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 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
12 When redis_servers is set to a list of > 1 element, and the Redis servers
13 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
14 case of MOVED into a DEFER case instead, thus moving the query onto the
15 next server in the list. For a cluster of N elements, all N servers must
16 be defined in redis_servers.
18 JH/01 Replace the store_release() internal interface with store_newblock(),
19 which internalises the check required to safely use the old one, plus
20 the allocate and data copy operations duplicated in both (!) of the
23 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
24 modifier. This matches the restriction on the commandline.
26 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
27 Previously only the last row was returned.
29 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
30 we assumed that tags in the header were well-formed, and parsed the
31 element content after inspecting only the first char of the tag.
32 Assumptions at that stage could crash the receive process on malformed
35 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
36 While running the DKIM ACL we operate on the Permanent memory pool so that
37 variables created with "set" persist to the DATA ACL. Also (at any time)
38 DNS lookups that fail create cache records using the Permanent pool. But
39 expansions release any allocations made on the current pool - so a dnsdb
40 lookup expansion done in the DKIM ACL releases the memory used for the
41 DNS negative-cache, and bad things result. Solution is to switch to the
42 Main pool for expansions.
43 While we're in that code, add checks on the DNS cache during store_reset,
44 active in the testsuite.
45 Problem spotted, and debugging aided, by Wolfgang Breyha.
47 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
48 When none of the hosts presented to a transport match an already-open
49 connection, close it and proceed with the list. Previously we would
50 queue the message. Spotted by Lena with Yahoo, probably involving
53 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
54 Previously a spurious "250 OK id=" response was appended to the proper
57 JH/08 The "support for" informational output now, which built with Content
58 Scanning support, has a line for the malware scanner interfaces compiled
59 in. Interface can be individually included or not at build time.
61 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
62 by the template makefile "src/EDITME". The "STREAM" support for an older
63 ClamAV interface method is removed.
65 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
66 rows affected is given instead).
68 JH/11 The runtime Berkeley DB library version is now additionally output by
69 "exim -d -bV". Previously only the compile-time version was shown.
71 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
72 SMTP connection. Previously, when one had more receipients than the
73 first, an abortive onward connection was made. Move to full support for
74 multiple onward connections in sequence, handling cutthrough connection
75 for all multi-message initiating connections.
77 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
78 routers. Previously, a multi-recipient message would fail to match the
79 onward-connection opened for the first recipient, and cause its closure.
81 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
82 a timeout on read on a GnuTLS initiating connection, resulting in the
83 initiating connection being dropped. This mattered most when the callout
84 was marked defer_ok. Fix to keep the two timeout-detection methods
87 JH/15 Relax results from ACL control request to enable cutthrough, in
88 unsupported situations, from error to silently (except under debug)
89 ignoring. This covers use with PRDR, frozen messages, queue-only and
92 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
94 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
95 metadata, resulting in a crash in free().
97 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
98 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
99 Broken also in d185889f4, with init system revamp.
101 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
102 Previously we abruptly closed the connection after reading a malware-
103 found indication; now we go on to read the "scan ok" response line,
106 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
107 ACL. Previously, a crash would result.
109 JH/19 Speed up macro lookups during configuration file read, by skipping non-
110 macro text after a replacement (previously it was only once per line) and
111 by skipping builtin macros when searching for an uppercase lead character.
113 JH/20 DANE support moved from Experimental to mainline. The Makefile control
114 for the build is renamed.
116 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
117 was allocated for every new TLS startup, meaning one per message. Fix
118 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
120 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
121 reported the original. Fix to report (as far as possible) the ACL
122 result replacing the original.
124 JH/23 Fix memory leak during multi-message connections using STARTTLS under
125 OpenSSL. Certificate information is loaded for every new TLS startup,
126 and the resources needed to be freed.
128 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
130 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
133 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
134 DATA response info to the (existing) per-recipient response info for
135 the "C=" log element. It can have useful tracking info from the
136 destination system. Patch from Simon Arlott.
138 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
139 length value. Previously this would segfault.
141 HS/02 Support Avast multiline protoocol, this allows passing flags to
142 newer versions of the scanner.
144 JH/28 Ensure that variables possibly set during message acceptance are marked
145 dead before release of memory in the daemon loop. This stops complaints
146 about them when the debug_store option is enabled. Discovered specifically
147 for sender_rate_period, but applies to a whole set of variables.
148 Do the same for the queue-runner loop, for variables set from spool
151 PP/02 DANE: add dane_require_tls_ciphers SMTP Transport option; if unset,
152 tls_require_ciphers is used as before.
158 JH/01 Rework error string handling in TLS interface so that the caller in
159 more cases is responsible for logging. This permits library-sourced
160 string to be attached to addresses during delivery, and collapses
161 pairs of long lines into single ones.
163 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
164 during configuration. Wildcards are allowed and expanded.
166 JH/02 Rework error string handling in DKIM to pass more info back to callers.
167 This permits better logging.
169 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
170 do not close it down and have the child transport start it up again on
171 the passed-on TCP connection. Instead, proxy the child (and any
172 subsequent ones) for TLS via a unix-domain socket channel. Logging is
173 affected: the continued delivery log lines do not have any DNSSEC, TLS
174 Certificate or OCSP information. TLS cipher information is still logged.
176 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
177 identical IP addresses on different listening ports. Will also affect
180 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
181 add noisy ifdef guards to special-case this sillyness.
182 Patch from Bernd Kuhls.
184 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
185 than 255 are no longer allowed.
187 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
188 Disable the session-cache too, which might reduce our load. Since we
189 currrectly use a new context for every connection, both as server and
190 client, there is no benefit for these.
191 GnuTLS appears to not support tickets server-side by default (we don't
192 call gnutls_session_ticket_enable_server()) but client side is enabled
193 by default on recent versions (3.1.3 +) unless the PFS priority string
196 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
197 <https://reproducible-builds.org/specs/source-date-epoch/>.
199 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
200 the check for any unsuccessful recipients did not notice the limit, and
201 erroneously found still-pending ones.
203 JH/08 Pipeline CHUNKING command and data together, on kernels that support
204 MSG_MORE. Only in-clear (not on TLS connections).
206 JH/09 Avoid using a temporary file during transport using dkim. Unless a
207 transport-filter is involved we can buffer the headers in memory for
208 creating the signature, and read the spool data file once for the
209 signature and again for transmission.
211 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
212 4.77 as the kernel support then wasn't solid, having issues in 64bit
213 mode. Now, it's been long enough. Add support for FreeBSD also.
215 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
216 case where the routing stage had gathered several addresses to send to
217 a host before calling the transport for the first, we previously failed
218 to close down TLS in the old transport process before passing the TCP
219 connection to the new process. The new one sent a STARTTLS command
220 which naturally failed, giving a failed delivery and bloating the retry
221 database. Investigation and fix prototype from Wolfgang Breyha.
223 JH/12 Fix check on SMTP command input synchronisation. Previously there were
224 false-negatives in the check that the sender had not preempted a response
225 or prompt from Exim (running as a server), due to that code's lack of
226 awareness of the SMTP input buffering.
228 PP/04 Add commandline_checks_require_admin option.
229 Exim drops privileges sanely, various checks such as -be aren't a
230 security problem, as long as you trust local users with access to their
231 own account. When invoked by services which pass untrusted data to
232 Exim, this might be an issue. Set this option in main configuration
233 AND make fixes to the calling application, such as using `--` to stop
236 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
237 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
238 responses to those, into a single TLS record each way (this usually means
239 a single packet). As a side issue, smtp_enforce_sync now works on TLS
242 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
243 affects you only if you're dancing at the edge of the param size limits.
244 If you are, and this message makes sense to you, then: raise the
245 configured limit or use OpenSSL 1.1. Nothing we can do for older
248 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
249 cmdline element to get the documented default one. Previously it was
252 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
253 are unsuable for non-message-specific reasons - eg. connection timeout,
254 banner-time rejection.
256 JH/16 Fix logging of delivery remote port, when specified by router, under
259 PP/06 Repair manualroute's ability to take options in any order, even if one
260 is the name of a transport.
263 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
265 JH/17 Change the list-building routines interface to use the expanding-string
266 triplet model, for better allocation and copying behaviour.
268 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
269 Previously it was constructed the first time a possibly-matching string
270 was met in the configuration file input during startup; now it is done
273 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
274 compatible one, to avoid the (poorly documented) possibility of a config
275 file in the working directory redirecting the DB files, possibly correpting
276 some existing file. CVE-2017-10140 assigned for BDB.
278 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
279 cache-hot. Previously, although the result was properly cached, the
280 initial verify call returned a defer.
282 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
283 the main verify for receipient in uncached-mode.
285 JH/22 Retire historical build files to an "unsupported" subdir. These are
286 defined as "ones for which we have no current evidence of testing".
288 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
289 if present. Previously it was ignored.
291 JH/24 Start using specified-initialisers in C structure init coding. This is
292 a C99 feature (it's 2017, so now considered safe).
294 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
295 if was a fixed-sized field and bitmask ops via macros; it is now more
298 PP/07 GitHub PR 56: Apply MariaDB build fix.
299 Patch provided by Jaroslav Škarvada.
301 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
302 during Coverity cleanups [4.87 JH/47]
303 Diagnosis and fix provided by Michael Fischer v. Mollard.
305 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
306 the right size to place the terminating semicolon on its own folded
307 line, the header hash was calculated to an incorrect value thanks to
308 the (relaxed) space the fold became.
310 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
311 and confused the parent.
313 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
314 which could crash as a result. This could lead to undeliverable messages.
316 JH/28 Logging: "next input sent too soon" now shows where input was truncated
319 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
320 matters on fast-turnover and PID-randomising systems, which were getting
321 out-of-order delivery.
323 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
324 a possibly-overlapping copy. The symptom was that "Remote host closed
325 connection in response to HELO" was logged instead of the actual 4xx
328 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
329 Previously only that bufferd was discarded, resulting in SYMTP command
332 JH/32 DKIM: when a message has multiple signatures matching an identity given
333 in dkim_verify_signers, run the dkim acl once for each. Previously only
334 one run was done. Bug 2189.
336 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
337 "panic the current process" to "deliberately defer". The panic log is
338 still written with the problem list name; the mail and reject logs now
339 get a temp-reject line for the message that was being handled, saying
340 something like "domains check lookup or other defer". The SMTP 451
341 message is still "Temporary local problem".
343 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
344 A crafted sequence of BDAT commands could result in in-use memory beeing
345 freed. CVE-2017-16943.
347 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
348 from SMTP input. Previously it was always done; now only done for DATA
349 and not BDAT commands. CVE-2017-16944.
351 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
352 to the message (such as an overlong header line). Previously this was
353 not done and we did not exit BDAT mode. Followon from the previous item
354 though a different problem.
360 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
361 than -2003 did; needs libidn2 in addition to libidn.
363 JH/02 The path option on a pipe transport is now expanded before use.
365 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
366 Patch provided by "Björn", documentation fix added too.
368 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
369 missing a wire-to-host endian conversion.
371 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
372 close after a BDAT command line could be taken as a following command,
373 giving a synch failure. Fix by only checking for synch immediately
374 before acknowledging the chunk.
376 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
377 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
378 macro. Patches provided by Josh Soref.
380 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
381 Previously we did not; the RFC seems ambiguous and VRFY is not listed
382 by IANA as a service extension. However, John Klensin suggests that we
385 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
386 the dkim code may be unix-mode line endings rather than smtp wire-format
387 CRLF, so prepend a CR to any bare LF.
389 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
390 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
392 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
393 the first were themselves being wrongly included in the feed into dkim
394 processing; with most chunk sizes in use this resulted in an incorrect
395 body hash calculated value.
397 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
398 DKIM signature block, for verification. Although advised against by
399 standards it is specifically not ruled illegal.
401 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
403 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
404 missing a body hash (the bh= tag).
406 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
407 It seems that HAProxy sends the Proxy Protocol information in clear and
408 only then does a TLS startup, so do the same.
410 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
411 TCP connections (such as for Spamd) unless the daemon successfully set
412 Fast Open mode on its listening sockets. This fixes breakage seen on
413 too-old kernels or those not configured for Fast Open, at the cost of
414 requiring both directions being enabled for TFO, and TFO never being used
415 by non-daemon-related Exim processes.
417 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
418 endings, at least on the first header line. Try to canonify any that get
419 past that check, despite the cost.
421 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
422 now limited to an arbitrary five deep, while parsing addresses with the
423 strip_excess_angle_brackets option enabled.
425 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
426 instead leave the unprompted TLS handshake in socket buffer for the
427 TLS library to consume.
429 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
431 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
433 JH/16 Drop variables when they go out of scope. Memory management drops a whole
434 region in one operation, for speed, and this leaves assigned pointers
435 dangling. Add checks run only under the testsuite which checks all
436 variables at a store-reset and panics on a dangling pointer; add code
437 explicitly nulling out all the variables discovered. Fixes one known
438 bug: a transport crash, where a dangling pointer for $sending_ip_address
439 originally assigned in a verify callout, is re-used.
441 PP/06 Drop '.' from @INC in various Perl scripts.
443 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
445 PP/08 Reduce a number of compilation warnings under clang; building with
446 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
447 should be warning-free.
449 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
451 HS/01 Fix portability problems introduced by PP/08 for platforms where
452 realloc(NULL) is not equivalent to malloc() [SunOS et al].
454 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
455 chunk. This allows us to accept broken chunked messages. We need a more
456 general solution here.
458 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
459 already-broken messages in the queue.
461 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
463 JH/19 Fix reference counting bug in routing-generated-address tracking.
469 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
470 supports it and a size is available (ie. the sending peer gave us one).
472 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
473 years of being deprecated). The replacements are the ACLs
474 acl_smtp_mime and acl_not_smtp_mime.
476 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
477 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
478 or even an in-clear connection were permitted. Now, if the host lookup
479 was dnssec and dane was requested then the host is only used if the
480 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
481 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
482 if one fails this test.
483 This means that a poorly-configured remote DNS will make it incommunicado;
484 but it protects against a DNS-interception attack on it.
486 JH/04 Bug 1810: make continued-use of an open smtp transport connection
487 non-noisy when a race steals the message being considered.
489 JH/05 If main configuration option tls_certificate is unset, generate a
490 self-signed certificate for inbound TLS connections.
492 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
493 in rewrites and routers.
495 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
496 and logged a warning sing 4.83; now they are a configuration file error.
498 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
499 (lacking @domain). Apply the same qualification processing as RCPT.
501 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
503 JH/10 Support ${sha256:} applied to a string (as well as the previous
506 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
507 a cutthrough deliver is pending, as we always want to make a connection.
508 This also avoids re-routing the message when later placing the cutthrough
509 connection after a verify cache hit.
510 Do not update it with the verify result either.
512 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
513 when routing results in more than one destination address.
515 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
516 signing (which inhibits the cutthrough capability). Previously only
517 the presence of an option was tested; now an expansion evaluating as
518 empty is permissible (obviously it should depend only on data available
519 when the cutthrough connection is made).
521 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
522 the relevant preceding SMTP command did not note the pipelining mode.
524 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
525 Previously they were not counted.
527 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
528 as one having no matching records. Previously we deferred the message
529 that needed the lookup.
531 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
532 distinguished as "(=".
534 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
535 for missing MX records. Previously it only worked for missing A records.
537 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
539 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
540 after the data-go-ahead and data-ack. Patch from Jason Betts.
542 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
543 even for a "none" policy. Patch from Tony Meyer.
545 JH/22 Fix continued use of a connection for further deliveries. If a port was
546 specified by a router, it must also match for the delivery to be
549 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
550 When one of the recipients of a message was unsuitable for the connection
551 (has no matching addresses), we lost track of needing to mark it
552 deferred. As a result mail would be lost.
554 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
556 JH/25 Decoding ACL controls is now done using a binary search; the source code
557 takes up less space and should be simpler to maintain. Merge the ACL
558 condition decode tables also, with similar effect.
560 JH/26 Fix problem with one_time used on a redirect router which returned the
561 parent address unchanged. A retry would see the parent address marked as
562 delivered, so not attempt the (identical) child. As a result mail would
565 JH/27 Fix a possible security hole, wherein a process operating with the Exim
566 UID can gain a root shell. Credit to http://www.halfdog.net/ for
567 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
570 JH/28 Enable {spool,log} filesystem space and inode checks as default.
571 Main config options check_{log,spool}_{inodes,space} are now
572 100 inodes, 10MB unless set otherwise in the configuration.
574 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
575 Previously it only applied to the main-section connection policy
578 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
580 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
581 by me. Added RFC7919 DH primes as an alternative.
583 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
584 are not in the system include path.
586 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
587 GnuTLS, when a session startup failed (eg because the client disconnected)
588 Exim did stdio operations after fclose. This was exposed by a recent
589 change which nulled out the file handle after the fclose.
591 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
592 signed directly by the cert-signing cert, rather than an intermediate
593 OCSP-signing cert. This is the model used by LetsEncrypt.
595 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
597 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
598 an incoming connection.
600 HS/02 Bug 1802: Do not half-close the connection after sending a request
603 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
604 fallback to "prime256v1".
606 JH/34 SECURITY: Use proper copy of DATA command in error message.
607 Could leak key material. Remotely exploitable. CVE-2016-9963.
613 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
614 and 3.4.4 - once the server is enabled to respond to an OCSP request
615 it does even when not requested, resulting in a stapling non-aware
616 client dropping the TLS connection.
618 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
619 support variable-length bit vectors. No functional change.
621 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
622 The I= interface field on outgoing lines is now after the H= remote
623 host field, same as incoming lines. There is a separate
624 outgoing_interface log selector which allows you to disable the
627 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
628 If not running log_selector +smtp_connection the mainlog would be held
629 open indefinitely after a "too many connections" event, including to a
630 deleted file after a log rotate. Leave the per net connection logging
631 leaving it open for efficiency as that will be quickly detected by the
632 check on the next write.
634 HS/01 Bug 1671: Fix post transport crash.
635 Processing the wait-<transport> messages could crash the delivery
636 process if the message IDs didn't exist for some reason. When
637 using 'split_spool_directory=yes' the construction of the spool
638 file name failed already, exposing the same netto behaviour.
640 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
641 mime_regex ACL conditions.
643 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
644 to DSN fail messages (bounces): remote IP, remote greeting, remote response
645 to HELO, local diagnostic string.
647 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
648 log line to debug. Even when configured with a tls authenticator many
649 client connections are expected to not authenticate in this way, so
650 an authenticate fail is not an error.
652 HS/02 Add the Exim version string to the process info. This way exiwhat
653 gives some more detail about the running daemon.
655 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
656 matter for fast-change records such as DNSBLs.
658 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
659 retry record, even if constant. There may be multiple transports with
660 different interface settings and the retry behaviour needs to be kept
663 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
665 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
667 JH/10 Bug 840: fix log_defer_output option of pipe transport
669 JH/11 Bug 830: use same host for all RCPTS of a message, even under
670 hosts_randomize. This matters a lot when combined with mua_wrapper.
672 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
673 ${quote_pgsql:<string>} operator.
675 JH/13 Bug 1708: avoid misaligned access in cached lookup.
677 JH/14 Change header file name for freeradius-client. Relevant if compiling
678 with Radius support; from the Gentoo tree and checked under Fedora.
680 JH/15 Bug 1712: Introduce $prdr_requested flag variable
682 JH/16 Bug 1714: Permit an empty string as expansion result for transport
683 option transport_filter, meaning no filtering.
685 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
687 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
688 defaults to "*" (all hosts). The variable is now available when not built
689 with TLS, default unset, mainly to enable keeping the testsuite sane.
690 If a server certificate is not supplied (via tls_certificate) an error is
691 logged, and clients will find TLS connections fail on startup. Presumably
692 they will retry in-clear.
693 Packagers of Exim are strongly encouraged to create a server certificate
694 at installation time.
696 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
697 with the $config_file variable.
699 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
700 in transport context, after the attempt, and per-recipient. The latter type
701 is per host attempted. The event data is the error message, and the errno
702 information encodes the lookup type (A vs. MX) used for the (first) host,
703 and the trailing two digits of the smtp 4xx response.
705 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
706 to write to mainlog (or rejectlog, paniclog) in the window between file
707 creation and permissions/ownership being changed. Particularly affects
708 installations where exicyclog is run as root, rather than exim user;
709 result is that the running daemon panics and dies.
711 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
713 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
714 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
715 "pri" and "weight". Note that the previous implicit priority given by the
716 list order is no longer honoured.
718 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
721 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
722 by defining SUPPORT_SOCKS.
724 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
725 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
726 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
727 variables are renamed to proxy_{local,external}_{address,port}.
729 JH/27 Move Internationalisation support from Experimental to mainline, enabled
730 for a build by defining SUPPORT_I18N
732 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
733 of the query string, and make ${quote_redis:} do that quoting.
735 JH/29 Move Events support from Experimental to mainline, enabled by default
736 and removable for a build by defining DISABLE_EVENT.
738 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
740 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
741 cached by the daemon.
743 JH/32 Move Redis support from Experimental to mainline, enabled for a build
744 by defining LOOKUP_REDIS. The libhiredis library is required.
746 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
747 keys are given for lookup.
749 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
750 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
751 only supported when built with TLS support. The PolarSSL SHA routines
752 are still used when the TLS library is too old for convenient support.
754 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
755 openssl_options), for security. OpenSSL forces this from version 1.1.0
756 server-side so match that on older versions.
758 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
759 allocation for $value could be released as the expansion processing
760 concluded, but leaving the global pointer active for it.
762 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
763 and to use the domains and local_parts ACL conditions.
765 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
766 incorrectly not doubled on cutthrough transmission, hence seen as a
767 body-termination at the receiving system - resulting in truncated mails.
768 Commonly the sender saw a TCP-level error, and retransmitted the message
769 via the normal store-and-forward channel. This could result in duplicates
770 received - but deduplicating mailstores were liable to retain only the
771 initial truncated version.
773 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
775 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
777 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
778 we're in there, support oversigning also; bug 1309.
780 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
782 HS/04 Add support for keep_environment and add_environment options.
784 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
785 either intentional arithmetic overflow during PRNG, or testing config-
788 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
789 delivery resulted in actual delivery. Cancel cutthrough before DATA
792 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
793 rejects a recipient: pass the reject to the originator.
795 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
796 Many were false-positives and ignorable, but it's worth fixing the
799 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
800 for the new environment-manipulation done at startup. Move the routines
801 from being local to tls.c to being global via the os.c file.
803 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
804 an extract embedded as result-arg for a map, the first arg for extract
805 is unavailable so we cannot tell if this is a numbered or keyed
806 extraction. Accept either.
812 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
815 JH/02 The smtp transport option "multi_domain" is now expanded.
817 JH/03 The smtp transport now requests PRDR by default, if the server offers
820 JH/04 Certificate name checking on server certificates, when exim is a client,
821 is now done by default. The transport option tls_verify_cert_hostnames
822 can be used to disable this per-host. The build option
823 EXPERIMENTAL_CERTNAMES is withdrawn.
825 JH/05 The value of the tls_verify_certificates smtp transport and main options
826 default to the word "system" to access the system default CA bundle.
827 For GnuTLS, only version 3.0.20 or later.
829 JH/06 Verification of the server certificate for a TLS connection is now tried
830 (but not required) by default. The verification status is now logged by
831 default, for both outbound TLS and client-certificate supplying inbound
834 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
837 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
838 Status Notification (bounce) messages are now MIME format per RFC 3464.
839 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
840 under the control of the dsn_advertise_hosts option, and routers may
841 have a dsn_lasthop option.
843 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
844 default, modifiable by a malware= option. The list separator for
845 the options can now be changed in the usual way. Bug 68.
847 JH/10 The smtp_receive_timeout main option is now expanded before use.
849 JH/11 The incoming_interface log option now also enables logging of the
850 local interface on delivery outgoing connections.
852 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
853 if the interface and destination host and port all match.
855 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
858 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
859 Patch from Andrew Lewis.
861 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
862 now supports optional time-restrictions, weighting, and priority
863 modifiers per server. Patch originally by <rommer@active.by>.
865 JH/16 The spamd_address main option now supports a mixed list of local
866 and remote servers. Remote servers can be IPv6 addresses, and
867 specify a port-range.
869 JH/17 Bug 68: The spamd_address main option now supports an optional
870 timeout value per server.
872 JH/18 Bug 1581: Router and transport options headers_add/remove can
873 now have the list separator specified.
875 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
878 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
881 JH/21 Support for the A6 type of dns record is withdrawn.
883 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
884 rather than the verbs used.
886 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
887 from 255 to 1024 chars.
889 JH/24 Verification callouts now attempt to use TLS by default.
891 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
892 are generic router options now. The defaults didn't change.
894 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
895 Original patch from Alexander Shikoff, worked over by JH.
897 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
898 files if ZCAT_COMMAND is not executable.
900 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
902 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
904 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
905 Normally benign, it bites when the pair was led to by a CNAME;
906 modern usage is to not canonicalize the domain to a CNAME target
907 (and we were inconsistent anyway for A-only vs AAAA+A).
909 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
911 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
912 when evaluating $sender_host_dnssec.
914 JH/31 Check the HELO verification lookup for DNSSEC, adding new
915 $sender_helo_dnssec variable.
917 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
919 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
921 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
923 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
924 documented as working, but never had. Support all but $spam_report.
926 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
927 added for tls authenticator.
929 HS/03 Add perl_taintmode main config option
935 TL/01 When running the test suite, the README says that variables such as
936 no_msglog_check are global and can be placed anywhere in a specific
937 test's script, however it was observed that placement needed to be near
938 the beginning for it to behave that way. Changed the runtest perl
939 script to read through the entire script once to detect and set these
940 variables, reset to the beginning of the script, and then run through
941 the script parsing/test process like normal.
943 TL/02 The BSD's have an arc4random API. One of the functions to induce
944 adding randomness was arc4random_stir(), but it has been removed in
945 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
946 function when detected.
948 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
949 cause callback expansion.
951 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
952 syntax errors in an expansion can be treated as a string instead of
953 logging or causing an error, due to the internal use of bool_lax
954 instead of bool when processing it.
956 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
957 server certificates when making smtp deliveries.
959 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
961 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
963 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
965 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
966 Merged patch from Sebastian Wiedenroth.
968 JH/05 Fix results-pipe from transport process. Several recipients, combined
969 with certificate use, exposed issues where response data items split
970 over buffer boundaries were not parsed properly. This eventually
971 resulted in duplicates being sent. This issue only became common enough
972 to notice due to the introduction of connection certificate information,
973 the item size being so much larger. Found and fixed by Wolfgang Breyha.
975 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
976 size buffer was used, resulting in syntax errors when an expansion
979 JH/07 Add support for directories of certificates when compiled with a GnuTLS
980 version 3.3.6 or later.
982 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
983 is EXPERIMENTAL_EVENT, the main-configuration and transport options
984 both become "event_action", the variables become $event_name, $event_data
985 and $event_defer_errno. There is a new variable $verify_mode, usable in
986 routers, transports and related events. The tls:cert event is now also
987 raised for inbound connections, if the main configuration event_action
990 TL/06 In test suite, disable OCSP for old versions of openssl which contained
991 early OCSP support, but no stapling (appears to be less than 1.0.0).
993 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
994 server certificate names available under the smtp transport option
995 "tls_verify_cert_hostname" now do not permit multi-component wildcard
998 JH/10 Time-related extraction expansions from certificates now use the main
999 option "timezone" setting for output formatting, and are consistent
1000 between OpenSSL and GnuTLS compilations. Bug 1541.
1002 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1003 encoded parameter in the incoming message. Bug 1558.
1005 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1006 include certificate info, eximon was claiming there were spoolfile
1009 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1011 JH/14 Log delivery-related information more consistently, using the sequence
1012 "H=<name> [<ip>]" wherever possible.
1014 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1015 are problematic for Debian distribution, omit them from the release
1018 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1020 JH/16 Fix string representation of time values on 64bit time_t architectures.
1023 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1024 output list separator was used.
1029 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1030 checkers that were complaining about end of non-void function with no
1033 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1034 This was a regression introduced in 4.83 by another bugfix.
1036 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1038 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1039 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1045 TF/01 Correctly close the server side of TLS when forking for delivery.
1047 When a message was received over SMTP with TLS, Exim failed to clear up
1048 the incoming connection properly after forking off the child process to
1049 deliver the message. In some situations the subsequent outgoing
1050 delivery connection happened to have the same fd number as the incoming
1051 connection previously had. Exim would try to use TLS and fail, logging
1052 a "Bad file descriptor" error.
1054 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1055 utilities have not been installed.
1057 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1058 temporary space as the ACL may create new global variables.
1060 TL/01 LDAP support uses per connection or global context settings, depending
1061 upon the detected version of the libraries at build time.
1063 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1064 to extract and use the src ip:port in logging and expansions as if it
1065 were a direct connection from the outside internet. PPv2 support was
1066 updated based on HAProxy spec change in May 2014.
1068 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1070 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1071 Properly escape header and check for NULL return.
1073 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1076 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1078 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1079 characters in header names, implemented as a verify condition.
1080 Contributed by Michael Fischer v. Mollard.
1082 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1083 results permerror and temperror. Previous values are deprecated but
1084 still accepted. In a future release, err_perm and err_temp will be
1085 completely removed, which will be a backward incompatibility if the
1086 ACL tests for either of these two old results. Patch contributed by
1087 user bes-internal on the mailing list.
1089 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1091 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1092 selectors, in both main and reject logs.
1094 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1097 JH/07 Add malware type "sock" for talking to simple daemon.
1099 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1101 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1102 routers/transports under cutthrough routing.
1104 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1105 numbers. Touch up "bool" conditional to keep the same definition.
1107 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1109 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1111 JH/12 Expand items in router/transport headers_add or headers_remove lists
1112 individually rather than the list as a whole. Bug 1452.
1114 Required for reasonable handling of multiple headers_ options when
1115 they may be empty; requires that headers_remove items with embedded
1116 colons must have them doubled (or the list-separator changed).
1118 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1119 view the policy declared in the DMARC record. Currently, $dmarc_status
1120 is a combined value of both the record presence and the result of the
1123 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1125 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1126 dnslookup router and the smtp transport (applying to the forward
1129 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1130 of ldap servers used for a specific lookup. Patch provided by Heiko
1133 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1134 New variable $lookup_dnssec_authenticated for observability.
1136 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1137 Patch submitted by Lars Timman.
1139 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1141 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1142 Requires trusted mode and valid format message id, aborts otherwise.
1143 Patch contributed by Heiko Schlichting.
1145 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1146 certextract with support for various fields. Bug 1358.
1148 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1149 is requested by default, modifiable by smtp transport option
1152 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1153 operate on certificate variables to give certificate fingerprints
1154 Also new ${sha256:cert_variable}.
1156 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1158 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1159 Christian Aistleitner.
1161 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1163 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1164 file. Patch from Wolfgang Breyha.
1166 JH/25 Expand the coverage of the delivery $host and $host_address to
1167 client authenticators run in verify callout. Bug 1476.
1169 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1170 align with daemon_smtp_ports. Bug 72.
1172 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1173 support and error reporting did not work properly.
1175 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1176 and is readable. Patch from Andrew Colin Kissa.
1178 TL/14 Enhance documentation of ${run expansion and how it parses the
1179 commandline after expansion, particularly in the case when an
1180 unquoted variable expansion results in an empty value.
1182 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1184 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1185 by renaming away from it. Addresses GH issue 15, reported by
1188 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1190 TL/15 SECURITY: prevent double expansion in math comparison functions
1191 (can expand unsanitized data). Not remotely exploitable.
1198 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1200 PP/02 Make -n do something, by making it not do something.
1201 When combined with -bP, the name of an option is not output.
1203 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1206 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1207 $sender_host_name and config options to manage this, and basic check
1210 PP/05 DSCP support for outbound connections and control modifier for inbound.
1212 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1213 (Only plugin which currently uses this is kerberos4, which nobody should
1214 be using, but we should make it available and other future plugins might
1215 conceivably use it, even though it would break NAT; stuff *should* be
1216 using channel bindings instead).
1218 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1219 name; added for Sendmail compatibility; requires admin caller.
1220 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1221 just ignore it); requires trusted caller.
1222 Also parse but ignore: -Ac -Am -X<logfile>
1225 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1227 TL/02 Add +smtp_confirmation as a default logging option.
1229 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1230 Patch by Magnus Holmgren from 2007-02-20.
1232 TL/04 Bugzilla 1281 - Spec typo.
1233 Bugzilla 1283 - Spec typo.
1234 Bugzilla 1290 - Spec grammar fixes.
1236 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1238 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1240 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1241 mailing list by by Dmitry Isaikin.
1243 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1245 JH/02 Support "G" suffix to numbers in ${if comparisons.
1247 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1249 NM/01 Bugzilla 1197 - Spec typo
1250 Bugzilla 1196 - Spec examples corrections
1252 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1254 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1255 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1258 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1259 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1261 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1262 "acl {{name}{arg}...}", and optional args on acl condition
1265 JH/05 Permit multiple router/transport headers_add/remove lines.
1267 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1269 JH/07 Avoid using a waiting database for a single-message-only transport.
1270 Performance patch from Paul Fisher. Bugzilla 1262.
1272 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1275 JH/09 Add $headers_added variable, with content from use of ACL modifier
1276 add_header (but not yet added to the message). Bugzilla 199.
1278 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1279 Pulled from Bugzilla 817 by Wolfgang Breyha.
1281 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1283 (nb: this is the same fix as in Exim 4.80.1)
1285 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1288 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1291 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1293 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1294 advertise SMTP AUTH mechanism to us, instead of a generic
1295 protocol violation error. Also, make Exim more robust to bad
1296 data from the Dovecot auth socket.
1298 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1300 When a queue runner is handling a message, Exim first routes the
1301 recipient addresses, during which it prunes them based on the retry
1302 hints database. After that it attempts to deliver the message to
1303 any remaining recipients. It then updates the hints database using
1306 So if a recipient address works intermittently, it can get repeatedly
1307 deferred at routing time. The retry hints record remains fresh so the
1308 address never reaches the final cutoff time.
1310 This is a fairly common occurrence when a user is bumping up against
1311 their storage quota. Exim had some logic in its local delivery code
1312 to deal with this. However it did not apply to per-recipient defers
1313 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1315 This change adds a proper retry rule check during routing so that the
1316 final cutoff time is checked against the message's age. We only do
1317 this check if there is an address retry record and there is not a
1318 domain retry record; this implies that previous attempts to handle
1319 the address had the retry_use_local_parts option turned on. We use
1320 this as an approximation for the destination being like a local
1321 delivery, as in LMTP.
1323 I suspect this new check makes the old local delivery cutoff check
1324 redundant, but I have not verified this so I left the code in place.
1326 TF/02 Correct gecos expansion when From: is a prefix of the username.
1328 Test 0254 submits a message to Exim with the header
1332 When I ran the test suite under the user fanf2, Exim expanded
1333 the header to contain my full name, whereas it should have added
1334 a Resent-Sender: header. It erroneously treats any prefix of the
1335 username as equal to the username.
1337 This change corrects that bug.
1339 GF/01 DCC debug and logging tidyup
1340 Error conditions log to paniclog rather than rejectlog.
1341 Debug lines prefixed by "DCC: " to remove any ambiguity.
1343 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1345 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1346 Bug spotted by Jeremy Harris; was flawed since initial commit.
1347 Would have resulted in OCSP responses post-SNI triggering an Exim
1348 NULL dereference and crash.
1350 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1352 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1353 Bug detection, analysis and fix by Samuel Thibault.
1354 Bugzilla 1331, Debian bug #698092.
1356 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1358 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1359 Server implementation by Todd Lyons, client by JH.
1360 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1361 config variable "prdr_enable" controls whether the server
1362 advertises the facility. If the client requests PRDR a new
1363 acl_data_smtp_prdr ACL is called once for each recipient, after
1364 the body content is received and before the acl_smtp_data ACL.
1365 The client is controlled by both of: a hosts_try_prdr option
1366 on the smtp transport, and the server advertisement.
1367 Default client logging of deliveries and rejections involving
1368 PRDR are flagged with the string "PRDR".
1370 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1371 fclose(). Diagnosis by Todd Lyons.
1373 PP/17 Update configure.default to handle IPv6 localhost better.
1374 Patch by Alain Williams (plus minor tweaks).
1377 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1378 This is now consistent with GnuTLS, and is now documented: the
1379 previous undocumented portable approach to treating the option as
1380 unset was to force an expansion failure. That still works, and
1381 an empty string is now equivalent.
1383 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1384 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1385 not performing validation itself.
1387 PP/20 Added force_command boolean option to pipe transport.
1388 Patch from Nick Koston, of cPanel Inc.
1390 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1393 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1395 PP/21 Fix eximon continuous updating with timestamped log-files.
1396 Broken in a format-string cleanup in 4.80, missed when I repaired the
1397 other false fix of the same issue.
1398 Report and fix from Heiko Schlichting.
1401 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1402 Report from Prashanth Katuri.
1404 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1405 It's SecureTransport, so affects any MacOS clients which use the
1406 system-integrated TLS libraries, including email clients.
1408 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1409 using a MIME ACL for non-SMTP local injection.
1410 Report and assistance in diagnosis by Warren Baker.
1412 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1414 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1416 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1417 last id that failed so it may be referenced in subsequent ACL's.
1419 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1422 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1423 ldap library initialization, allowing self-signed CA's to be
1424 used. Also properly sets require_cert option later in code by
1425 using NULL (global ldap config) instead of ldap handle (per
1426 session). Bug diagnosis and testing by alxgomz.
1428 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1429 the src/util/ subdirectory.
1431 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1432 renamed to Transport Post Delivery Action by Jeremy Harris, as
1435 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1436 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1437 redis_servers = needs to be configured which will be used by the redis
1438 lookup. Patch from Warren Baker, of The Packet Hub.
1440 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1442 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1443 hostname or reverse DNS when processing a host list. Used suggestions
1444 from multiple comments on this bug.
1446 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1448 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1449 Missed a few lines, added it to make the runtest require no keyboard
1452 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1453 contains upper case chars. Make router use caseful_local_part.
1455 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1456 support when GnuTLS has been built with p11-kit.
1462 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1464 This, or similar/improved, will also be change PP/11 of 4.82.
1470 PP/01 Handle short writes when writing local log-files.
1471 In practice, only affects FreeBSD (8 onwards).
1472 Bugzilla 1053, with thanks to Dmitry Isaikin.
1474 NM/01 Bugzilla 949 - Documentation tweak
1476 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1479 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1481 PP/02 Implemented gsasl authenticator.
1483 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1485 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1486 `pkg-config foo` for cflags/libs.
1488 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1489 with rest of GSASL and with heimdal_gssapi.
1491 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1492 `pkg-config foo` for cflags/libs for the TLS implementation.
1494 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1495 properties get this fed in as external SSF. A number of robustness
1496 and debugging improvements to the cyrus_sasl authenticator.
1498 PP/08 cyrus_sasl server now expands the server_realm option.
1500 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1501 Patch by Jeremy Harris.
1503 PP/10 Added dbmjz lookup type.
1505 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1507 PP/12 MAIL args handles TAB as well as SP, for better interop with
1508 non-compliant senders.
1509 Analysis and variant patch by Todd Lyons.
1511 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1512 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1513 Patch from Dirk Mueller <dmueller@suse.com>
1515 PP/13 tls_peerdn now print-escaped for spool files.
1516 Observed some $tls_peerdn in wild which contained \n, which resulted
1517 in spool file corruption.
1519 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1520 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1521 or write after TLS renegotiation, which otherwise led to messages
1524 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1525 as a tracking header (ie: a signed header comes before the signature).
1526 Patch from Wolfgang Breyha.
1528 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1529 comma-sep list; embedded commas doubled.
1531 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1533 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1535 Report and patch from Dmitry Banschikov.
1537 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1538 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1539 failures. We appear to now support TLS1.1+ with Exim.
1541 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1542 lets Exim select keys and certificates based upon TLS SNI from client.
1543 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1544 before an outbound SMTP session. New log_selector, +tls_sni.
1546 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1547 NULL dereference. Report and patch from Alun Jones.
1549 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1550 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1551 Not seeing resolver debug output on NetBSD, but suspect this is a
1552 resolver implementation change.
1554 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1555 Left warnings. Added "eximon gdb" invocation mode.
1557 PP/21 Defaulting "accept_8bitmime" to true, not false.
1559 PP/22 Added -bw for inetd wait mode support.
1561 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1562 locate the relevant includes and libraries. Made this the default.
1564 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1565 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1567 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1568 This may cause build issues on older platforms.
1570 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1571 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1572 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1573 Added SNI support via GnuTLS too.
1574 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1576 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1578 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1579 Applied second patch from Janne, implementing suggestion to default
1580 multiple-strings-in-record handling to match SPF spec.
1582 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1584 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1585 read-only, out of scope).
1586 Patch from Wolfgang Breyha, report from Stuart Northfield.
1588 PP/29 Fix three issues highlighted by clang analyser static analysis.
1589 Only crash-plausible issue would require the Cambridge-specific
1590 iplookup router and a misconfiguration.
1591 Report from Marcin Mirosław.
1593 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1595 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1596 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1597 As part of this, removing so much warning spew let me fix some minor
1598 real issues in debug logging.
1600 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1601 assignment on my part. Fixed.
1603 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1604 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1605 Janne Snabb (who went above and beyond: thank you).
1607 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1608 string otherwise requires a connection and a bunch more work and it's
1609 relatively easy to get wrong. Should also expose TLS library linkage
1612 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1613 64-bit ${eval} (JH/03).
1615 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1616 GNU libc to support some of the 64-bit stuff, should not lead to
1617 conflicts. Defined before os.h is pulled in, so if a given platform
1618 needs to override this, it can.
1620 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1621 protection layer was required, which is not implemented.
1622 Bugzilla 1254, patch from Wolfgang Breyha.
1624 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1625 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1626 tls_dhparam take prime identifiers. Also unbreak combination of
1627 OpenSSL+DH_params+TLSSNI.
1629 PP/39 Disable SSLv2 by default in OpenSSL support.
1635 PP/01 Solaris build fix for Oracle's LDAP libraries.
1636 Bugzilla 1109, patch from Stephen Usher.
1638 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1640 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1643 TF/02 Fix a couple more cases where we did not log the error message
1644 when unlink() failed. See also change 4.74-TF/03.
1646 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1647 lock up or crash if it happened to be inside a call to libc when it
1648 got a SIGUSR1 from exiwhat.
1650 The SIGUSR1 handler appends the current process status to the process
1651 log which is later printed by exiwhat. It used to use the general
1652 purpose logging code to do this, but several functions it calls are
1653 not safe for signals.
1655 The new output code in the SIGUSR1 handler is specific to the process
1656 log, and simple enough that it's easy to inspect for signal safety.
1657 Removing some special cases also simplifies the general logging code.
1658 Removing the spurious timestamps from the process log simplifies
1661 TF/04 Improved ratelimit ACL condition.
1663 The /noupdate option has been deprecated in favour of /readonly which
1664 has clearer semantics. The /leaky, /strict, and /readonly update modes
1665 are mutually exclusive. The update mode is no longer included in the
1666 database key; it just determines when the database is updated. (This
1667 means that when you upgrade Exim will forget old rate measurements.)
1669 Exim now checks that the per_* options are used with an update mode that
1670 makes sense for the current ACL. For example, when Exim is processing a
1671 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1672 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1673 must specify per_mail/readonly. If you omit the update mode it defaults to
1674 /leaky where that makes sense (as before) or /readonly where required.
1676 The /noupdate option is now undocumented but still supported for
1677 backwards compatibility. It is equivalent to /readonly except that in
1678 ACLs where /readonly is required you may specify /leaky/noupdate or
1679 /strict/noupdate which are treated the same as /readonly.
1681 A useful new feature is the /count= option. This is a generalization
1682 of the per_byte option, so that you can measure the throughput of other
1683 aggregate values. For example, the per_byte option is now equivalent
1684 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1686 The per_rcpt option has been generalized using the /count= mechanism
1687 (though it's more complicated than the per_byte equivalence). When it is
1688 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1689 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1690 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1691 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1692 Note that using per_rcpt with a non-readonly update mode in more than
1693 one ACL will cause the recipients to be double-counted. (The per_mail
1694 and per_byte options don't have this problem.)
1696 The handling of very low rates has changed slightly. If the computed rate
1697 is less than the event's count (usually one) then this event is the first
1698 after a long gap. In this case the rate is set to the same as this event's
1699 count, so that the first message of a spam run is counted properly.
1701 The major new feature is a mechanism for counting the rate of unique
1702 events. The new per_addr option counts the number of different
1703 recipients that someone has sent messages to in the last time period. It
1704 behaves like per_rcpt if all the recipient addresses are different, but
1705 duplicate recipient addresses do not increase the measured rate. Like
1706 the /count= option this is a general mechanism, so the per_addr option
1707 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1708 example, measure the rate that a client uses different sender addresses
1709 with the options per_mail/unique=$sender_address. There are further
1710 details in the main documentation.
1712 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1714 TF/06 Removed a few PCRE remnants.
1716 TF/07 Automatically extract Exim's version number from tags in the git
1717 repository when doing development or release builds.
1719 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1720 Bugzilla 879. Patch from Paul Fisher.
1722 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1723 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1726 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1728 PP/05 Variable $av_failed, true if the AV scanner deferred.
1729 Bugzilla 1078. Patch from John Horne.
1731 PP/06 Stop make process more reliably on build failure.
1732 Bugzilla 1087. Patch from Heiko Schlittermann.
1734 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1735 Bugzilla 1089. Patch from Heiko Schlittermann.
1737 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1738 Bugzilla 1131. Patch from Holger Weiß.
1740 PP/09 Handle IPv6 addresses with SPF.
1741 Bugzilla 860. Patch from Wolfgang Breyha.
1743 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1745 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1748 PP/11 match_* no longer expand right-hand-side by default.
1749 New compile-time build option, EXPAND_LISTMATCH_RHS.
1750 New expansion conditions, "inlist", "inlisti".
1752 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1754 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1756 PP/14 fix log_write() format string regression from TF/03.
1757 Bugzilla 1152. Patch from Dmitry Isaikin.
1763 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1765 PP/02 Harmonised TLS library version reporting; only show if debugging.
1766 Layout now matches that introduced for other libraries in 4.74 PP/03.
1768 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1770 PP/04 New "dns_use_edns0" global option.
1772 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1775 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1776 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1778 TK/01 Updated PolarSSL code to 0.14.2.
1779 Bugzilla 1097. Patch from Andreas Metzler.
1781 PP/07 Catch divide-by-zero in ${eval:...}.
1782 Fixes bugzilla 1102.
1784 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1787 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1788 format-string attack -- SECURITY: remote arbitrary code execution.
1790 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1791 time unintentionally subject to list matching rules, letting the header
1792 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1793 arbitrary string expansion). This allowed for information disclosure.
1795 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1796 INT_MIN/-1 -- value coerced to INT_MAX.
1802 NM/01 Workaround for PCRE version dependency in version reporting
1805 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1806 This fixes portability to compilers other than gcc, notably
1807 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1809 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1810 makefiles for portability to HP-UX and POSIX correctness.
1812 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1813 Also via indented variable definition in the Makefile.
1814 (Debugging by Oliver Heesakkers).
1816 PP/02 Restore caching of spamd results with expanded spamd_address.
1817 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1819 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1820 Improves build reliability. Fix from: Frank Elsner
1822 NM/02 Fix wide character breakage in the rfc2047 coding
1823 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1825 NM/03 Allow underscore in dnslist lookups
1826 Fixes bug 1026. Patch from Graeme Fowler
1828 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1829 Code patches from Adam Ciarcinski of NetBSD.
1831 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1834 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1835 is logged, to avoid truncation. Patch from John Horne.
1837 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1838 Patch from Jakob Hirsch.
1840 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1841 SQL string expansion failure details.
1842 Patch from Andrey Oktyabrski.
1844 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1845 Patch from Simon Arlott.
1847 PP/09 New lookups functionality failed to compile on old gcc which rejects
1848 extern declarations in function scope.
1849 Patch from Oliver Fleischmann
1851 PP/10 Use sig_atomic_t for flags set from signal handlers.
1852 Check getgroups() return and improve debugging.
1853 Fixed developed for diagnosis in bug 927 (which turned out to be
1856 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1857 Patch from Mark Zealey.
1859 PP/12 Bugzilla 1056: Improved spamd server selection.
1860 Patch from Mark Zealey.
1862 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1863 Based on patch from Heiko Schlittermann.
1865 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1866 Patch from Uwe Doering, sign-off by Michael Haardt.
1868 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1869 variable declaration deep within a block. Bug and patch from
1872 PP/15 lookups-Makefile IRIX compatibility coercion.
1874 PP/16 Make DISABLE_DKIM build knob functional.
1876 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1877 Patch by Simon Arlott
1879 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1880 variable argument macros. Our copy now differs from upstream.
1886 TF/01 Failure to get a lock on a hints database can have serious
1887 consequences so log it to the panic log.
1889 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1890 controlled using the smtp_confirmation log selector.
1892 TF/03 Include the error message when we fail to unlink a spool file.
1894 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1895 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1896 for maintaining out-of-tree patches for some time.
1898 PP/01 Bugzilla 139: Documentation and portability issues.
1899 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1900 Handle per-OS dynamic-module compilation flags.
1902 PP/02 Let /dev/null have normal permissions.
1903 The 4.73 fixes were a little too stringent and complained about the
1904 permissions on /dev/null. Exempt it from some checks.
1905 Reported by Andreas M. Kirchwitz.
1907 PP/03 Report version information for many libraries, including
1908 Exim version information for dynamically loaded libraries. Created
1909 version.h, now support a version extension string for distributors
1910 who patch heavily. Dynamic module ABI change.
1912 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1913 privilege escalation vulnerability whereby the Exim run-time user
1914 can cause root to append content of the attacker's choosing to
1917 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1920 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1921 If dropping privileges for untrusted macros, we disabled normal logging
1922 on the basis that it would fail; for the Exim run-time user, this is not
1923 the case, and it resulted in successful deliveries going unlogged.
1924 Fixed. Reported by Andreas Metzler.
1930 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1931 only prepend for the Resent-* case. Fixes regression introduced in
1932 Exim 4.70 by NM/22 for Bugzilla 607.
1934 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1935 increasing numbers of administrators be bitten by this.
1937 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1939 PP/03 Bugzilla 994: added openssl_options main configuration option.
1941 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1943 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1945 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1947 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1948 without a peer certificate, leading to a segfault because of an
1949 assumption that peers always have certificates. Be a little more
1950 paranoid. Problem reported by Martin Tscholak.
1952 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1953 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1954 NB: ClamAV planning to remove STREAM in "middle of 2010".
1955 CL also introduces -bmalware, various -d+acl logging additions and
1956 more caution in buffer sizes.
1958 PP/09 Implemented reverse_ip expansion operator.
1960 PP/10 Bugzilla 937: provide a "debug" ACL control.
1962 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1964 PP/12 Bugzilla 973: Implement --version.
1966 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1968 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1970 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1972 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1973 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1974 ignore trailing whitespace.
1976 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1978 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1981 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1982 Notification from Dr Andrew Aitchison.
1984 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1985 ExtendedDetectionInfo response format.
1986 Notification from John Horne.
1988 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1991 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1992 XSL and documented dependency on system catalogs, with examples of how
1995 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1998 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1999 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2000 configuration file which is writeable by the Exim user or group.
2002 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2003 of configuration files to cover files specified with the -C option if
2004 they are going to be used with root privileges, not just the default
2007 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2008 option (effectively making it always true).
2010 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2011 files to be used while preserving root privileges.
2013 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2014 that rogue child processes cannot use them.
2016 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2017 run-time user, instead of root.
2019 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2020 Exim run-time user without dropping privileges.
2022 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2023 result string, instead of calling string_vformat() twice with the same
2026 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2027 for other users. Others should always drop root privileges if they use
2028 -C on the command line, even for a whitelisted configure file.
2030 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2032 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2038 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2039 $data_path, and $header_path variables; fixed documentation bugs and
2042 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2043 exipick to access non-standard spools, including the "frozen" queue
2046 NM/01 Bugzilla 965: Support mysql stored procedures.
2047 Patch from Alain Williams
2049 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2051 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2052 Patch from Andreas Metzler
2054 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2055 Patch from Kirill Miazine
2057 NM/05 Bugzilla 671: Added umask to procmail example.
2059 JJ/03 installed exipick 20100323.0, fixing doc bug
2061 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2062 directory. Notification and patch from Dan Rosenberg.
2064 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2066 TK/02 Improve log output when DKIM signing operation fails.
2068 MH/01 Treat the transport option dkim_domain as a colon separated
2069 list, not as a single string, and sign the message with each element,
2070 omitting multiple occurences of the same signer.
2072 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2073 Bugzilla 985, 986. Patch by Simon Arlott
2075 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2076 Patch by Simon Arlott
2078 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2079 MBX locking. Notification from Dan Rosenberg.
2085 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2087 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2089 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2091 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2093 NM/04 Bugzilla 915: Flag broken perl installation during build.
2099 TK/01 Added patch by Johannes Berg that expands the main option
2100 "spamd_address" if it starts with a dollar sign.
2102 TK/02 Write list of recipients to X-Envelope-Sender header when building
2103 the mbox-format spool file for content scanning (suggested by Jakob
2106 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2107 (http://www.dcc-servers.net/) support via dccifd. Activated by
2108 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2110 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2111 by Mark Daniel Reidel <mr@df.eu>.
2113 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2114 When building exim an external PCRE library is now needed -
2115 PCRE is a system library on the majority of modern systems.
2116 See entry on PCRE_LIBS in EDITME file.
2118 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2119 conversation. Added nologin parameter to request.
2120 Patch contributed by Kirill Miazine.
2122 TF/01 Do not log submission mode rewrites if they do not change the address.
2124 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2126 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2127 log files in place. Contributed by Roberto Lima.
2129 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2131 TF/03 Bugzilla 615: When checking the local_parts router precondition
2132 after a local_part_suffix or local_part_prefix option, Exim now
2133 does not use the address's named list lookup cache, since this
2134 contains cached lookups for the whole local part.
2136 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2137 Robert Millan. Documentation is in experimental-spec.txt.
2139 TF/04 Bugzilla 668: Fix parallel build (make -j).
2141 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2143 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2144 Patch provided by Jan Srzednicki.
2146 TF/05 Leading white space used to be stripped from $spam_report which
2147 wrecked the formatting. Now it is preserved.
2149 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2150 that they are available at delivery time.
2152 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2154 TF/08 TLS error reporting now respects the incoming_interface and
2155 incoming_port log selectors.
2157 TF/09 Produce a more useful error message if an SMTP transport's hosts
2158 setting expands to an empty string.
2160 NM/06 Bugzilla 744: EXPN did not work under TLS.
2161 Patch provided by Phil Pennock.
2163 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2164 Patch provided by Richard Godbee.
2166 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2167 acl_smtp_notquit, added index entry.
2169 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2170 Patch provided by Eugene Bujak.
2172 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2173 accept(). Patch provided by Maxim Dounin.
2175 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2176 Patch provided by Phil Pennock.
2178 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2180 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2181 Patch provided by Brad "anomie" Jorsch.
2183 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2184 Patch provided by Dean Brooks.
2186 TK/05 Add native DKIM support (does not depend on external libraries).
2188 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2189 Patch provided by Graeme Fowler.
2191 NM/16 Bugzilla 851: Documentation example syntax fix.
2193 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2195 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2198 NM/19 Bugzilla 745: TLS version reporting.
2199 Patch provided by Phil Pennock.
2201 NM/20 Bugzilla 167: bool: condition support.
2202 Patch provided by Phil Pennock.
2204 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2205 clients. Patch provided by Phil Pennock.
2207 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2208 Patch provided by Brad "anomie" Jorsch.
2210 NM/23 Bugzilla 687: Fix misparses in eximstats.
2211 Patch provided by Heiko Schlittermann.
2213 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2214 Patch provided by Heiko Schlittermann.
2216 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2217 plus update to original patch.
2219 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2221 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2222 Patch provided by David Brownlee.
2224 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2226 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2228 NM/30 Bugzilla 888: TLS documentation bugfixes.
2230 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2232 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2233 Unlike the original bugzilla I have changed all shell scripts in src tree.
2235 NM/33 Bugzilla 898: Transport filter timeout fix.
2236 Patch by Todd Rinaldo.
2238 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2239 Patch by Serge Demonchaux.
2241 NM/35 Bugzilla 39: Base64 decode bug fixes.
2242 Patch by Jakob Hirsch.
2244 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2246 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2248 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2250 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2256 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2257 ALT-N's libdkim that I have put here:
2258 http://duncanthrax.net/exim-experimental/
2260 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2261 were called 'true' and it seems that C99 defines that as a reserved
2262 keyword to be used with 'bool' variable types. That means you could
2263 not include C99-style headers which use bools without triggering
2264 build errors in sieve.c.
2266 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2267 as mailq or other aliases. Changed the --help handling significantly
2268 to do whats expected. exim_usage() emits usage/help information.
2270 SC/01 Added the -bylocaldomain option to eximstats.
2272 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2274 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2276 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2282 PH/01 Another patch from the Sieve maintainer.
2284 PH/02 When an IPv6 address is converted to a string for single-key lookup
2285 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2286 dots are used instead of colons so that keys in lsearch files need not
2287 contain colons. This was done some time before quoting was made available
2288 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2289 (notated using the quote facility) so as to distinguish them from IPv4
2290 keys. This meant that lookups for IP addresses in host lists did not work
2291 for iplsearch lookups.
2293 This has been fixed by arranging for IPv6 addresses to be expressed with
2294 colons if the lookup type is iplsearch. This is not incompatible, because
2295 previously such lookups could never work.
2297 The situation is now rather anomalous, since one *can* have colons in
2298 ordinary lsearch keys. However, making the change in all cases is
2299 incompatible and would probably break a number of configurations.
2301 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2304 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2305 conversion specification without a maximum field width, thereby enabling
2306 a rogue spamd server to cause a buffer overflow. While nobody in their
2307 right mind would setup Exim to query an untrusted spamd server, an
2308 attacker that gains access to a server running spamd could potentially
2309 exploit this vulnerability to run arbitrary code as the Exim user.
2311 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2312 $primary_hostname instead of what libspf2 thinks the hosts name is.
2314 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2315 a directory entry by the name of the lookup key. Previously, if a
2316 symlink pointed to a non-existing file or a file in a directory that
2317 Exim lacked permissions to read, a lookup for a key matching that
2318 symlink would fail. Now it is enough that a matching directory entry
2319 exists, symlink or not. (Bugzilla 503.)
2321 PH/03 The body_linecount and body_zerocount variables are now exported in the
2324 PH/04 Added the $dnslist_matched variable.
2326 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2327 This means they are set thereafter only if the connection becomes
2330 PH/06 Added the client_condition to authenticators so that some can be skipped
2331 by clients under certain conditions.
2333 PH/07 The error message for a badly-placed control=no_multiline_responses left
2334 "_responses" off the end of the name.
2336 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2338 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2339 (without spaces) instead of just copying the configuration text.
2341 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2343 PH/11 Added $max_received_linelength.
2345 PH/12 Added +ignore_defer and +include_defer to host lists.
2347 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2348 way in which PCRE > 7.0 is built.
2350 PH/14 Implemented queue_only_load_latch.
2352 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2353 MAIL command. The effect was to mangle the value on 64-bit systems.
2355 PH/16 Another patch from the Sieve maintainer.
2357 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2359 PH/18 If a system quota error occurred while trying to create the file for
2360 a maildir delivery, the message "Mailbox is full" was not appended to the
2361 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2362 only to a quota excession during the actual writing of the file.
2364 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2365 characters?) which causes problems in log lines. The DN values are now
2366 passed through string_printing() before being added to log lines.
2368 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2369 and InterBase are left for another time.)
2371 PH/21 Added message_body_newlines option.
2373 PH/22 Guard against possible overflow in moan_check_errorcopy().
2375 PH/23 POSIX allows open() to be a macro; guard against that.
2377 PH/24 If the recipient of an error message contained an @ in the local part
2378 (suitably quoted, of course), incorrect values were put in $domain and
2379 $local_part during the evaluation of errors_copy.
2385 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2386 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2389 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2390 issue a MAIL command.
2392 PH/02 In an ACL statement such as
2394 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2396 if a client was not listed at all, or was listed with a value other than
2397 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2398 the condition was not true (as it should be), so access was not denied.
2399 The bug was that the ! inversion was incorrectly passed on to the second
2400 item. This has been fixed.
2402 PH/03 Added additional dnslists conditions == and =& which are different from
2403 = and & when the dns lookup returns more than one IP address.
2405 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2406 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2408 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2409 FSYNC, which compiles an option called disable_fsync that allows for
2410 bypassing fsync(). The documentation is heavily laced with warnings.
2412 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2414 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2415 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2416 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2417 including adding "make clean"; (3) Added -fPIC when compiling the test
2418 dynamically loaded module, to get rid of a warning.
2420 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2421 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2422 = 0s. The bug is otherwise harmless.
2424 PH/07 There was a bug in the dovecot authenticator such that the value of
2425 $auth1 could be overwritten, and so not correctly preserved, after a
2426 successful authentication. This usually meant that the value preserved by
2427 the server_setid option was incorrect.
2429 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2431 PH/09 Installed PCRE release 7.0.
2433 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2434 run for batched SMTP input. It is now run at the start of every message
2435 in the batch. While fixing this I discovered that the process information
2436 (output by running exiwhat) was not always getting set for -bs and -bS
2437 input. This is fixed, and it now also says "batched" for BSMTP.
2439 PH/11 Added control=no_pipelining.
2441 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2442 patch, slightly modified), and move the expansion of helo_data till after
2443 the connection is made in the smtp transport (so it can use these
2446 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2448 PH/14 Added log_selector = +pid.
2450 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2452 PH/16 Add ${if forany and ${if forall.
2454 PH/17 Added dsn_from option to vary the From: line in DSNs.
2456 PH/18 Flush SMTP output before performing a callout, unless control =
2457 no_callout_flush is set.
2459 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2460 was true (the default) a successful delivery failed to delete the retry
2461 item, thus causing premature timeout of the address. The bug is now
2464 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2466 PH/21 Long custom messages for fakedefer and fakereject are now split up
2467 into multiline responses in the same way that messages for "deny" and
2468 other ACL rejections are.
2470 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2471 with slight modification.
2473 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2474 draft, changing the syntax and factoring some duplicate code".
2476 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2477 for deliveries of the second and subsequent messages over the same SMTP
2480 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2481 ${reduce, with only minor "tidies".
2483 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2485 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2486 expansion side effects.
2488 PH/27 When a message times out after an over-quota error from an Exim-imposed
2489 quota, the bounce message says "mailbox is full". This message was not
2490 being given when it was a system quota that was exceeded. It now should
2493 MH/03 Made $recipients available in local_scan(). local_scan() already has
2494 better access to the recipient list through recipients_list[], but
2495 $recipients can be useful in postmaster-provided expansion strings.
2497 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2498 in the case of a MAIL command with additional options following the
2499 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2500 were accidentally chopped off.
2502 PH/29 SMTP synchronization checks are implemented when a command is read -
2503 there is a check that no more input is waiting when there shouldn't be
2504 any. However, for some commands, a delay in an ACL can mean that it is
2505 some time before the response is written. In this time, more input might
2506 arrive, invalidly. So now there are extra checks after an ACL has run for
2507 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2508 pipelining has not been advertised.
2510 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2512 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2513 returned in the SMTP response when smtp_return_error_details was false.
2514 This has been fixed.
2516 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2517 instead of the C I/O that was originally supplied, because problems were
2518 reported on Solaris.
2520 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2521 Exim which did not show up earlier: it was assuming that a call to
2522 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2523 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2524 was a macro that became an assignment, so it seemed to work. This has
2525 changed to a proper function call with a void return, hence the compile
2526 error. Exim's code has been fixed.
2528 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2531 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2533 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2535 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2536 because it turns out to be too big or there is a timeout) there is no
2537 "Completed" line in the log. When some messages of this type were
2538 selected by exigrep, they were listed as "not completed". Others were
2539 picked up by some special patterns. I have improved the selection
2540 criteria to be more general.
2542 PH/38 The host_find_failed option in the manualroute router can now be set
2543 to "ignore", to completely ignore a host whose IP address cannot be
2544 found. If all hosts are ignored, the behaviour is controlled by the new
2545 host_all_ignored option.
2547 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2548 homing or because of multiple MX records with /mx) generated more than
2549 one IP address, and the following item turned out to be the local host,
2550 all the secondary addresses of the first item were incorrectly removed
2551 from the list, along with the local host and any following hosts (which
2552 is what is supposed to happen).
2554 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2555 whoever called Exim into the -H file. In the case of the daemon it was
2556 behaving confusingly. When first started, it used values for whoever
2557 started the daemon, but after a SIGHUP it used the Exim user (because it
2558 calls itself on a restart). I have changed the code so that it now always
2561 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2562 message are rejected with the same error (e.g. no authentication or bad
2563 sender address), and a DATA command is nevertheless sent (as can happen
2564 with PIPELINING or a stupid MUA), the error message that was given to the
2565 RCPT commands is included in the rejection of the DATA command. This is
2566 intended to be helpful for MUAs that show only the final error to their
2569 PH/42 Another patch from the Sieve maintainer.
2571 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2572 Eximstats - Fixed some broken HTML links and added missing column headers
2574 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2575 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2577 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2579 PH/43 Yet another patch from the Sieve maintainer.
2581 PH/44 I found a way to check for a TCP/IP connection going away before sending
2582 the response to the final '.' that terminates a message, but only in the
2583 case where the client has not sent further data following the '.'
2584 (unfortunately, this is allowed). However, in many cases there won't be
2585 any further data because there won't be any more messages to send. A call
2586 to select() can be used: if it shows that the input is "ready", there is
2587 either input waiting, or the socket has been closed. An attempt to read
2588 the next input character can distinguish the two cases. Previously, Exim
2589 would have sent an OK response which the client would never have see.
2590 This could lead to message repetition. This fix should cure that, at
2591 least in a lot of common cases.
2593 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2594 advertised in response to EHLO.
2600 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2601 fixed by 4.65/MH/01 (is this a record?) are fixed:
2603 (i) An empty string was always treated as zero by the numeric comparison
2604 operators. This behaviour has been restored.
2606 (ii) It is documented that the numeric comparison operators always treat
2607 their arguments as decimal numbers. This was broken in that numbers
2608 starting with 0 were being interpreted as octal.
2610 While fixing these problems I realized that there was another issue that
2611 hadn't been noticed. Values of message_size_limit (both the global option
2612 and the transport option) were treated as octal if they started with 0.
2613 The documentation was vague. These values are now always treated as
2614 decimal, and I will make that clear in the documentation.
2620 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2621 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2624 MH/01 Don't check that the operands of numeric comparison operators are
2625 integers when their expansion is in "skipping" mode (fixes bug
2626 introduced by 4.64-PH/07).
2628 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2629 child addresses, Exim now panics and dies. Previously, because the count
2630 is held in a short int, deliveries were likely to be lost. As such a
2631 large number of recipients for a single message is ridiculous
2632 (performance will be very, very poor), I have chosen to impose a limit
2633 rather than extend the field.
2639 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2640 leftover -K file (the existence of which was triggered by #402).
2641 While we were at it, introduced process PID as part of the -K
2642 filename. This should rule out race conditions when creating
2645 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2646 processing considerably. Previous code took too long for large mails,
2647 triggering a timeout which in turn triggers #401.
2649 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2650 in the DK code in transports.c. sendfile() is not really portable,
2651 hence the _LINUX specificness.
2653 TF/01 In the add_headers option to the mail command in an Exim filter,
2654 there was a bug that Exim would claim a syntax error in any
2655 header after the first one which had an odd number of characters
2658 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2659 callout verification, Exim cached a "reject" for the entire domain. This
2660 is correct for most verifications, but it is not correct for a recipient
2661 verification with use_sender or use_postmaster set, because in that case
2662 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2663 case of MAIL FROM:<> rejection from other early rejections (e.g.
2664 rejection of HELO). When verifying a recipient using a non-null MAIL
2665 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2666 Whatever the result of the callout, the value of the domain cache is
2667 left unchanged (for any other kind of callout, getting as far as trying
2668 RCPT means that the domain itself is ok).
2670 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2673 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2674 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2675 session, the daemon ignores ECONNECT errors and logs others; it now
2676 ignores EPIPE as well.
2678 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2679 (quoted-printable decoding).
2681 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2682 later the small subsequent patch to fix an introduced bug.
2684 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2686 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2688 PH/08 An error is now given if message_size_limit is specified negative.
2690 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2691 to be given (somewhat) arbitrary names.
2693 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2696 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2697 miscellaneous code fixes
2699 PH/10 Added the log_reject_target ACL modifier to specify where to log
2702 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2703 hostname. This is wrong, because it relates to the incoming message (and
2704 probably the interface on which it is arriving) and not to the outgoing
2705 callout (which could be using a different interface). This has been
2706 changed to use the value of the helo_data option from the smtp transport
2707 instead - this is what is used when a message is actually being sent. If
2708 there is no remote transport (possible with a router that sets up host
2709 addresses), $smtp_active_hostname is used.
2711 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2712 tweaks were necessary in order to get it to work (see also 21 below):
2713 (a) The code assumed that strncpy() returns a negative number on buffer
2714 overflow, which isn't the case. Replaced with Exim's string_format()
2716 (b) There were several signed/unsigned issues. I just did the minimum
2717 hacking in of casts. There is scope for a larger refactoring.
2718 (c) The code used strcasecmp() which is not a standard C function.
2719 Replaced with Exim's strcmpic() function.
2720 (d) The code set only $1; it now sets $auth1 as well.
2721 (e) A simple test gave the error "authentication client didn't specify
2722 service in request". It would seem that Dovecot has changed its
2723 interface. Fortunately there's a specification; I followed it and
2724 changed what the client sends and it appears to be working now.
2726 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2729 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2730 address A is aliased to B and C, where B exists and C does not. Without
2731 -v the output is "A verified" because verification stops after a
2732 successful redirection if more than one address is generated. However,
2733 with -v the child addresses are also verified. Exim was outputting "A
2734 failed to verify" and then showing the successful verification for C,
2735 with its parentage. It now outputs "B failed to verify", showing B's
2736 parentage before showing the successful verification of C.
2738 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2739 look up a TXT record in a specific list after matching in a combined
2742 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2743 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2744 they consult the DNS. I had assumed they would set it the way they
2745 wanted; and indeed my experiments on Linux seem to show that in some
2746 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2747 To be on the safe side, however, I have now made the interface to
2748 host_find_byname() similar to host_find_bydns(), with an argument
2749 containing the DNS resolver options. The host_find_byname() function now
2750 sets these options at its start, just as host_find_bydns() does. The smtp
2751 transport options dns_qualify_single and dns_search_parents are passed to
2752 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2753 of host_find_byname() use the default settings of RES_DEFNAMES
2754 (qualify_single) but not RES_DNSRCH (search_parents).
2756 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2757 spool_read_header() do less string testing, by means of a preliminary
2758 switch on the second character of optional "-foo" lines. (This is
2759 overdue, caused by the large number of possibilities that now exist.
2760 Originally there were few.) While I was there, I also converted the
2761 str(n)cmp tests so they don't re-test the leading "-" and the first
2762 character, in the hope this might squeeze out yet more improvement.
2764 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2765 flag allowing group syntax was set by the header_syntax check but not
2766 turned off, possible causing trouble later; (2) The flag was not being
2767 set at all for the header_verify test, causing "group"-style headers to
2768 be rejected. I have now set it in this case, and also caused header_
2769 verify to ignore an empty address taken from a group. While doing this, I
2770 came across some other cases where the code for allowing group syntax
2771 while scanning a header line wasn't quite right (mostly, not resetting
2772 the flag correctly in the right place). These bugs could have caused
2773 trouble for malformed header lines. I hope it is now all correct.
2775 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2776 with the "reply" argument non-NULL. The code, however (which originally
2777 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2778 but it didn't always do it. This confused somebody who was copying the
2779 code for some other use. I have removed all the tests.
2781 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2782 feature that was used to support insecure browsers during the U.S. crypto
2783 embargo. It requires special client support, and Exim is probably the
2784 only MTA that supported it -- and would never use it because real RSA is
2785 always available. This code has been removed, because it had the bad
2786 effect of slowing Exim down by computing (never used) parameters for the
2787 RSA_EXPORT functionality.
2789 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2790 authenticator to fail if there's a tab character in the incoming data
2791 (there should never be unless someone is messing about, as it's supposed
2792 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2793 if the connection is using TLS or if the remote IP is the same as the
2794 local IP, and the "valid-client-cert option" if a client certificate has
2797 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2798 authenticators. This can be used for authorization after authentication
2799 succeeds. (In the case of plaintext, it servers for both authentication
2802 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2803 if any retry times were supplied.
2805 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2806 connection, where there is no client IP address to check. In this
2807 situation, the verify now always succeeds.
2809 PH/25 Applied John Jetmore's -Mset patch.
2811 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2813 PH/27 In a string expansion for a processed (not raw) header when multiple
2814 headers of the same name were present, leading whitespace was being
2815 removed from all of them, but trailing whitespace was being removed only
2816 from the last one. Now trailing whitespace is removed from each header
2817 before concatenation. Completely empty headers in a concatenation (as
2818 before) are ignored.
2820 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2821 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2823 PH/29 [Removed. This was a change that I later backed out, and forgot to
2824 correct the ChangeLog entry (that I had efficiently created) before
2825 committing the later change.]
2827 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2828 address errors (4xx response to RCPT) over the same connection as other
2829 messages routed to the same hosts. Such deliveries are always "forced",
2830 so retry times are not inspected. This resulted in far too many retries
2831 for the affected addresses. The effect occurred only when there were more
2832 hosts than the hosts_max_try setting in the smtp transport when it had
2833 the 4xx errors. Those hosts that it had tried were not added to the list
2834 of hosts for which the message was waiting, so if all were tried, there
2835 was no problem. Two fixes have been applied:
2837 (i) If there are any address or message errors in an SMTP delivery, none
2838 of the hosts (tried or untried) are now added to the list of hosts
2839 for which the message is waiting, so the message should not be a
2840 candidate for sending over the same connection that was used for a
2841 successful delivery of some other message. This seems entirely
2842 reasonable: after all the message is NOT "waiting for some host".
2843 This is so "obvious" that I'm not sure why it wasn't done
2844 previously. Hope I haven't missed anything, but it can't do any
2845 harm, as the worst effect is to miss an optimization.
2847 (ii) If, despite (i), such a delivery is accidentally attempted, the
2848 routing retry time is respected, so at least it doesn't keep
2849 hammering the server.
2851 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2852 in ${readsocket because some servers need this prod.
2854 PH/32 Added some extra debug output when updating a wait-xxx database.
2856 PH/33 The hint "could be header name not terminated by colon", which has been
2857 given for certain expansion errors for a long time, was not being given
2858 for the ${if def:h_colon_omitted{... case.
2860 PH/34 The spec says: "With one important exception, whenever a domain list is
2861 being scanned, $domain contains the subject domain." There was at least
2862 one case where this was not true.
2864 PH/35 The error "getsockname() failed: connection reset by peer" was being
2865 written to the panic log as well as the main log, but it isn't really
2866 panic-worthy as it just means the connection died rather early on. I have
2867 removed the panic log writing for the ECONNRESET error when getsockname()
2870 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2871 runs only) independently of the message's sender address. This meant
2872 that, if the 4xx error was in fact related to the sender, a different
2873 message to the same recipient with a different sender could confuse
2874 things. In particular, this can happen when sending to a greylisting
2875 server, but other circumstances could also provoke similar problems.
2876 I have changed the default so that the retry time for these errors is now
2877 based a combination of the sender and recipient addresses. This change
2878 can be overridden by setting address_retry_include_sender=false in the
2881 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2882 remote server are returned as part of bounce messages. This was not
2883 happening for LMTP over a pipe (the lmtp transport), but now it is the
2884 same for both kinds of LMTP.
2886 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2887 in header lines that were in fact CNAMEs. This is no longer the case.
2889 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2890 and queue runs started by the daemon processed all messages. This has
2891 been fixed so that -R and -S can now usefully be given with -q<time>.
2893 PH/40 Import PCRE release 6.7 (fixes some bugs).
2895 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2897 PH/42 Give an error if -q is specified more than once.
2899 PH/43 Renamed the variables $interface_address and $interface_port as
2900 $received_ip_address and $received_port, to make it clear that these
2901 values apply to message reception, and not to the outgoing interface when
2902 a message is delivered. (The old names remain recognized, of course.)
2904 PH/44 There was no timeout on the connect() call when using a Unix domain
2905 socket in the ${readsocket expansion. There now is.
2907 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2908 be meaningful with "accept".
2910 SC/01 Eximstats V1.43
2911 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2913 SC/02 Eximstats V1.44
2914 Use a glob alias rather than an array ref in the generated
2915 parser. This improves both readability and performance.
2917 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2918 Collect SpamAssassin and rejection statistics.
2919 Don't display local sender or destination tables unless
2920 there is data to show.
2921 Added average volumes into the top table text output.
2923 SC/04 Eximstats V1.46
2924 Collect data on the number of addresses (recipients)
2925 as well as the number of messages.
2927 SC/05 Eximstats V1.47
2928 Added 'Message too big' to the list of mail rejection
2929 reasons (thanks to Marco Gaiarin).
2931 SC/06 Eximstats V1.48
2932 Mainlog lines which have GMT offsets and are too short to
2933 have a flag are now skipped.
2935 SC/07 Eximstats V1.49 (Alain Williams)
2936 Added the -emptyok flag.
2938 SC/08 Eximstats V1.50
2939 Fixes for obtaining the IP address from reject messages.
2941 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2942 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2943 whitespace changes from 4.64-PH/27
2945 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2948 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2949 are found, allow negative numbers in numeric criteria)
2951 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2953 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2954 to match changes made in 4.64-PH/43
2956 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2958 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2959 "domain" is actually an IP address, and force a failure. This locks out
2960 those revolvers/nameservers that support "A-for-A" lookups, in
2961 contravention of the specifications.
2963 PH/48 When a host name was looked up from an IP address, and the subsequent
2964 forward lookup of the name timed out, the host name was left in
2965 $sender_host_name, contrary to the specification.
2967 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2968 restricted to single-key lookups, Exim was not diagnosing an error if
2969 * or *@ was used with a query-style lookup.
2971 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2973 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2974 long ago, but noone interested enough thought of it. Let's just say that
2975 the "1.1" means that there are some new functions that weren't there at
2976 some point in the past.
2978 PH/51 Error processing for expansion failure of helo_data from an smtp
2979 transport during callout processing was broken.
2981 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2982 tested/used via the -bh/-bhc/-bs options.
2984 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2985 bug, fixed in subsequent PCRE releases).
2987 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2988 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2990 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2996 SC/01 Use a glob alias rather than an array ref in eximstats generated
2997 parser. This improves both readability and performance.
2999 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3000 Don't display local sender or destination tables in eximstats unless
3001 there is data to show.
3002 Added average volumes into the eximstats top table text output.
3004 SC/03 Collect data on the number of addresses (recipients) as well
3005 as the number of messages in eximstats.
3007 TF/01 Correct an error in the documentation for the redirect router. Exim
3008 does (usually) call initgroups() when daemonizing.
3010 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3011 with consistent privilege compared to when running as a daemon.
3013 TF/03 Note in the spec that $authenticated_id is not set for local
3014 submissions from trusted users.
3016 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3017 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3019 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3020 by adding some example configuration directives to the default
3021 configuration file. A little bit of work is required to uncomment the
3022 directives and define how usernames and passwords are checked, but
3023 there is now a framework to start from.
3025 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3026 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3027 without this. I don't know how relevant this is to other LDAP libraries.
3029 PH/02 Add the verb name to the "unknown ACL verb" error.
3031 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3033 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3035 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3036 directory not expanded when it should be if an expanded home directory
3037 was set for the address (which is overridden by the transport).
3039 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3042 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3043 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3044 because it is too late at that time, and has no effect.
3046 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3047 security issue with \' (bugzilla #107). I could not use the
3048 PQescapeStringConn() function, because it needs a PGconn value as one of
3051 PH/08 When testing addresses using -bt, indicate those final addresses that
3052 are duplicates that would not cause an additional delivery. At least one
3053 person was confused, thinking that -bt output corresponded to deliveries.
3054 (Suppressing duplicates isn't a good idea as you lose the information
3055 about possibly different redirections that led to the duplicates.)
3057 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3058 systems where poll() doesn't work, in particular OS X.
3060 PH/10 Added more information to debugging output for retry time not reached.
3062 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3063 operations in malware.c.
3065 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3068 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3069 syslog_duplication set false, log lines that would normally be written
3070 both the the main log and to the reject log were not written to syslog at
3073 PH/14 In the default configuration, change the use of "message" in ACL warn
3074 statements to "add_header".
3076 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3077 not followed by a command (e.g. "seen endif").
3079 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3080 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3083 PH/17 Added extra conditions to the default value of delay_warning_condition
3087 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3088 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3089 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3092 The Auto-Submitted: and various List- headers are standardised, whereas I
3093 don't think Precedence: ever was.
3095 PH/18 Refactored debugging code in route_finduser() to show more information,
3096 in particular, the error code if getpwnam() issues one.
3098 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3099 This is apparently needed in addition to the PH/07 change above to avoid
3100 any possible encoding problems.
3102 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3103 but not after initializing Perl.
3105 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3106 output them only if debugging. By default they are written stderr,
3107 apparently, which is not desirable.
3109 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3112 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3115 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3117 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3118 authenticated or an ident call has been made. Suppress the default
3119 values for $authenticated_id and $authenticated_sender (but permit -oMai
3120 and -oMas) when testing with -bh.
3122 PH/24 Re-jigged the order of the tests in the default configuration so that the
3123 tests for valid domains and recipients precede the DNS black list and CSA
3124 tests, on the grounds that those ones are more expensive.
3126 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3127 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3128 This bug exists in every version of Exim that I still have, right back to
3131 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3132 However, an attempt to turn on case-sensitivity in a regex key by
3133 including (?-i) didn't work because the subject string was already
3134 lowercased, and the effects were non-intuitive. It turns out that a
3135 one-line patch can be used to allow (?-i) to work as expected.
3141 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3142 other effects) broke the use of negated acl sub-conditions.
3144 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3147 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3148 "Deny" causes Exim to reject the incoming connection with a 554 error.
3149 Unfortunately, if there is a major crisis, such as a disk failure,
3150 tcp-wrappers gives "deny", whereas what one would like would be some
3151 kind of temporary error. A kludge has been added to help with this.
3152 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3153 554 error is used if errno is still zero or contains ENOENT (which occurs
3154 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3157 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3159 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3160 errors. Otherwise a message that provokes a temporary error (when other
3161 messages do not) can cause a whole host to time out.
3163 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3164 addresses were routed directly to files or pipes from a redirect router.
3165 File deliveries just didn't batch; pipe deliveries might have suffered
3168 PH/06 A failure to get a lock for a hints database would erroneously always say
3169 "Failed to get write lock", even when it was really a read lock.
3171 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3172 of 0600. This has been changed to use the value of the lockfile_mode
3173 option (which defaults to 0600).
3175 PH/08 Applied small patch from the Sieve maintainer.
3177 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3178 folder from quota calculations, a direct delivery into this folder messed
3179 up the contents of the maildirsize file. This was because the regex was
3180 used only to exclude .Trash (or whatever) when the size of the mailbox
3181 was calculated. There was no check that a delivery was happening into an
3182 excluded directory. This bug has been fixed by ignoring all quota
3183 processing for deliveries into excluded directories.
3185 PH/10 Added the maildirfolder_create_regex option to appendfile.
3191 PH/01 The code for finding all the local interface addresses on a FreeBSD
3192 system running IPv6 was broken. This may well have applied to all BSD
3193 systems, as well as to others that have similar system calls. The broken
3194 code found IPv4 interfaces correctly, but gave incorrect values for the
3195 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3196 that it would not match correctly against @[] and not recognize the IPv6
3199 PH/02 The ipliteral router was not recognizing addresses of the form user@
3200 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3202 PH/03 Added disable_ipv6.
3204 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3205 decoded form, because it is most often used to construct To: headers
3206 lines in autoreplies, and the decoded form may well be syntactically
3207 invalid. However, $reply_address has leading white space removed, and all
3208 newlines turned into spaces so that the autoreply transport does not
3211 PH/05 If group was specified without a user on a router, and no group or user
3212 was specified on a transport, the group from the router was ignored.
3214 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3215 for visible compile-time settings that can be used to change these
3216 numbers, for those that want even more. Backwards compatibility with old
3217 spool files has been maintained. However, going back to a previous Exim
3218 release will lost any variables that are in spool files.
3220 PH/07 Two small changes when running in the test harness: increase delay when
3221 passing a TCP/IP connection to a new process, in case the original
3222 process has to generate a bounce, and remove special handling of
3223 127.0.0.2 (sic), which is no longer necessary.
3225 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3226 be the same on different OS.
3228 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3231 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3232 whether --show-vars was specified or not
3234 JJ/02 exipick: Added support for new ACL variable spool format introduced
3237 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3238 syntactically invalid From: or Reply-to: line, and a filter used this to
3239 generate an autoreply, and therefore failed to obtain an address for the
3240 autoreply, Exim could try to deliver to a non-existent relative file
3241 name, causing unrelated and misleading errors. What now happens is that
3242 it logs this as a hard delivery error, but does not attempt to create a
3245 PH/11 The exinext utility has a -C option for testing purposes, but although
3246 the given file was scanned by exinext itself; it wasn't being passed on
3247 when Exim was called.
3249 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3250 an end-of-file indication when reading a command response.
3252 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3253 compiled. In many other places in Exim, IPv6 addresses are always
3254 recognized, so I have changed this. It also means that IPv4 domain
3255 literals of the form [IPV4:n.n.n.n] are now always recognized.
3257 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3258 used if the router is not running as root, for example, when verifying at
3259 ACL time, or when using -bh. The debugging output from this situation was
3260 non-existent - all you got was a failure to exec. I have made two
3263 (a) Failures to set uid/gid, the current directory, or a process leader
3264 in a subprocess such as that created by queryprogram now generate
3265 suitable debugging output when -d is set.
3267 (b) The queryprogram router detects when it is not running as root,
3268 outputs suitable debugging information if -d is set, and then runs
3269 the subprocess without attempting to change uid/gid.
3271 PH/15 Minor change to Makefile for building test_host (undocumented testing
3274 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3275 additional section of a DNS packet that returns MX or SRV records.
3276 Instead, it always explicitly searches for A/AAAA records. This avoids
3277 major problems that occur when a DNS server includes only records of one
3278 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3279 fixed another bug: if SRV records were looked up and the corresponding
3280 address records were *not* found in the additional section, the port
3281 values from the SRV records were lost.
3283 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3284 using the correct key (the original address) when searching the retry
3285 rules in order to find which one to use for generating the retry hint.
3287 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3288 adding the default one. Similarly, if it contains a Reply-To: header, the
3289 errors_reply_to option, if set, is not used.
3291 PH/19 When calculating a retry time, Exim used to measure the "time since
3292 failure" by looking at the "first failed" field in the retry record. Now
3293 it does not use this if it is later than than the arrival time of the
3294 message. Instead it uses the arrival time. This makes for better
3295 behaviour in cases where some deliveries succeed, thus re-setting the
3296 "first failed" field. An example is a quota failure for a huge message
3297 when small messages continue to be delivered. Without this change, the
3298 "time since failure" will always be short, possible causing more frequent
3299 delivery attempts for the huge message than are intended.
3300 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3302 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3303 $1, $2, $3) because the numerical variables can be reset during some
3304 expansion items (e.g. "match"), thereby losing the authentication data.
3306 PH/21 Make -bV show the size of off_t variables so that the test suite can
3307 decide whether to run tests for quotas > 2G.
3309 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3310 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3311 filecount value is greater than 2G or if a quota value is greater than 2G
3312 on a system where the size of off_t is not greater than 4, a panic error
3315 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3316 never match. The debug and -bh output now contains an explicit error
3317 message indicating a malformed IPv4 address or mask.
3319 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3320 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3321 PH/23 above applies.
3323 PH/25 Do not write to syslog when running in the test harness. The only
3324 occasion when this arises is a failure to open the main or panic logs
3325 (for which there is an explicit test).
3327 PH/26 Added the /no_tell option to "control=freeze".
3329 PH/27 If a host name lookup failed very early in a connection, for example, if
3330 the IP address matched host_lookup and the reverse lookup yielded a name
3331 that did not have a forward lookup, an error message of the form "no IP
3332 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3333 could be logged. Now it outputs the IP address instead of "NULL".
3335 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3336 allows the sender and the authenticated sender to be set when
3337 submitting a message from within Exim. Since child_open_exim() is
3338 documented for local_scan(), the new function should be too.
3340 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3341 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3342 results in an empty string is now treated as unset.
3344 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3346 PH/31 Added sender_verify_fail logging option.
3348 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3349 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3350 tidied the source and removed it altogether.
3352 PH/33 When a queue run was abandoned because the load average was too high, a
3353 log line was always written; now it is written only if the queue_run log
3354 selector is set. In addition, the log line for abandonment now contains
3355 information about the queue run such as the pid. This is always present
3356 in "start" and "stop" lines but was omitted from the "abandon" line.
3358 PH/34 Omit spaces between a header name and the colon in the error message that
3359 is given when verify = headers_syntax fails (if there are lots of them,
3360 the message gets confusing).
3362 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3363 names, as there are now some PTR records that contain slashes. This check
3364 is only to protect against broken name servers that fall over on strange
3365 characters, so the fact that it applies to all lookups doesn't matter.
3367 PH/36 Now that the new test suite is complete, we can remove some of the
3368 special code in Exim that was needed for the old test suite. For example,
3369 sorting DNS records because real resolvers return them in an arbitrary
3370 order. The new test suite's fake resolver always returns records in the
3373 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3374 bounces) except when queue_only is set, to avoid logging races between
3375 the different processes.
3377 PH/38 Panic-die if .include specifies a non-absolute path.
3379 PH/39 A tweak to the "H" retry rule from its user.
3381 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3382 a label. They prevented compilation on older perls.
3384 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3385 a warning to be raised on newish perls.
3387 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3388 on queue. Changes to match documented behaviour of showing count of
3389 messages matching specified criteria.
3391 PH/40 Changed the default ident timeout from 30s to 5s.
3393 PH/41 Added support for the use of login_cap features, on those BSD systems
3394 that have them, for controlling the resources used by pipe deliveries.
3396 PH/42 The content-scanning code uses fopen() to create files in which to put
3397 message data. Previously it was not paying any attention to the mode of
3398 the files. Exim runs with umask(0) because the rest of the code creates
3399 files with open(), and sets the required mode explicitly. Thus, these
3400 files were ending up world-writeable. This was not a big issue, because,
3401 being within the spool directory, they were not world-accessible. I have
3402 created a function called modefopen, which takes an additional mode
3403 argument. It sets umask(777), creates the file, chmods it to the required
3404 mode, then resets the umask. All the relevant calls to fopen() in the
3405 content scanning code have been changed to use this function.
3407 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3408 to 24 hours. This avoids potential overflow problems when processing G
3409 and H retry rules. I suspect nobody ever tinkers with this value.
3411 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3413 PH/45 When the plaintext authenticator is running as a client, the server's
3414 challenges are checked to ensure they are valid base64 strings. By
3415 default, the authentication attempt is cancelled if an invalid string is
3416 received. Setting client_ignore_invalid_base64 true ignores these errors.
3417 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3418 they are received. Thus, the responses can be made to depend on the
3419 challenges. If an invalid string is ignored, an empty string is placed in
3422 PH/46 Messages that are created by the autoreply transport now contains a
3423 References: header, in accordance with RFCs 2822 and 3834.
3425 PH/47 Added authenticated_sender_force to the smtp transport.
3427 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3429 PH/49 Installed latest patch from the Sieve maintainer.
3431 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3432 was also set, the appendfile transport was unnecessarily scanning a
3433 directory of message files (e.g. for maildir delivery) to find the count
3434 of files (along with the size), even though it did not need this
3435 information. It now does the scan only if it needs to find either the
3436 size of the count of files.
3438 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3440 PH/52 Two bugs concerned with error handling when the smtp transport is
3443 (i) Exim was not creating retry information for temporary errors given
3444 for individual recipients after the DATA command when the smtp transport
3445 was used in LMTP mode. This meant that they could be retried too
3446 frequently, and not timed out correctly.
3448 (ii) Exim was setting the flag that allows error details to be returned
3449 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3450 recipients that were returned after the DATA command.
3452 PH/53 This is related to PH/52, but is more general: for any failing address,
3453 when detailed error information was permitted to be returned to the
3454 sender, but the error was temporary, then after the final timeout, only
3455 "retry timeout exceeded" was returned. Now it returns the full error as
3456 well as "retry timeout exceeded".
3458 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3459 do this, and (what is worse) MTAs that accept it.
3461 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3462 will now be deprecated.
3464 PH/56 New os.c-cygwin from the Cygwin maintainer.
3466 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3467 formats (previously only available in exim formats via -bpr, -bpru,
3468 and -bpra. Now also available in native and exiqgrep formats)
3470 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3471 with very large, slow to parse queues
3473 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3475 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3477 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3478 responses to authentication challenges, though it was showing the
3479 challenges; (ii) I've removed the CR characters from the debug output for
3482 PH/58 Allow for the insertion of a newline as well as a space when a string
3483 is turned into more than one encoded-word during RFC 2047 encoding. The
3484 Sieve code now uses this.
3486 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3487 data_4xx, lost_connection, tls_required.
3489 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3490 message was being sent as an SMTP response.
3492 PH/61 Add -l and -k options to exicyclog.
3494 PH/62 When verifying, if an address was redirected to one new address, so that
3495 verification continued, and the new address failed or deferred after
3496 having set something in $address_data, the value of $address_data was not
3497 passed back to the ACL. This was different to the case when no
3498 redirection occurred. The value is now passed back in both cases.
3500 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3501 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3502 use login_cap.h, so on its own it isn't the distinguishing feature. The
3503 new name refers directly to the setclassresources() function.
3505 PH/65 Added configuration files for NetBSD3.
3507 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3509 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3510 is preferred over IPv4.
3512 PH/68 The bounce_return_message and bounce_return_body options were not being
3513 honoured for bounces generated during the reception of non-SMTP messages.
3514 In particular, this applied to messages rejected by the ACL. This bug has
3515 been fixed. However, if bounce_return_message is true and bounce_return_
3516 body is false, the headers that are returned for a non-SMTP message
3517 include only those that have been read before the error was detected.
3518 (In the case of an ACL rejection, they have all been read.)
3520 PH/69 The HTML version of the specification is now built in a directory called
3521 spec_html instead of spec.html, because the latter looks like a path with
3522 a MIME-type, and this confuses some software.
3524 PH/70 Catch two compiler warnings in sieve.c.
3526 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3527 function verify_get_ident() calls ip_connect() to connect a socket, but
3528 if the "connect()" function timed out, ip_connect() used to close the
3529 socket. However, verify_get_ident() also closes the socket later, and in
3530 between Exim writes to the log, which may get opened at this point. When
3531 the socket was closed in ip_connect(), the log could get the same file
3532 descriptor number as the socket. This naturally causes chaos. The fix is
3533 not to close the socket in ip_connect(); the socket should be closed by
3534 the function that creates it. There was only one place in the code where
3535 this was missing, in the iplookup router, which I don't think anybody now
3536 uses, but I've fixed it anyway.
3538 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3539 well as to direct DNS lookups. Otherwise the handling of names in host
3540 lists is inconsistent and therefore confusing.
3546 PH/01 Two changes to the default runtime configuration:
3548 (1) Move the checks for relay_from_hosts and authenticated clients from
3549 after to before the (commented out) DNS black list checks.
3551 (2) Add control=submission to the relay_from_hosts and authenticated
3552 clients checks, on the grounds that messages accepted by these
3553 statements are most likely to be submissions.
3555 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3557 (1) Generate an error if the third argument for the ${prvs expansion is
3560 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3563 (3) Reset the variables that are obtained from the first argument of
3564 ${prvscheck and used in the second argument before leaving the code,
3565 because their memory is reclaimed, so using them afterwards may do
3568 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3569 one (it's much easier than Tom thought :-).
3571 (5) Because of (4), we can now allow for the use of $prvscheck_result
3572 inside the third argument.
3574 PH/03 For some reason, the default setting of PATH when running a command from
3575 a pipe transport was just "/usr/bin". I have changed it to
3578 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3579 anything to be listed in the output from -bV.
3581 PH/05 When a filter generated an autoreply, the entire To: header line was
3582 quoted in the delivery log line, like this:
3584 => >A.N.Other <ano@some.domain> <original@ddress> ...
3586 This has been changed so that it extracts the operative address. There
3587 may be more than one such address. If so, they are comma-separated, like
3590 => >ano@some.domain,ona@other.domain <original@ddress> ...
3592 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3593 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3594 not being looked up in the rDNS to get a host name, Exim was showing the
3595 IP address twice in Received: lines, even though the IP addresses were
3596 identical. For example:
3598 Received: from [1.2.3.4] (helo=[1.2.3.4])
3600 However, if the real host name was known, it was omitting the HELO data
3601 if it matched the actual IP address. This has been tidied up so that it
3602 doesn't show the same IP address twice.
3604 PH/07 When both +timestamp and +memory debugging was on, the value given by
3605 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3606 called by the debug printing, thereby overwriting the timestamp buffer.
3607 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3609 PH/08 When the original message was included in an autoreply transport, it
3610 always said "this is a copy of the message, including all the headers",
3611 even if body_only or headers_only was set. It now gives an appropriate
3614 PH/09 Applied a patch from the Sieve maintainer which:
3616 o fixes some comments
3617 o adds the (disabled) notify extension core
3618 o adds some debug output for the result of if/elsif tests
3619 o points to the current vacation draft in the documentation
3620 and documents the missing references header update
3624 o fixes a bug in processing the envelope test (when testing
3625 multