false and not-0=true, but is not correct code.
4773 68. The host_aton() function was not handling scoped IPv6 addresses (those
4774 with, for example, "%eth0" on the end) correctly.
4776 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4777 negated items (that is, ~something) in unsigned ints. Some compilers
4778 apparently mutter when there is no cast.
4780 70. If an address verification called from an ACL failed, and did not produce a
4781 user-specific message (i.e. there was only a "system" message), nothing was
4782 put in $acl_verify_message. In this situation, it now puts the system
4785 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4786 session; change 4.31/43 added the unwanted input to the log line - except
4787 that it did not do this in the start of session case. It now does.
4789 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4790 This is wrong and can cause the other end to generate a synchronization
4791 error if it is another Exim or anything else that does the synchronization
4792 check. A QUIT command is no longer sent after a timeout.
4794 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4795 during host lookups.
4797 74. The defer_ok option of callout verification was not working if it was used
4798 when verifying addresses in header lines, that is, for this case:
4800 verify = header_sender/callout=defer_ok
4802 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4803 those file descriptors could be used for SMTP connections. If anything
4804 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4805 could be sent to the SMTP client, causing chaos. The daemon now opens
4806 stdin, stdout, and stderr to /dev/null when it puts itself into the
4809 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4810 log by default. The user can override this with suitable Perl magic.
4812 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4813 the log message when discard triggers, was not working for the DATA ACL or
4814 for the non-SMTP ACL.
4816 78. Error message wording change in sieve.c.
4818 79. If smtp_accept_max_per_host was set, the number of connections could be
4819 restricted to fewer than expected, because the daemon was trying to set up
4820 a new connection before checking whether the processes handling previous
4821 connections had finished. The check for completed processes is now done
4822 earlier. On busy systems, this bug wouldn't be noticed because something
4823 else would have woken the daemon, and it would have reaped the completed
4826 80. If a message was submitted locally by a user whose login name contained one
4827 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4828 It caused a spool format error. I have fixed the spool reading code. A
4829 related problem was that the "from" clause in the Received: line became
4830 illegal because of the space(s). It is now covered by ${quote_local_part.
4832 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4835 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4837 83. Patch from Sieve maintainer to fix unterminated string problem in
4838 "vacation" handling.
4840 84. Some minor changes to the Linux configuration files to help with other
4841 OS variants using glibc.
4843 85. One more patch for Sieve to update vacation handling to latest spec.
4846 ----------------------------------------------------
4847 See the note above about the 4.44 and 4.50 releases.
4848 ----------------------------------------------------
4854 1. Change 4.43/35 introduced a bug that caused file counts to be
4855 incorrectly computed when quota_filecount was set in an appendfile
4858 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4859 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4862 3. Give more explanation in the error message when the command for a transport
4863 filter fails to execute.
4865 4. There are several places where Exim runs a non-Exim command in a
4866 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4867 was being done only for the command run by the queryprogram router. It is
4868 now done for all such subprocesses. The other cases are: ${run, transport
4869 filters, and the commands run by the lmtp and pipe transports.
4871 5. Some older OS have a limit of 256 on the maximum number of file
4872 descriptors. Exim was using setrlimit() to set 1000 as a large value
4873 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4874 systems. I've change it so that if it can't get 1000, it tries for 256.
4876 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4877 was an oversight, and furthermore, ever since the addition of extra
4878 controls (e.g. 4.43/32), the checks on when to allow different forms of
4879 "control" were broken. There should now be diagnostics for all cases when a
4880 control that does not make sense is encountered.
4882 7. $recipients is now available in the predata ACL (oversight).
4884 8. Tidy the search cache before the fork to do a delivery from a message
4885 received from the command line. Otherwise the child will trigger a lookup
4886 failure and thereby defer the delivery if it tries to use (for example) a
4887 cached ldap connection that the parent has called unbind on.
4889 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4890 of $address_data from the recipient verification was clobbered by the
4891 sender verification.
4893 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4894 was its contents. (It was OK if the option was not defined at all.)
4896 11. A "Completed" log line is now written for messages that are removed from
4897 the spool by the -Mrm option.
4899 12. $host_address is now set to the target address during the checking of
4900 ignore_target_hosts.
4902 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4903 being passed; this would have caused $sender_host_name to have been used if
4904 matching the list had actually called for a host name (not very likely,
4905 since this list is usually IP addresses). A host name is now passed as
4908 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4909 code that specifies a non-restarting handler (typically sigaction() in
4910 modern systems) in an attempt to fix a rare and obscure crash bug.
4912 15. Narrowed the window for a race in the daemon that could cause it to ignore
4913 SIGCHLD signals. This is not a major problem, because they are used only to
4914 wake it up if nothing else does.
4916 16. A malformed maildirsize file could cause Exim to calculate negative values
4917 for the mailbox size or file count. Odd effects could occur as a result.
4918 The maildirsize information is now recalculated if the size or filecount
4921 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4922 support for a long time. Removed HAVE_SYS_VFS_H.
4924 18. Updated exipick to current release from John Jetmore.
4926 19. Allow an empty sender to be matched against a lookup in an address list.
4927 Previously the only cases considered were a regular expression, or an
4930 20. Exim went into a mad DNS lookup loop when doing a callout where the
4931 host was specified on the transport, if the DNS lookup yielded more than
4934 21. The RFC2047 encoding function was originally intended for short strings
4935 such as real names; it was not keeping to the 75-character limit for
4936 encoded words that the RFC imposes. It now respects the limit, and
4937 generates multiple encoded words if necessary. To be on the safe side, I
4938 have increased the buffer size for the ${rfc2047: expansion operator from
4941 22. Failure to deliver a bounce message always caused it to be frozen, even if
4942 there was an errors_to setting on the router. The errors_to setting is now
4945 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4946 canonical form (fully expanded) before being placed in
4947 $sender_host_address.
4949 24. Updated eximstats to version 1.33
4951 25. Include certificate and key file names in error message when GnuTLS fails
4952 to set them up, because the GnuTLS error message doesn't include the name
4953 of the failing file when there is a problem reading it.
4955 26. Expand error message when OpenSSL has problems setting up cert/key files.
4958 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4959 (this can affect the format of dates).
4961 28. exim_tidydb, when checking for the continued existence of a message for
4962 which it has found a message-specific retry record, was not finding
4963 messages that were in split spool directories. Consequently, it was
4964 deleting retry records that should have stayed in existence.
4966 29. eximstats updated to version 1.35
4967 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4968 1.35 - bugfix such that pie charts by volume are generated correctly
4970 30. The SPA authentication driver was not abandoning authentication and moving
4971 on to the next authenticator when an expansion was forced to fail,
4972 contradicting the general specification for all authenticators. Instead it
4973 was generating a temporary error. It now behaves as specified.
4975 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4976 (the order specifies the preference for clients). The order is now AES256,
4977 AES128, 3DES, ARCFOUR128.
4979 31. Small patch to Sieve code - explicitly set From: when generating an
4982 32. Exim crashed if a remote delivery caused a very long error message to be
4983 recorded - for instance if somebody sent an entire SpamAssassin report back
4984 as a large number of 550 error lines. This bug was coincidentally fixed by
4985 increasing the size of one of Exim's internal buffers (big_buffer) that
4986 happened as part of the Exiscan merge. However, to be on the safe side, I
4987 have made the code more robust (and fixed the comments that describe what
4990 33. Some experimental protocols are using DNS PTR records for new purposes. The
4991 keys for these records are domain names, not reversed IP addresses. The
4992 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4993 leaves it alone. Component reversal etc. now happens only for IP addresses.
4996 34. The host_aton() function is supposed to be passed a string that is known
4997 to be a valid IP address. However, in the case of IPv6 addresses, it was
4998 not checking this. This is a hostage to fortune. Exim now panics and dies
4999 if the condition is not met. A case was found where this could be provoked
5000 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5001 components; fortuitously, this particular loophole had already been fixed
5002 by change 4.50/55 or 4.44/33 above.
5004 If there are any other similar loopholes, the new check in host_aton()
5005 itself should stop them being exploited. The report I received stated that
5006 data on the command line could provoke the exploit when Exim was running as
5007 exim, but did not say which command line option was involved. All I could
5008 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5009 running as the user.
5012 35. There was a buffer overflow vulnerability in the SPA authentication code
5013 (which came originally from the Samba project). I have added a test to the
5014 spa_base64_to_bits() function which I hope fixes it.
5017 36. The daemon start-up calls getloadavg() while still root for those OS that
5018 need the first call to be done as root, but it missed one case: when
5019 deliver_queue_load_max is set with deliver_drop_privilege. This is
5020 necessary for the benefit of the queue runner, because there is no re-exec
5021 when deliver_drop_privilege is set.
5023 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5024 was in use, was not putting the data itself into the right store pool;
5025 consequently, it could be overwritten for a subsequent message in the same
5026 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5029 38. Sometimes the final signoff response after QUIT could fail to get
5030 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5031 < 0 before doing a fflush(). This bug looks as though it goes back to the
5032 introduction of TLS in release 3.20, but "sometimes" must have been rare
5033 because the tests only now provoked it.
5039 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5040 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5041 Now it returns an integer. A number of calls were still expecting a T/F
5042 return. Fortuitously, in all cases, the tests worked in OK situations,
5043 which is the norm. However, things would have gone wrong on any write
5044 failures on the smtp file descriptor. This function is used when sending
5045 messages over SMTP and also when doing verify callouts.
5047 2. When Exim is called to do synchronous delivery of a locally submitted
5048 message (the -odf or -odi options), it no longer closes stderr before doing
5051 3. Implemented the mua_wrapper option.
5053 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5055 5. Implemented the functions header_remove(), header_testname(),
5056 header_add_at_position(), and receive_remove_recipient(), and exported them
5059 6. If an ACL "warn" statement specified the addition of headers, Exim already
5060 inserted X-ACL-Warn: at the start if there was no header name. However, it
5061 was not making this test for the second and subsequent header lines if
5062 there were newlines in the string. This meant that an invalid header could
5063 be inserted if Exim was badly configured.
5065 7. Allow an ACL "warn" statement to add header lines at the start or after all
5066 the Received: headers, as well as at the end.
5068 8. Added the rcpt_4xx retry error code.
5070 9. Added postmaster_mailfrom=xxx to callout verification option.
5072 10. Added mailfrom=xxxx to the callout verification option, for verify=
5075 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5076 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5078 12. Inserted some casts to stop certain compilers warning when using pointer
5079 differences as field lengths or precisions in printf-type calls (mostly
5080 affecting debugging statements).
5082 13. Added optional readline() support for -be (dynamically loaded).
5084 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5085 same clock tick as a message's arrival, so that its received time was the
5086 same as the "first fail" time on the retry record, and that message
5087 remained on the queue past the ultimate address timeout, every queue runner
5088 would try a delivery (because it was past the ultimate address timeout) but
5089 after another failure, the ultimate address timeout, which should have then
5090 bounced the address, did not kick in. This was a "< instead of <=" error;
5091 in most cases the first failure would have been in the next clock tick
5092 after the received time, and all would be well.
5094 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5095 being recognized when the domain list was tested by the match_domain
5096 condition in an expansion string.
5098 16. Added the ${str2b64: operator.
5100 17. Exim was always calling setrlimit() to set a large limit for the number of
5101 processes, without checking whether the existing limit was already
5102 adequate. (It did check for the limit on file descriptors.) Furthermore,
5103 errors from getrlimit() and setrlimit() were being ignored. Now they are
5104 logged to the main and panic logs, but Exim does carry on, to try to do its
5105 job under whatever limits there are.
5107 18. Imported PCRE 5.0.
5109 19. Trivial typo in log message " temporarily refused connection" (the leading
5112 20. If the log selector return_path_on_delivery was set and an address was
5113 redirected to /dev/null, the delivery process crashed because it assumed
5114 that a return path would always be set for a "successful" delivery. In this
5115 case, the whole delivery is bypassed as an optimization, and therefore no
5118 21. Internal re-arrangement: the function for sending a challenge and reading
5119 a response while authentication was assuming a zero-terminated challenge
5120 string. It's now changed to take a pointer and a length, to allow for
5121 binary data in such strings.
5123 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5125 23. Exim was not respecting finduser_retries when seeking the login of the
5126 uid under which it was called; it was always trying 10 times. (The default
5127 setting of finduser_retries is zero.) Also, it was sleeping after the final
5128 failure, which is pointless.
5130 24. Implemented tls_on_connect_ports.
5132 25. Implemented acl_smtp_predata.
5134 26. If the domain in control=submission is set empty, Exim assumes that the
5135 authenticated id is a complete email address when it generates From: or
5136 Sender: header lines.
5138 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5139 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5140 chown and chgrp in /bin and hostname in /usr/bin.
5142 28. Exim was keeping the "process log" file open after each use, just as it
5143 does for the main log. This opens the possibility of it remaining open for
5144 long periods when the USR1 signal hits a daemon. Occasional processlog
5145 errors were reported, that could have been caused by this. Anyway, it seems
5146 much more sensible not to leave this file open at all, so that is what now
5149 29. The long-running daemon process does not normally write to the log once it
5150 has entered its main loop, and it closes the log before doing so. This is
5151 so that log files can straightforwardly be renamed and moved. However,
5152 there are a couple of unusual error situations where the daemon does write
5153 log entries, and I had neglected to close the log afterwards.
5155 30. The text of an SMTP error response that was received during a remote
5156 delivery was being truncated at 512 bytes. This is too short for some of
5157 the long messages that one sometimes sees. I've increased the limit to
5160 31. It is now possible to make retry rules that apply only when a message has a
5161 specific sender, in particular, an empty sender.
5163 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5164 it possible to be selective about when SMTP synchronization is enforced.
5166 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5168 32. Implemented hosts_connection_nolog.
5170 33. Added an ACL for QUIT.
5172 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5175 35. Added mailbox_size and mailbox_filecount to appendfile.
5177 36. Added control = no_multiline_responses to ACLs.
5179 37. There was a bug in the logic of the code that waits for the clock to tick
5180 in the case where the clock went backwards by a substantial amount such
5181 that the microsecond fraction of "now" was more than the microsecond
5182 fraction of "then" (but the whole seconds number was less).
5184 38. Added support for the libradius Radius client library this is found on
5185 FreeBSD (previously only the radiusclient library was supported).
5191 1. When certain lookups returned multiple values in the form name=value, the
5192 quoting of the values was not always being done properly. Specifically:
5193 (a) If the value started with a double quote, but contained no whitespace,
5195 (b) If the value contained whitespace other than a space character (i.e.
5196 tabs or newlines or carriage returns) it was not quoted.
5197 This fix has been applied to the mysql and pgsql lookups by writing a
5198 separate quoting function and calling it from the lookup code. The fix
5199 should probably also be applied to nisplus, ibase and oracle lookups, but
5200 since I cannot test any of those, I have not disturbed their existing code.
5202 2. A hit in the callout cache for a specific address caused a log line with no
5203 reason for rejecting RCPT. Now it says "Previous (cached) callout
5204 verification failure".
5206 3. There was an off-by-one bug in the queryprogram router. An over-long
5207 return line was truncated at 256 instead of 255 characters, thereby
5208 overflowing its buffer with the terminating zero. As well as fixing this, I
5209 have increased the buffer size to 1024 (and made a note to document this).
5211 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5212 when Exim is waiting for an SMTP response from a remote server, Exim
5213 restarts its select() call on the socket, thereby resetting its timeout.
5214 This is not a problem when such interrupts are rare. Somebody set up a cron
5215 job to run exiwhat every 2 minutes, which is less than the normal select()
5216 timeout (5 or 10 minutes). This meant that the select() timeout never
5217 kicked in because it was always reset. I have fixed this by comparing the
5218 time when an interrupt arrives with the time at the start of the first call
5219 to select(). If more time than the timeout has elapsed, the interrupt is
5220 treated as a timeout.
5222 5. Some internal re-factoring in preparation for the addition of Sieve
5223 extensions (by MH). In particular, the "personal" test is moved to a
5224 separate function, and given an option for scanning Cc: and Bcc: (which is
5225 not set for Exim filters).
5227 6. When Exim created an email address using the login of the caller as the
5228 local part (e.g. when creating a From: or Sender: header line), it was not
5229 quoting the local part when it contained special characters such as @.
5231 7. Installed new OpenBSD configuration files.
5233 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5234 try to make them clearer.
5236 9. Callout options, other than the timeout value, were being ignored when
5237 verifying sender addresses in header lines. For example, when using
5239 verify = header_sender/callout=no_cache
5241 the cache was (incorrectly) being used.
5243 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5244 only the Cygwin environment.
5246 11. When return_path_on_delivery was set as a log selector, if different remote
5247 addresses in the same message used different return paths and parallel
5248 remote delivery occurred, the wrong values would sometimes be logged.
5249 (Whenever a remote delivery process finished, the return path value from
5250 the most recently started remote delivery process was logged.)
5252 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5253 lines when AUTH and/or TLS are in use. This is the "received protocol"
5254 field. Exim used to use "asmtp" for authenticated SMTP, without any
5255 indication (in the protocol name) for TLS use. Now it follows the RFC and
5256 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5257 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5258 names appear in log lines as well as in Received: header lines.
5260 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5261 extensions, and comparison tests, and to fix some bugs.
5263 14. Changes to the "personal" filter test:
5265 (1) The test was buggy in that it was just doing the equivalent of
5266 "contains" tests on header lines. For example, if a user's address was
5267 anne@some.where, the "personal" test would incorrectly be true for
5269 To: susanne@some.where
5271 This test is now done by extracting each address from the header in turn,
5272 and checking the entire address. Other tests that are part of "personal"
5273 are now done using regular expressions (for example, to check local parts
5274 of addresses in From: header lines).
5276 (2) The list of non-personal local parts in From: addresses has been
5277 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5278 taken from the Sieve specification recommendations.
5280 (3) If the message contains any header line starting with "List-" it is
5281 treated as non-personal.
5283 (4) The test for "circular" in the Subject: header line has been removed
5284 because it now seems ill-conceived.
5286 15. Minor typos in src/EDITME comments corrected.
5288 16. Installed latest exipick from John Jetmore.
5290 17. If headers_add on a router specified a text string that was too long for
5291 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5292 of string_sprintf() is now avoided.
5294 18. $message_body_size was not set (it was always zero) when running the DATA
5295 ACL and the local_scan() function.
5297 19. For the "mail" command in an Exim filter, no default was being set for
5298 the once_repeat time, causing a random time value to be used if "once" was
5299 specified. (If the value happened to be <= 0, no repeat happened.) The
5300 default is now 0s, meaning "never repeat". The "vacation" command was OK
5301 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5302 (I found it when inspecting the code).
5304 20. There is now an overall timeout for performing a callout verification. It
5305 defaults to 4 times the callout timeout, which applies to individual SMTP
5306 commands during the callout. The overall timeout applies when there is more
5307 than one host that can be tried. The timeout is checked before trying the
5308 next host. This prevents very long delays if there are a large number of
5309 hosts and all are timing out (e.g. when the network connections are timing
5310 out). The value of the overall timeout can be changed by specifying an
5311 additional sub-option for "callout", called "maxwait". For example:
5313 verify = sender/callout=5s,maxwait=20s
5315 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5316 to the end before writing, but this should make it even safer).
5318 22. Exim was forgetting that it had advertised PIPELINING for the second and
5319 subsequent messages on an SMTP connection. It was also not resetting its
5320 memory on STARTTLS and an internal HELO.
5322 23. When Exim logs an SMTP synchronization error within a session, it now
5323 records whether PIPELINING has been advertised or not.
5325 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5326 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5327 rather than long int.
5329 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5331 26. Added the never_mail option to autoreply.
5337 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5338 crash if the getsockname() call failed; this can happen if a connection is
5339 closed very soon after it is established. The problem was simply in the
5340 order in which certain operations were done, causing Exim to try to write
5341 to the SMTP stream before it had set up the file descriptor. The bug has
5342 been fixed by making things happen in the correct order.
5348 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5349 before the connection was closed, thus losing the rejection response.
5351 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5352 some early Solaris releases, but causes trouble in current releases where
5353 socklen_t is defined.
5355 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5358 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5361 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5362 that releasing the top of it at the end releases what was used for sub-
5363 expansions (unless the block got too big). However, discard this block if
5364 the first thing is a variable or header, so that we can use its block when
5365 it is dynamic (useful for very large $message_headers, for example).
5367 6. Lookups now cache *every* query, not just the most recent. A new, separate
5368 store pool is used for this. It can be recovered when all lookup caches are
5369 flushed. Lookups now release memory at the end of their result strings.
5370 This has involved some general refactoring of the lookup sources.
5372 7. Some code has been added to the store_xxx() functions to reduce the amount
5373 of flapping under certain conditions.
5375 8. log_incoming_interface used to affect only the <= reception log lines. Now
5376 it causes the local interface and port to be added to several more SMTP log
5377 lines, for example "SMTP connection from", and rejection lines.
5379 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5381 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5383 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5384 could be overwritten at the end of the current message (or the start of a
5385 new message if it was set in a HELO ACL). The value is now preserved for
5386 the duration of the SMTP connection.
5388 12. If a transport had a headers_rewrite setting, and a matching header line
5389 contained an unqualified address, that address was qualified, even if it
5390 did not match any rewriting rules. The underlying bug was that the values
5391 of the flags that permit the existence of unqualified sender and recipient
5392 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5393 non-local messages, and by -bnq for local messages) were not being
5394 preserved with the message after it was received.
5396 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5397 "next input=" as part of the text comprising the host identity instead of
5398 the correct text. The code was using the same buffer for two different
5399 strings. However, depending on which order the printing function evaluated
5400 its arguments, the bug did not always show up. Under Linux, for example, my
5401 test suite worked just fine.
5403 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5404 doesn't work with some older versions of Perl. It has been changed to "my",
5405 which in any case is probably the better facility to use.
5407 15. A really picky compiler found some instances of statements for creating
5408 error messages that either had too many or two few arguments for the format
5411 16. The size of the buffer for calls to the DNS resolver has been increased
5412 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5413 for addresses that have a lot of PTR records. This alleviates a problem; it
5414 does not fully solve it.
5416 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5417 buffer now truncates the list and logs the incident, which is the same
5418 action as happens when Exim is looking up a host name and its aliases.
5419 Previously in this situation something unpredictable would happen;
5420 sometimes it was "internal error: store_reset failed".
5422 18. If a server dropped the connection unexpectedly when an Exim client was
5423 using GnuTLS and trying to read a response, the client delivery process
5424 crashed while trying to generate an error log message.
5426 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5427 string, for example:
5429 warn message = H1: something\nH2: something
5431 the text was added as a single header line from Exim's point of view
5432 though it ended up OK in the delivered message. However, searching for the
5433 second and subsequent header lines using $h_h2: did not work. This has been
5434 fixed. Similarly, if a system filter added multiple headers in this way,
5435 the routers could not see them.
5437 20. Expanded the error message when iplsearch is called with an invalid key to
5438 suggest using net-iplsearch in a host list.
5440 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5441 ACLs are no longer actually imposed (and a message to that effect is
5444 22. If a "gecos" field in a passwd entry contained escaped characters, in
5445 particular, if it contained a \" sequence, Exim got it wrong when building
5446 a From: or a Sender: header from that name. A second bug also caused
5447 incorrect handling when an unquoted " was present following a character
5448 that needed quoting.
5450 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5451 was not being matched caselessly.
5453 24. Arranged for all hyphens in the exim.8 source to be escaped with
5456 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5457 properly. Recipient callout cache records were still being keyed to include
5458 the sender, even when use_sender was set false. This led to far more
5459 callouts that were necessary. The sender is no longer included in the key
5460 when use_sender is false.
5462 26. Added "control = submission" modifier to ACLs.
5464 27. Added the ${base62d: operator to decode base 62 numbers.
5466 28. dnsdb lookups can now access SRV records.
5468 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5469 the configuration file.
5471 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5472 (-v) mode. This makes the output for a verbose queue run more intelligible.
5474 31. Added a use_postmaster feature to recipient callouts.
5476 32. Added the $body_zerocount variable, containing the number of binary zero
5477 bytes in the message body.
5479 33. The time of last modification of the "new" subdirectory is now used as the
5480 "mailbox time last read" when there is a quota error for a maildir
5483 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5485 35. Added +ignore_unknown as a special item in host lists.
5487 36. Code for decoding IPv6 addresses in host lists is now included, even if
5488 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5489 address was recognized as an IP address, but was then not correctly decoded
5490 into binary, causing unexpected and incorrect effects when compared with
5497 1. Very minor rewording of debugging text in manualroute to say "list of
5498 hosts" instead of "hostlist".
5500 2. If verify=header_syntax was set, and a header line with an unqualified
5501 address (no domain) and a large number of spaces between the end of the
5502 name and the colon was received, the reception process suffered a buffer
5503 overflow, and (when I tested it) crashed. This was caused by some obsolete
5504 code that should have been removed. The fix is to remove it!
5506 3. When running in the test harness, delay a bit after writing a bounce
5507 message to get a bit more predictability in the log output.
5509 4. Added a call to search_tidyup() just before forking a reception process. In
5510 theory, someone could use a lookup in the expansion of smtp_accept_max_
5511 per_host which, without the tidyup, could leave open a database connection.
5513 5. Added the variables $recipient_data and $sender_data which get set from a
5514 lookup success in an ACL "recipients" or "senders" condition, or a router
5515 "senders" option, similar to $domain_data and $local_part_data.
5517 6. Moved the writing of debug_print from before to after the "senders" test
5520 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5521 problems for message scanning, either using a data ACL, or using
5522 local_scan() because the Received: header was not generated till after they
5523 were called (in order to set the time as the time of reception completion).
5524 I have revised the way this works. The header is now generated after the
5525 body is received, but before the ACL or local_scan() are called. After they
5526 are run, the timestamp in the header is updated.
5532 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5533 before starting a queue runner without re-exec. This happened only when
5534 deliver_drop_privilege was set or when the Exim user was set to root. The
5535 effect of the bug was that timeouts during subsequent deliveries caused
5536 crashes instead of being properly handled. The handler is now left at its
5537 default (and expected) setting.
5539 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5540 message, again when deliver_drop_privilege is set or Exim is run as root.
5541 The bug described in (1) was not present in this case, but the tidying up
5542 of the other signals was missing. I have made the two cases consistent.
5544 3. The ignore_target_hosts setting on a manualroute router was being ignored
5545 for hosts that were looked up using the /MX notation.
5547 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5550 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5551 operated on the sender address. After changing the $sender_address to <>
5552 for the sender address verify, Exim was re-instated it as the original
5553 (before rewriting) address, but remembering that it had rewritten it, so it
5554 wasn't rewriting it again. This bug also had the effect of breaking the
5555 sender address verification caching when the sender address was rewritten.
5557 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5558 This has been changed so that if the ip literal address matches
5559 ignore_target_hosts, the router declines.
5561 7. Added expansion conditions match_domain, match_address, and match_local_
5562 part (NOT match_host).
5564 8. The placeholder for the Received: header didn't have a length field set.
5566 9. Added code to Exim itself and to exim_lock to test for a specific race
5567 condition that could lead to file corruption when using MBX delivery. The
5568 issue is with the lockfile that is created in /tmp. If this file is removed
5569 after a process has opened it but before that process has acquired a lock,
5570 there is the potential for a second process to recreate the file and also
5571 acquire a lock. This could lead to two Exim processes writing to the file
5572 at the same time. The added code performs the same test as UW imapd; it
5573 checks after acquiring the lock that its file descriptor still refers to
5574 the same named file.
5576 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5577 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5580 11. Added the smtp_active_hostname option. If used, this will typically be made
5581 to depend on the incoming interface address. Because $interface_address is
5582 not set up until the daemon has forked a reception process, error responses
5583 that can happen earlier (such as "too many connections") no longer contain
5586 12. If an expansion in a condition on a "warn" statement fails because a lookup
5587 defers, the "warn" statement is abandoned, and the next ACL statement is
5588 processed. Previously this caused the whole ACL to be aborted.
5590 13. Added the iplsearch lookup type.
5592 14. Added ident_timeout as a log selector.
5594 15. Added tls_certificate_verified as a log selector.
5596 16. Added a global option tls_require_ciphers (compare the smtp transport
5597 option of the same name). This controls incoming TLS connections.
5599 17. I finally figured out how to make tls_require_ciphers do a similar thing
5600 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5601 before starting the TLS session.
5603 18. Tabs are now shown as \t in -bP output.
5605 19. If the log selector return_path_on_delivery was set, Exim crashed when
5606 bouncing a message because it had too many Received: header lines.
5608 20. If two routers both had headers_remove settings, and the first one included
5609 a superfluous trailing colon, the final name in the first list and the
5610 first name in the second list were incorrectly joined into one item (with a
5611 colon in the middle).
5617 1. Added -C and -D options to the exinext utility, mainly to make it easier
5618 to include in the automated testing, but these could be helpful when
5619 multiple configurations are in use.
5621 2. The exinext utility was not formatting the output nicely when there was
5622 an alternate port involved in the retry record key, nor when there was a
5623 message id as well (for retries that were specific to a specific message
5624 and a specific host). It was also confused by IPv6 addresses, because of
5625 the additional colons they contain. I have fixed the IPv4 problem, and
5626 patched it up to do a reasonable job for IPv6.
5628 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5629 delivery, the log line now contains "pipelined" if PIPELINING was used.
5631 4. An SMTP transport process used to panic and die if the bind() call to set
5632 an explicit outgoing interface failed. This has been changed; it is now
5633 treated in the same way as a connect() failure.
5635 5. A reference to $sender_host_name in the part of a conditional expansion
5636 that was being skipped was still causing a DNS lookup. This no longer
5639 6. The def: expansion condition was not recognizing references to header lines
5640 that used bh_ and bheader_.
5642 7. Added the _cache feature to named lists.
5644 8. The code for checking quota_filecount in the appendfile transport was
5645 allowing one more file than it should have been.
5647 9. For compatibility with Sendmail, the command line option
5655 and sets the incoming protocol and host name (for trusted callers). The
5656 host name and its colon can be omitted when only the protocol is to be set.
5657 Note the Exim already has two private options, -pd and -ps, that refer to
5658 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5659 "s", but I don't think that's a major issue.
5661 10. A number of refactoring changes to the code, none of which should affect
5664 (a) The number of logging options was getting close to filling up the
5665 32-bit word that was used as a bit map. I have split them into two classes:
5666 those that are passed in the argument to log_write(), and those that are
5667 only ever tested independently outside of that function. These are now in
5668 separate 32-bit words, so there is plenty of room for expansion again.
5669 There is no change in the user interface or the logging behaviour.
5671 (b) When building, for example, log lines, the code previously used a
5672 macro that called string_cat() twice, in order to add two strings. This is
5673 not really sufficiently general. Furthermore, there was one instance where
5674 it was actually wrong because one of the argument was used twice, and in
5675 one call a function was used. (As it happened, calling the function twice
5676 did not affect the overall behaviour.) The macro has been replaced by a
5677 function that can join an arbitrary number of extra strings onto a growing
5680 (c) The code for expansion conditions now uses a table and a binary chop
5681 instead of a serial search (which was left over from when there were very
5682 few conditions). Also, it now recognizes conditions like "pam" even when
5683 the relevant support is not compiled in: a suitably worded error message is
5684 given if an attempt is made to use such a condition.
5686 11. Added ${time_interval:xxxxx}.
5688 12. A bug was causing one of the ddress fields not to be passed back correctly
5689 from remote delivery subprocesses. The field in question was not being
5690 subsequently used, so this caused to problems in practice.
5692 13. Added new log selectors queue_time and deliver_time.
5694 14. Might have fixed a bug in maildirsizefile handling that threw up
5695 "unexpected character" debug warnings, and recalculated the data
5696 unnecessarily. In any case, I expanded the warning message to give more
5699 15. Added the message "Restricted characters in address" to the statements in
5700 the default ACL that block characters like @ and % in local parts.
5702 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5703 Three changes have been made:
5705 (a) There was a serious bug; a negative response to MAIL caused the whole
5706 recipient domain to be cached as invalid, thereby blocking all messages
5707 to all local parts at the same domain, from all senders. This bug has
5708 been fixed. The domain is no longer cached after a negative response to
5709 MAIL if the sender used is not empty.
5711 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5714 (c) A new callout option, "use_sender" has been added for people who want
5715 the modified behaviour.
5721 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5724 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5725 indeed breaks things for older releases.
5727 3. Added additional logging to the case where there is a problem reading data
5728 from a filter that is running in a subprocess using a pipe, in order to
5729 try to track down a specific problem.
5731 4. Testing facility fudge: when running in the test harness and attempting
5732 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5733 getting "No route to host". Convert this to a timeout.
5735 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5738 6. Some OS don't have socklen_t but use size_t instead. This affects the
5739 fifth argument of getsockopt() amongst other things. This is now
5740 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5741 can be set for individual OS. I have set it for SunOS5, OSF1, and
5742 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5743 some earlier ones do not.
5745 7. Change 4.30/15 was not doing the test caselessly.
5747 8. The standard form for an IPv6 address literal was being rejected by address
5748 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5749 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5750 this, as well as the form without the "IPv6" on the front (but only when
5751 address literals are enabled, of course).
5753 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5755 10. Exim crashed if a message with an empty sender address specified by -f
5756 encountered a router with an errors_to setting. This could be provoked only
5757 by a command such as
5761 where an empty string was supplied; "<>" did not hit this bug.
5763 11. Installed PCRE release 4.5.
5765 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5766 remained set. It is now erased.
5768 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5769 times from message ids (which are base 36 rather than the normal 62).
5771 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5772 were being counted as actual protocol errors, and logged if the log
5773 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5774 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5775 and DATA following a set of rejected RCPTs do not count as protocol errors.
5776 In other words, Exim assumes they were pipelined, though this may not
5777 actually be the case. Of course, in all cases the client gets an
5778 appropriate error code.
5780 15. If a lookup fails in an ACL condition, a message about the failure may
5781 be available; it is used if testing the ACL cannot continue, because most
5782 such messages specify what the cause of the deferral is. However, some
5783 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5784 that caused an old message to be retained and used if a later statement
5785 caused a defer, replacing the real cause of the deferral.
5787 16. If an IP address had so many PTR records that the DNS lookup buffer
5788 was not large enough to hold them, Exim could crash while trying to process
5789 the truncated data. It now detects and logs this case.
5791 17. Further to 4.21/58, another change has been made: if (and only if) the
5792 first line of a message (the first header line) ends with CRLF, a bare LF
5793 in a subsequent header line has a space inserted after it, so as not to
5794 terminate the header.
5796 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5797 unnecessarily, used atoi() instead of strtol(), and didn't check the
5798 termination when getting file sizes from file names by regex.
5800 19. Completely re-implemented the support for maildirsize files, in the light
5801 of a number of problems with the previous contributed implementation
5802 (4.30/29). In particular:
5804 . If the quota is zero, the maildirsize file is maintained, but no quota is
5807 . If the maildir directory does not exist, it is created before any attempt
5808 to write a maildirsize file.
5810 . The quota value in the file is just a cache; if the quota is changed in
5811 the transport, the new value overrides.
5813 . A regular expression is available for excluding directories from the
5816 20. The autoreply transport checks the characters in options that define the
5817 message's headers; it allows continued headers, but it was checking with
5818 isspace() after an embedded newline instead of explicitly looking for a
5821 21. If all the "regular" hosts to which an address was routed had passed their
5822 expiry times, and had not reached their retry times, the address was
5823 bounced, even if fallback hosts were defined. Now Exim should go on to try
5826 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5827 equivalent code in the SMTP transport. Some hosts send humungous responses
5828 to HELO/EHLO, more than 1024 it seems.
5830 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5831 gives compiler warnings in some environments. I've now done it "properly",
5834 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5835 (because of problems with the built-in one) was declared to return uschar *
5836 instead of char *, causing compiler failure.
5838 25. Fixed a file descriptor leak when processing alias/forward files.
5840 26. Fixed a minor format string issue in dbfn.c.
5842 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5844 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5845 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5848 29. When a sender address is verified, it is cached, to save repeating the test
5849 when there is more than one recipient in a message. However, when the
5850 verification involves a callout, it is possible for different callout
5851 options to be set for different recipients. It is too complicated to keep
5852 track of this in the cache, so now Exim always runs a verification when a
5853 callout is required, relying on the callout cache for the optimization.
5854 The overhead is duplication of the address routing, but this should not be
5857 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5858 to be verified with callout=postmaster, and the main callout worked but the
5859 postmaster check failed, the verification correctly failed. However, if a
5860 subsequent RCPT command asked for sender verification *without* the
5861 postmaster check, incorrect caching caused this verification also to fail,
5864 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5865 it was not caching the DNS options (qualify_single, search_parents) that
5866 were used when the lookup failed. A subsequent lookup with different
5867 options therefore always gave the same answer, though there were cases
5868 where it should not have. (Example: a "domains = !$mx_any" option on a
5869 dnslookup router: the "domains" option is always processed without any
5870 widening, but the router might have qualify_single set.) Now Exim uses the
5871 cached value only when the same options are set.
5873 32. Added John Jetmore's "exipick" utility to the distribution.
5875 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5876 than a timeout (e.g. a certificate is required, and is not provided), an
5877 Exim server now closes the connection immediately. Previously it waited for
5878 the client to close - but if the client is SSL, it seems that they each
5879 wait for each other, leading to a delay before one of them times out.
5881 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5882 maintained 0.8.x compatibility because I don't think many are using it, and
5883 it is clearly obsolete.
5885 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5888 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5889 client certificate was expired. A simple patch fixes this, though I don't
5890 understand the full logic of why the verify callback is called multiple
5893 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5894 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5895 which causes problems with some clients (such as the Certicom SSL Plus
5896 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5897 disables the coutermeasure allowing Eudora to connect."
5899 38. Exim was not checking that a write() to a log file succeeded. This could
5900 lead to Bad Things if a log got too big, in particular if it hit a file
5901 size limit. Exim now panics and dies if it cannot write to a log file, just
5902 as it does if it cannot open a log file.
5904 39. Modified OS/Makefile-Linux so that it now contains
5906 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5908 The two -D definitions ensure that Exim is compiled with large file
5909 support, which makes it possible to handle log files that are bigger than
5912 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5913 instance) a domain was checked against a named list that involved a lookup,
5914 causing $domain_data to be set, then another domain was checked against the
5915 same list, then the first domain was re-checked, the value of $domain_data
5916 after the final check could be wrong. In particular, if the second check
5917 failed, it could be set empty. This bug probably also applied to
5920 41. The strip_trailing_dot option was not being applied to the address given
5921 with the -f command-line option.
5923 42. The code for reading a message's header from the spool was incrementing
5924 $received_count, but never initializing it. This meant that the value was
5925 incorrect (doubled) while delivering a message in the same process in which
5926 it was received. In the most common configuration of Exim, this never
5927 happens - a fresh exec is done - but it can happen when
5928 deliver_drop_privilege is set.
5930 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5931 it now includes up to 150 characters of the unexpected data in the log
5934 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5935 and building data strings. The size of both of these buffers was 10 000
5936 bytes - far larger than anybody would *ever* want, thought I. Needless to
5937 say, somebody hit the limit. I have increased the maximum line length to
5938 20 000 and the maximum data length of concatenated lines to 100 000. I have
5939 also fixed two bugs, because there was no checking on these buffers. Tsk,
5940 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5941 buffer is too small.
5943 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5944 lsearch lookups. Now it does.
5946 46. When parsing a route_list item in a manualroute router, a fixed-length
5947 buffer was used for the list of hosts. I made this 1024 bytes long,
5948 thinking that nobody would ever have a list of hosts that long. Wrong.
5949 Somebody had a whole pile of complicated expansion conditions, and the
5950 string was silently truncated, leading to an expansion error. It turns out
5951 that it is easier to change to an unlimited length (owing to other changes
5952 that have happened since this code was originally written) than to build
5953 structure for giving a limitation error. The length of the item that
5954 expands into the list of hosts is now unlimited.
5956 47. The lsearch lookup could not handle data where the length of text line was
5957 more than 4095 characters. Such lines were truncated, leading to shortened
5958 data being returned. It should now handle lines of any length.
5960 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5961 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5964 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5965 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5966 no longer happens in comment lines. A list of replacements is now placed
5967 at the head of all of the source files, except those whose only change is
5968 to replace PERL_COMMAND in the very first #! line.
5970 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5971 messages on the queue, with a bottom-up merge sort, using code contributed
5972 by Michael Haardt. This should make operations like -bp somewhat faster on
5973 large queues. It won't affect queue runners, except when queue_run_in_order
5976 51. Installed eximstats 1.31 in the distribution.
5978 52. Added support for SRV lookups to the dnslookup router.
5980 53. If an ACL referred to $message_body or $message_body_end, the value was not
5981 reset for any messages that followed in the same SMTP session.
5983 54. The store-handling optimization for building very long strings was not
5984 differentiating between the different store pools. I don't think this
5985 actually made any difference in practice, but I've tidied it.
5987 55. While running the routers to verify a sender address, $sender_address
5988 was still set to the sender address. This is wrong, because when routing to
5989 send a bounce to the sender, it would be empty. Therefore, I have changed
5990 it so that, while verifying a sender address, $sender_address is set to <>.
5991 (There is no change to what happens when verifying a recipient address.)
5993 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5994 target A or AAAA records (if not already returned) without resetting the
5995 qualify_single or search_parents options of the DNS resolver. These are
5996 inappropriate in this case because the targets of MX and SRV records must
5997 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5998 target that, when qualified, matched something in the local domain. These
5999 two options are now turned off when doing these lookups.
6001 57. It seems that at least some releases of Reiserfs (which does not have the
6002 concept of a fixed number of inodes) returns zero and not -1 for the
6003 number of available inodes. This interacted badly with check_spool_inodes,
6004 which assumed that -1 was the "no such thing" setting. What I have done is
6005 to check that the total number of inodes is greater than zero before doing
6006 the test of how many are available.
6008 58. When a "warn" ACL statement has a log_message modifier, the message is
6009 remembered, and not repeated. This is to avoid a lot of repetition when a
6010 message has many recipients that cause the same warning to be written.
6011 However, Exim was preserving the list of already written lines for an
6012 entire SMTP session, which doesn't seem right. The memory is now reset if a
6013 new message is started.
6015 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6016 debugging output unless log_rewrite was also set.
6018 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6019 of (int)(handle) when we know that handle contains (void *)(-1).
6021 61. The Exim daemon panic-logs an error return when it closes the incoming
6022 connection. However "connection reset by peer" seems to be common, and
6023 isn't really an error worthy of noting specially, so that particular error
6026 62. When Exim is trying to find all the local interfaces, it used to panic and
6027 die if the ioctl to get the interface flags failed. However, it seems that
6028 on at least one OS (Solaris 9) it is possible to have an interface that is
6029 included in the list of interfaces, but for which you get a failure error
6030 for this call. This happens when the interface is not "plumbed" into a
6031 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6032 failure of the "get flags" call assumes that the interface is down.
6034 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6035 makes life easier for people who are doing arithmetic on fields extracted
6036 from dates, where you often get leading zeros that should not be
6037 interpreted as octal.
6039 64. Added qualify_domain to the redirect router, to override the global
6042 65. If a pathologically long header line contained very many addresses (the
6043 report of this problem mentioned 10 000) and each of them was rewritten,
6044 Exim could use up a very large amount of memory. (It kept on making new
6045 copies of the header line as it rewrote, and never released the old ones.)
6046 At the expense of a bit more processing, the header rewriting function has
6047 been changed so that it no longer eats memory in this way.
6049 66. The generation of the Received: header has been moved from the time that a
6050 message starts to be received, to the time that it finishes. The timestamp
6051 in the Received: header should now be very close to that of the <= log
6052 line. There are two side-effects of this change:
6054 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6055 logged header lines no longer include the local Received: line, because
6056 it has not yet been created. The same applies to a copy of the message
6057 that is returned to a non-SMTP sender when a message is rejected.
6059 (b) When a filter file is tested using -bf, no additional Received: header
6060 is added to the test message. After some thought, I decided that this
6063 This change does not affect the value of $received_for. It is still set
6064 after address rewriting, but before local_scan() is called.
6066 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6068 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6069 gave an unhelpful panic error message, and a defer error. I have managed to
6070 change this behaviour so that it now rejects any supplied certificate,
6071 which seems right, as the list of acceptable certificates is empty.
6073 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6074 gave an unhelpful defer error. I have not managed to make this reject any
6075 supplied certificates, but the error message it gives is "no certificate
6076 supplied", which is not helpful.
6078 70. exigrep's output now also includes lines that are not associated with any
6079 message, but which match the given pattern. Implemented by a patch from
6080 Martin Sluka, which also tidied up the Perl a bit.
6082 71. Recipient callout verification, like sender verification, was using <> in
6083 the MAIL FROM command. This isn't really the right thing, since the actual
6084 sender may affect whether the remote host accepts the recipient or not. I
6085 have changed it to use the actual sender in the callout; this means that
6086 the cache record is now keyed on a recipient/sender pair, not just the
6087 recipient address. There doesn't seem to be a real danger of callout loops,
6088 since a callout by the remote host to check the sender would use <>.
6089 [SEE ABOVE: changed after hitting problems.]
6091 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6092 temporary errors. However, in the case of such a code being given after
6093 the end of a data transmission (i.e. after ".") Exim was failing to write
6094 a retry record for the message. (Yes, there was some broken host that was
6095 actually sending 8xx at this point.)
6097 73. An unknown lookup type in a host list could cause Exim to panic-die when
6098 the list was checked. (An example that provoked this was putting <; in the
6099 middle of a list instead of at the start.) If this happened during a DATA
6100 ACL check, a -D file could be left lying around. This kind of configuration
6101 error no longer causes Exim to die; instead it causes a defer error. The
6102 incident is still logged to the main and panic logs.
6104 74. Buglet left over from Exim 3 conversion. The message "too many messages
6105 in one connection" was written to the rejectlog but not the mainlog, except
6106 when address rewriting (yes!) was being logged.
6108 75. Added write_rejectlog option.
6110 76. When a system filter was run not as root (that is, when system_filter_user
6111 was set), the values of the $n variables were not being returned to the
6112 main process; thus, they were not subsequently available in the $sn
6115 77. Added +return_path_on_delivery log selector.
6117 78. A connection timeout was being treated differently from recipients deferred
6118 when testing hosts_max_try with a message that was older than the host's
6119 retry timeout. (The host should not be counted, thus allowing all hosts to
6120 be tried at least once before bouncing.) This may have been the cause of an
6121 occasionally reported bug whereby a message would remain on the queue
6122 longer than the retry timeout, but would be bounced if a delivery was
6123 forced. I say "may" because I never totally pinned down the problem;
6124 setting up timeout/retry tests is difficult. See also the next item.
6126 79. The ultimate address timeout was not being applied to errors that involved
6127 a combination of host plus message (for example, a timeout on a MAIL
6128 command). When an address resolved to a number of possible hosts, and they
6129 were not all tried for each delivery (e.g. because of hosts_max_try), a
6130 message could remain on the queue longer than the retry timeout.
6132 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6135 81. Fixed an obscure SMTP outgoing bug which required at least the following
6136 conditions: (a) there was another message waiting for the same server;
6137 (b) the server returned 5xx to all RCPT commands in the first message so
6138 that the message was not completed; (c) the server dropped the connection
6139 or gave a negative response to the RSET that Exim sends to abort the
6140 transaction. The observed case was a dropped connection after DATA that had
6141 been sent in pipelining mode. That is, the server had advertised PIPELINING
6142 but was not implementing it correctly. The effect of the bug was incorrect
6143 behaviour, such as trying another host, and this could lead to a crash.
6149 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6150 and daemon.c were passed as pointers to ints; they should have been
6151 pointers to socklen_t variables (which are typically unsigned ints).
6153 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6156 3. Fixed a really odd bug that affected only the testing scheme; patching a
6157 certain fixed string in the binary changed the value of another string that
6158 happened to be identical to the end of the original first string.
6160 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6161 name", it returns that address as the IP address. On some operating
6162 systems (e.g. Solaris), it also passes back the IP address string as the
6163 "host name". However, on others (e.g. Linux), it passes back an empty
6164 string. Exim wasn't checking for this, and was changing the host name to an
6165 empty string, assuming it had been canonicalized.
6167 5. Although rare, it is permitted to have more than one PTR record for a given
6168 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6169 all the names associated with an address, because they do in Solaris.
6170 However, it seems that they do not in Linux for data that comes from the
6171 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6172 I found this out when I moved to a new Linux workstation and tried to run
6173 the Exim test suite.
6175 To get round this problem I have changed the code so that it now does its
6176 own call to the DNS to look up PTR records when searching for a host name.
6177 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6178 addresses that are only in /etc/hosts are still found.
6180 This behaviour is, however, controlled by an option called host_lookup_
6181 order, which defaults to "bydns:byaddr". If people want to use the other
6182 order, or indeed, just use one or the other means of lookup, they can
6183 specify it in this variable.
6185 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6186 some operating systems, this comes back from gethostbyaddr() as an empty
6187 string, and this is what Exim used to test for. However, it seems that in
6188 other systems, "." is yielded. Exim now tests for this case too.
6190 7. The values of check_spool_space and check_log_space are now held internally
6191 as a number of kilobytes instead of an absolute number of bytes. If a
6192 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6193 kilobyte. This means that much larger values can be stored.
6195 8. Exim monitor: an attempt to get the action menu when not actually pointing
6196 at a message produces an empty menu entitled "No message selected". This
6197 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6198 no entries in it ("Shell widget menu has zero width and/or height"). So I
6199 have added a single, blank menu entry in this case.
6201 9. Added ${quote_local_part.
6203 10. MIME decoding is now applied to the contents of Subject: header lines when
6206 11. Now that a reference to $sender_host_address automatically causes a reverse
6207 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6208 host lookup before query-style lookups in lists that might use this
6209 variable. This has therefore been abolished, and the "net-" prefix is no
6210 longer necessary for query-style lookups.
6212 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6213 have been a typo for LFLAGS, so it has been changed.
6215 13. The install script calls Exim with "-C /dev/null" in order to find the
6216 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6217 to be output. However, since Exim outputs its version number before the
6218 error, it didn't break the script. It just looked ugly. I fixed this by
6219 always allowing "-C /dev/null" if the caller is root.
6221 14. Ignore overlarge ACL variable number when reading spool file - insurance
6222 against a later release with more variables having written the file.
6224 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6225 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6226 this, as well as the form without the "IPv6" on the front.
6228 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6229 OS/Makefile-Darwin file.
6231 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6232 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6234 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6235 variable. However, some OS use ints for inodes. I've added cast to long int
6236 to get rid of the compiler warning.
6238 19. I had forgotten to lock out "/../" in configuration file names when
6239 ALT_CONFIG_PREFIX was set.
6241 20. Routers used for verification do not need to specify transports. However,
6242 if such a router generated a host list, and callout was configured, Exim
6243 crashed, because it could not find a port number from the (non-existent)
6244 transport. It now assumes port 25 in this circumstance.
6246 21. Added the -t option to exigrep.
6248 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6249 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6250 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6251 them is set, LOOKUP_LSEARCH is forced.
6253 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6255 24. Added sender and host information to the "rejected by local_scan()" log
6256 line; previously there was no indication of these.
6258 25. Added .include_if_exists.
6260 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6261 Linux. It turns out that not all file systems support this. Apparently some
6262 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6263 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6264 not supported on the file descriptor, is now ignored when Exim is trying to
6265 sync a directory. This applies only to Linux.
6267 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6269 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6270 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6273 29. Added support for maildirsize files from supplied patch (modified a bit).
6275 30. The use of :fail: followed by an empty string could lead Exim to respond to
6276 sender verification failures with (e.g.):
6278 550 Verification failed for <xxx>
6279 550 Sender verify failed
6281 where the first response line was missing the '-' that indicates it is not
6282 the final line of the response.
6284 31. The loop for finding the name of the user that called Exim had a hardwired
6285 limit of 10; it now uses the value of finduser_retries, which is used for
6286 all other user lookups.
6288 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6291 33. Exim was neglecting to zero errno before one call of strtol() when
6292 expanding a string and expecting an integer value. On some systems this
6293 resulted in spurious "integer overflow" errors. Also, it was casting the
6294 result into an int without checking.
6296 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6297 did not work. The code looks as if it has *never* worked, though it appears
6298 to have been documented since at least release 1.62. I have made it work.
6300 35. The "timeout_DNS" error in retry rules, also documented since at least
6301 1.62, also never worked. As it isn't clear exactly what this means, and
6302 clearly it isn't a major issue, I have abolished the feature by treating it
6303 as "timeout", and writing a warning to the main and panic logs.
6305 36. The display of retry rules for -brt wasn't always showing the error code
6308 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6309 timeout_connect_A, timeout_connect_MX.
6311 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6312 to the empty sender.
6314 39. The daemon was not analysing the content of -oX till after it had closed
6315 stderr and disconnected from the controlling terminal. This meant that any
6316 syntax errors were only noted on the panic log, and the return code from
6317 the command was 0. By re-arranging the code a little, I've made the
6318 decoding happen first, so such errors now appear on stderr, and the return
6319 code is 1. However, the actual setting up of the sockets still happens in
6320 the disconnected process, so errors there are still only recorded on the
6323 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6324 connections (as happens on some IP stacks) was logged at start up time as
6325 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6326 it from "IPv6 and IPv4", which means that two separate sockets are being
6329 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6330 says whether AF_INET or AF_INET6 was passed as an argument.
6332 42. Exiwhat output was messed up when time zones were included in log
6335 43. Exiwhat now gives more information about the daemon's listening ports,
6336 and whether -tls-on-connect was used.
6338 44. The "port" option of the smtp transport is now expanded.
6340 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6341 silently ignored. Now an error message is written to the main and panic
6344 46. There's a new ACL modifier called "logwrite" which writes to a log file
6345 as soon as it is encountered.
6347 47. Added $local_user_uid and $local_user_gid at routing time.
6349 48. Exim crashed when trying to verify a sender address that was being
6352 49. Exim was recognizing only a space character after ".include". It now also
6353 recognizes a tab character.
6355 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6356 extracting the relevant information from the specification. The man page no
6357 longer contains scrambled data for the -d option, and I've added a section
6358 at the front about calling Exim under different names.
6360 51. Added "extra_headers" argument to the "mail" command in filter files.
6362 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6365 53. Installed eximstats 1.29.
6367 54. Added transport_filter_timeout as a generic transport option.
6369 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6370 Cc: header lines. This was required by RFC 822, but it not required by RFC
6373 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6374 incoming messages that did not have them. Now it does so only if the
6375 message originates locally, that is, if there is no associated remote host
6376 address. When Resent- header lines are present, this applies to the Resent-
6377 lines rather than the non-Resent- lines.
6379 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6380 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6382 58. Messages for configuration errors now include the name of the main
6383 configuration file - useful now that there may be more than one file in a
6384 list (.included file names were always shown).
6386 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6387 for those rare installations that do not start the daemon as root or run it
6388 setuid root. I've cut out the call to initgroups() if the daemon is not
6391 60. The Exim user and group can now be bound into the binary as text strings
6392 that are looked up at the start of Exim's processing.
6394 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6396 62. Added $mailstore_basename variable.
6398 63. Installed patch to sieve.c from Michael Haardt.
6400 64. When Exim failed to open the panic log after failing to open the main log,
6401 the original message it was trying to log was written to stderr and debug
6402 output, but if they were not available (the usual case in production), it
6403 was lost. Now it is written to syslog before the two lines that record the
6404 failures to open the logs.
6406 65. Users' Exim filters run in subprocesses under the user's uid. It is
6407 possible for a "deliver" command or an alias in a "personal" command to
6408 provoke an address rewrite. If logging of address rewriting is configured,
6409 this fails because the process is not running as root or exim. There may be
6410 a better way of dealing with this, but for the moment (because 4.30 needs
6411 to be released), I have disabled address rewrite logging when running a
6412 filter in a non-root, non-exim process.
6418 1. The buildconfig auxiliary program wasn't quoting the value set for
6419 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6420 not defined. This bug was masked in 4.22 by the effect that was fixed in
6423 2. Some messages that were rejected after a message id was allocated were
6424 shown as "incomplete" by exigrep. It no longer does this for messages that
6425 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6427 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6428 have allow_domain_literals set, the ID did not get logged in the <= line.
6429 Domain literals are now always recognized in Message-ID: header lines.
6431 4. The first argument for a ${extract expansion item is the key name or field
6432 number. Leading and trailing spaces in this item were not being ignored,
6433 causing some misleading effects.
6435 5. When deliver_drop_privilege was set, single queue runner processes started
6436 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6437 same command in the process it spins off) were not dropping privilege.
6439 6. When the daemon running as "exim" started a queue runner, it always
6440 re-executed Exim in the spun-off process. This is a waste of effort when
6441 deliver_drop_privilege is set. The new process now just calls the
6442 queue-runner function directly.
6448 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6451 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6452 ignored. Though the use of -oP was forcing the writing of a pid file, it
6453 was always written to the default place.
6455 3. If the message "no IP address found for host xxxx" is generated during
6456 incoming verification, it is now followed by identification of the incoming
6457 connection (so you can more easily find what provoked it).
6459 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6461 5. Added some features to "harden" Exim a bit more against certain attacks:
6463 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6464 be put in Local/Makefile. This is like the never_users runtime option,
6465 but it cannot be overridden. The default setting is "root".
6467 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6468 prefix string with which any file named in a -C command line option
6471 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6472 is retained for -C and -D only if the caller of Exim is root. Without
6473 it, the exim user may also use -C and -D and retain privilege.
6475 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6476 command line option is disabled.
6478 6. Macro names set by the -D option must start with an upper case letter, just
6479 like macro names defined in the configuration file.
6481 7. Added "dereference=" facility to LDAP.
6483 8. Two instances of the typo "uknown" in the source files are fixed.
6485 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6486 the Configure-Makefile script screwed up while processing it.
6488 10. Incorporated PCRE 4.4.
6490 11. The SMTP synchronization check was not operating right at the start of an
6491 SMTP session. For example, it could not catch a HELO sent before the client
6492 waited for the greeting. There is now a check for outstanding input at the
6493 point when the greeting is written. Because of the duplex, asynchronous
6494 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6495 way, but not yet received, when the check is performed.
6497 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6498 on TCP/IP sockets, because this apparently causes some broken clients to
6501 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6502 unchanged) from the Cygwin maintainer.
6504 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6505 was supported instead of testing for mbx. Effectively a typo.
6507 15. The spa authenticator server code was not checking that the input it
6508 received was valid base64.
6510 16. The debug output line for the "set" modifier in ACLs was not showing the
6511 name of the variable that was being set.
6513 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6515 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6516 lookup on its own. Something else was needed to trigger the lookup. For
6517 example, a match in host_lookup or the need for a host name in a host list.
6518 Now, if $sender_host_name is referenced and the host name has not yet been
6519 looked up, a lookup is performed. If the lookup fails, the variable remains
6520 empty, and $host_lookup_failed is set to "1".
6522 19. Added "eqi" as a case-independent comparison operator.
6524 20. The saslauthd authentication condition could segfault if neither service
6525 nor realm was specified.
6527 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6528 error message that was logged was misleading, and incoming SMTP
6529 connections were dropped. The message is now more accurate, and temporary
6530 errors are given to SMTP connections.
6532 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6533 (except RSET) with a 421 error, until QUIT is received. However, it was
6534 failing to send a response to QUIT.
6536 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6537 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6538 ACL is now run after the helo_try_verify_hosts code.
6540 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6541 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6542 case-independent, so other case variants are also recognized). Apparently
6543 some systems use these upper case variants.
6545 25. If more than two messages were waiting for the same host, and a transport
6546 filter was specified for the transport, Exim sent two messages over the
6547 same TCP/IP connection, and then failed with "socket operation on non-
6548 socket" when it tried to send the third.
6550 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6552 27. The extern definition of crypt16() in expand.c was not being excluded when
6553 the OS had its own crypt16() function.
6555 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6556 as a preferred synonym for return_size_limit, both as an option and as an
6559 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6561 30. Changed the default configuration ACL to relax the local part checking rule
6562 for addresses that are not in any local domains. For these addresses,
6563 slashes and pipe symbols are allowed within local parts, but the sequence
6564 /../ is explicitly forbidden.
6566 31. SPA server authentication was not clearing the challenge buffer before
6569 32. log_message in a "warn" ACL statement was writing to the reject log as
6570 well as to the main log, which contradicts the documentation and doesn't
6571 seem right (because no rejection is happening). So I have stopped it.
6573 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6574 However, I am unable to do any testing of this.
6576 34. Fixed an infelicity in the appendfile transport. When checking directories
6577 for a mailbox, to see if any needed to be created, it was accidentally
6578 using path names with one or more superfluous leading slashes; tracing
6579 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6581 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6582 contents are added to the log line that is written for every discarded
6583 recipient. (Previously a log_message setting was ignored.)
6585 36. The ${quote: operator now quotes the string if it is empty.
6587 37. The install script runs exim in order to find its version number. If for
6588 some reason other than non-existence or emptiness, which it checks, it
6589 could not run './exim', it was installing it with an empty version number,
6590 i.e. as "exim-". This error state is now caught, and the installation is
6593 38. An argument was missing from the function that creates an error message
6594 when Exim fails to connect to the socket for saslauthd authentication.
6595 This could cause Exim to crash, or give a corrupted message.
6597 39. Added isip, isip4, and isip6 to ${if conditions.
6599 40. The ACL variables $acl_xx are now saved with the message, and can be
6600 accessed later in routers, transports, and filters.
6602 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6603 strings in the file are not string-expanded.
6605 42. If a MAIL command specified a SIZE value that was too large to fit into an
6606 int variable, the check against message_size_limit failed. Such values are
6607 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6608 day this will have to be increased, but I don't think I want to be around
6609 when emails are that large.
6616 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6617 iconv() is not standard in FreeBSD.
6619 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6620 IPv6 enabled. The observed symptom was a segmentation fault on return from
6621 the function os_common_find_running_interfaces() in src/os.c.
6623 3. In the check_special_case() function in daemon.c I had used "errno" as an
6624 argument name, which causes warnings on some systems. This was basically a
6625 typo, since it was named "eno" in the comments!
6627 4. The code that waits for the clock to tick (at a resolution of some fraction
6628 of a second) so as to ensure message-id uniqueness was always waiting for
6629 at least one whole tick, when it could have waited for less. [This is
6630 almost certainly not relevant at current processor speeds, where it is
6631 unlikely to ever wait at all. But we try to future-proof.]
6633 5. The function that sleeps for a time interval that includes fractions of a
6634 second contained a race. It did not block SIGALRM between setting the
6635 timer, and suspending (a couple of lines later). If the interval was short
6636 and the sigsuspend() was delayed until after it had expired, the suspension
6637 never ended. On busy systems this could lead to processes getting stuck for
6640 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6641 process, before it forks any delivery processes. The open lookup caching
6642 mechanism meant that the open file or database connection was passed into
6643 the delivery process. The problem was that delivery processes always tidy
6644 up cached lookup data. This could cause a problem for the next delivery
6645 process started by the queue runner, because the external queue runner
6646 process does not know about the closure. So the next delivery process
6647 still has data in the lookup cache. In the case of a file lookup, there was
6648 no problem because closing a file descriptor in a subprocess doesn't affect
6649 the parent. However, if the lookup was caching a connection to a database,
6650 the connection was closed, and the second delivery process was likely to
6651 see errors such as "PGSQL: query failed: server closed the connection
6652 unexpectedly". The problem has been fixed by closing all cached lookups
6653 in a queue runner before running a delivery process.
6655 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6656 seem to have the "const" qualifier which it has on other OS. I've
6659 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6660 *of the same type* (client or server) with the same public name that an
6661 error should be diagnosed.
6663 9. When Exim looked up a host name for an IP address, but failed to find the
6664 original IP address when looking up the host name (a safety check), it
6665 output the message "<ip address> does not match any IP for NULL", which was
6666 confusing, to say the least. The bug was that the host name should have
6667 appeared instead of "NULL".
6669 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6670 user that is built into the binary, and the -C or -D options is used, root
6671 privilege is dropped before the configuration file is read. In addition,
6672 logging is switched to stderr instead of the normal log files. If the
6673 configuration then re-defines the Exim user, the unprivileged environment
6674 is probably not what is expected, so Exim logs a panic warning message (but
6677 However, if deliver_drop_privilege is set, the unprivileged state may well
6678 be exactly what is intended, so the warning has been cut out in that case,
6679 and Exim is allowed to try to write to its normal log files.
6685 1. smtp_return_error_details was not giving details for temporary sender
6686 or receiver verification errors.
6688 2. Diagnose a configuration error if two authenticators have the same public
6691 3. Exim used not to create the message log file for a message until the first
6692 delivery attempt. This could be confusing when incoming messages were held
6693 for policy or load reasons. The message log file is now created at the time
6694 the message is received, and an initial "Received" line is written to it.
6696 4. The automatically generated man page for command line options had a minor
6697 bug that caused no ill effects; however, a more serious problem was that
6698 the procedure for building the man page automatically didn't always
6699 operate. Consequently, release 4.20 contains an out-of-date version. This
6700 shouldn't happen again.
6702 5. When building Exim with embedded Perl support, the script that builds the
6703 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6704 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6706 6. The freeze_tell option was not being used for messages that were frozen on
6707 arrival, either by an ACL or by local_scan().
6709 7. Added the smtp_incomplete_transaction log selector.
6711 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6712 was accepting AUTH without a new EHLO.
6714 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6715 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6718 10. Logging of TCP/IP connections (when configured) now happens in the main
6719 daemon process instead of the child process, so that the TCP/IP connection
6720 count is more accurate (but it can never be perfect).
6722 11. The use of "drop" in a nested ACL was not being handled correctly in the
6723 outer ACL. Now, if condition failure induced by the nested "drop" causes
6724 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6725 or "require"), the connection is dropped.
6727 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6728 that yield "discard" can now be used with an "accept" or a "discard" verb,
6729 but an error is generated for any others (because I can't see a useful way
6730 to define what should happen).
6732 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6733 are now processed in the same way as lines in the Exim configuration file.
6734 In particular, continuation lines are supported.
6736 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6738 15. Added -ti meaning -t -i.
6740 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6741 domains, and warn by logging if others are found.
6743 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6744 returned from GIFCONF when Exim is trying to find the list of interfaces on
6745 a host. The code in os.c has been modified to copy each ifreq to an aligned
6746 structure in all cases.
6748 Also, in some cases, the returned ifreq's were being copied to a 'struct
6749 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6750 means the last couple of bytes of an IPv6 address could be chopped if the
6751 ifreq contained only a normal sockaddr (14 bytes storage).
6753 18. Named domain lists were not supported in the hosts_treat_as_local option.
6754 An entry such as +xxxx was not recognized, and was treated as a literal
6757 19. Ensure that header lines added by a DATA ACL are included in the reject log
6758 if the ACL subsequently rejects the message.
6760 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6761 MD5 (which is deprecated).
6763 21. When testing a filter file using -bf, Exim was writing a message when it
6764 took the sender from a "From " line in the message, but it was not doing so
6765 when it took $return_path from a Return-Path: header line. It now does.
6767 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6768 with a valid header line field name (a series of printing characters
6769 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6771 23. Changed "disc" in the source to "disk" to conform to the documentation and
6772 the book and for uniformity.
6774 24. Ignore Sendmail's -Ooption=value command line item.
6776 25. When execve() failed while trying to run a command in a pipe transport,
6777 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6778 could be confused with a return value of 69 from the command itself. This
6779 has been changed to 127, the value the shell returns if it is asked to run
6780 a non-existent command. The wording for the related log line suggests a
6781 non-existent command as the problem.
6783 26. If received_header_text expands to an empty string, do not add a Received:
6784 header line to the message. (Well, it adds a token one on the spool, but
6785 marks it "old" so that it doesn't get used or transmitted.)
6787 27. Installed eximstats 1.28 (addition of -nt option).
6789 28. There was no check for failure on the call to getsockname() in the daemon
6790 code. This can fail if there is a shortage of resources on the system, with
6791 ENOMEM, for example. A temporary error is now given on failure.
6793 29. Contrary to the C standard, it seems that in some environments, the
6794 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6795 program. Exim now does this explicitly; it affects the formatting of
6796 timestamps using strftime().
6798 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6799 complaints. I've now initialized all the variables, to avoid this.
6801 32. Header lines added by a system filter were not being "seen" during
6802 transport-time rewrites.
6804 33. The info_callback() function passed to OpenSSL is set up with type void
6805 (*)(SSL *, int, int), as described somewhere. However, when calling the
6806 function (actually a macro) that sets it up, the type void(*)() is
6807 expected. I've put in a cast to prevent warnings from picky compilers.
6809 34. If a DNS black list lookup found a CNAME record, but there were no A
6810 records associated with the domain it pointed at, Exim crashed.
6812 35. If a DNS black list lookup returned more than one A record, Exim ignored
6813 all but the first. It now scans all returned addresses if a particular IP
6814 value is being sought. In this situation, the contents of the
6815 $dnslist_value variable are a list of all the addresses, separated by a
6818 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6819 to accept a host name and all its aliases if the forward lookup for any of
6820 them yielded the IP address of the incoming connection. Now it accepts only
6821 those names whose forward lookup yields the correct IP address. Any other
6822 names are discarded. This closes a loophole whereby a rogue DNS
6823 administrator could create reverse DNS records to break through a
6824 wildcarded host restriction in an ACL.
6826 37. If a user filter or a system filter that ran in a subprocess used any of
6827 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6828 the wrong values were passed to the pipe command ($thisaddress had the
6829 value of $0, $0 had the value of $1, etc). This bug was introduced by
6830 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6832 38. Improved the line breaking for long SMTP error messages from ACLs.
6833 Previously, if there was no break point between 40 and 75 characters, Exim
6834 left the rest of the message alone. Two changes have been made: (a) I've
6835 reduced the minimum length to 35 characters; (b) if it can't find a break
6836 point between 35 and 75 characters, it looks ahead and uses the first one
6837 that it finds. This may give the occasional overlong line, but at least the
6838 remaining text gets split now.
6840 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6841 file descriptors might be low, and that setting 1000 would always raise it.
6842 It turns out that in some environments, the limit is already over 1000 and
6843 that lowering it causes trouble. So now Exim takes care not to decrease it.
6845 40. When delivering a message, the value of $return_path is set to $sender_
6846 address at the start of routing (routers may change the value). By an
6847 oversight, this default was not being set up when an address was tested by
6848 -bt or -bv, which affected the outcome if any router or filter referred to
6851 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6852 log only when "message" or "log_message" are set. However, if one of the
6853 conditions was an address verification, or a call to a nested ACL, the
6854 messages generated by the underlying test were being passed through. This
6855 no longer happens. The underlying message is available in $acl_verify_
6856 message for both "message" and "log_message" expansions, so it can be
6857 passed through if needed.
6859 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6860 new expansion $bh_ to give the encoded byte string without charset
6861 translation. Translation happens only if iconv() is available; HAVE_ICONV
6862 indicates this at build time. HEADERS_CHARSET gives the charset to
6863 translate to; headers_charset can change it in the configuration, and
6864 "headers charset" can change it in an individual filter file.
6866 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6867 that creates RFC 2047 encoded "words" labels them as that charset instead
6868 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6869 expansion operator; (ii) when Exim creates a From: line for a local
6870 message; (iii) when a header line is rewritten to include a "phrase" part.
6872 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6873 buggy, causing it to skip the first lines of messages whose message ID
6874 ended in 'D'. This would not have bitten before Exim release 4.14, because
6875 message IDs were unlikely to end in 'D' before then. The effect was to have
6876 incorrect size information for certain domains.
6878 45. #include "config.h" was missing at the start of the crypt16.c module. This
6879 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6882 46. If there was a timeout during a "random" callout check, Exim treated it as
6883 a failure of the random address, and carried on sending RSET and the real
6884 address. If the delay was just some slowness somewhere, the response to the
6885 original RCPT would be taken as a response to RSET and so on, causing
6886 mayhem of various kinds.
6888 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6889 when I implemented it. It didn't allow for the fact that some option values
6890 may legitimately be negative (e.g. size_addition), and it didn't even do
6891 the right test for positive values.
6893 48. Domain names in DNS records are case-independent. Exim always looks them up
6894 in lower case. Some resolvers return domain names in exactly the case they
6895 appear in the zone file, that is, they may contain uppercase letters. Not
6896 all resolvers do this - some return always lower case. Exim was treating a
6897 change of case by a resolver as a change of domain, similar to a widening
6898 of a domain abbreviation. This triggered its re-routing code and so it was
6899 trying to route what was effectively the same domain again. This normally
6900 caused routing to fail (because the router wouldn't handle the domain
6901 twice). Now Exim checks for this case specially, and just changes the
6902 casing of the domain that it ultimately uses when it transmits the message
6905 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6908 50. If a filter generated a file delivery with a non-absolute name (possible if
6909 no home directory exists for the router), the forbid_file option was not
6912 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6913 the existing equality matching.
6915 52. Exim was using ints instead of ino_t variables in some places where it was
6916 dealing with inode numbers.
6918 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6919 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6920 called TMPDIR, and if it finds it is different, it changes its value.
6922 54. The smtp_printf() function is now made available to local_scan() so
6923 additional output lines can be written before returning. There is also an
6924 smtp_fflush() function to enable the detection of a dropped connection.
6925 The variables smtp_input and smtp_batched_input are exported to
6928 55. Changed the default runtime configuration: the message "Unknown user"
6929 has been removed from the ACL, and instead placed on the localuser router,
6930 using the cannot_route_message feature. This means that any verification
6931 failures that generate their own messages won't get overridden. Similarly,
6932 the "Unrouteable address" message that was in the ACL for unverifiable
6933 relay addresses has also been removed.
6935 56. Added hosts_avoid_esmtp to the smtp transport.
6937 57. The exicyclog script was not checking for the esoteric option
6938 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6939 will work only if exicyclog is run under the appropriate euid.
6941 58. Following a discussion on the list, the rules by which Exim recognises line
6942 endings on incoming messages have been changed. The -dropcr and drop_cr
6943 options are now no-ops, retained only for backwards compatibility. The
6944 following line terminators are recognized: LF CRLF CR. However, special
6945 processing applies to CR:
6947 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6948 nor a local message in the state where . is a terminator.
6950 (ii) If a bare CR is encountered in a header line, an extra space is added
6951 after the line terminator so as not to end the header. The reasoning
6952 behind this is that bare CRs in header lines are most likely either
6953 to be mistakes, or people trying to play silly games.
6955 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6956 was being incorrectly given as 18 bytes larger than it should have been.
6957 This is a VOB (very old bug).
6959 60. This may never have affected anything current, but just in case it has:
6960 When the local host is found other than at the start of a list of hosts,
6961 the local host, those with the same MX, and any that follow, are discarded.
6962 When the list in question was part of a longer list of hosts, the following
6963 hosts (not currently being processed) were also being discarded. This no
6964 longer happens. I'm not sure if this situation could ever has previously
6967 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6970 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6971 header. This is something that is recommended in a new Internet Draft, and
6972 is something that is documented as being done by Sendmail. There are two
6973 possible values. For messages generated by the autoreply transport, Exim
6976 Auto-Submitted: auto-replied
6978 whereas for all other generated messages (e.g. bounces) it adds
6980 Auto-Submitted: auto-generated
6982 63. The "personal" condition in filters now includes a test for the
6983 Auto-Submitted: header. If it contains the string "auto-" the message it
6984 not considered personal.
6986 64. Added rcpt_include_affixes as a generic transport option.
6988 65. Added queue_only_override (default true).
6990 66. Added the syslog_duplication option.
6992 67. If what should have been the first header line of a message consisted of
6993 a space followed by a colon, Exim was mis-interpreting it as a header line.
6994 It isn't of course - it is syntactically invalid and should therefore be
6995 treated as the start of the message body. The misbehaviour could have
6996 caused a number of strange effects, including loss of data in subsequent
6997 header lines, and spool format errors.
6999 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7000 client host had authenticated. This control can now be exercised by an ACL
7001 for more flexibility.
7003 69. By default, callouts do not happen when testing with -bh. There is now a
7004 variant, -bhc, which does actually run the callout code, including
7005 consulting and updating the callout cache.
7007 70. Added support for saslauthd authentication, courtesy of Alexander
7010 71. If statvfs() failed on the spool or log directories while checking their
7011 size for availability, Exim confusingly gave the error "space shortage".
7012 Furthermore, in debugging mode it crashed with a floating point exception.
7013 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7014 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7015 serious problem, Exim now writes to the main and panic logs when this
7016 happens, with details of the failure. It then refuses to accept the
7017 incoming message, giving the message "spool directory problem" or "log
7018 directory problem" with a 421 code for SMTP messages.
7020 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7021 0, 1, and 2 exist, because some OS complain for execs without them (see
7022 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7023 descriptors. However, the code omitted to check that the open succeeded,
7024 causing mysterious errors if for some reason the permissions on /dev/null
7025 got screwed. Now Exim writes a message to the main and panic logs, and
7026 bombs out if it can't open /dev/null.
7028 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7029 interact so that it is all more flexible. It is supposed to remain
7030 backwards compatible. Also added extra_local_interfaces.
7032 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7033 to bomb out with an assertion failure - to the client this appears as a
7034 connection drop. This problem occurs in the part of the code that was taken
7035 from the Samba project. Fortunately, the assertion is in a very simple
7036 function, so I have fixed this by reproducing the function inline in the
7037 one place where it is called, and arranging for authentication to fail
7038 instead of killing the process with assert().
7040 75. The SPA client code was not working when the server requested OEM rather
7041 than Unicode encoding.
7043 76. Added code to make require_files with a specific uid setting more usable in
7044 the case where statting the file as root fails - usually a non-root-mounted
7045 NFS file system. When this happens and the failure is EACCES, Exim now
7046 forks a subprocess and does the per-uid checking as the relevant uid.
7048 77. Added process_log_path.
7050 78. If log_file_path was not explicitly set, a setting of check_log_space or
7051 check_log_inodes was ignored.
7053 79. If a space check for the spool or log partitions fails, the incident is now
7054 logged. Of course, in the latter case the data may get lost...
7056 80. Added the %p formatting code to string_format() so that it can be used to
7057 print addresses in debug_print(). Adjusted all the address printing in the
7058 debugging in store.c to use %p rather than %d.
7060 81. There was a concern that a line of code in smtp_in.c could overflow a
7061 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7062 initially expressed, the concern was not well-founded, because trailing
7063 spaces are removed early. However, if the trailing spaces were followed by
7064 a NULL, they did not get removed, so the overflow was possible. Two fixes
7067 (a) I re-wrote the offending code in a cleaner fashion.
7068 (b) If an incoming SMTP command contains a NULL character, it is rejected
7071 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7072 runs initgroups(), so that if the Exim user is in any additional groups,
7073 they will be used during message reception.
7079 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob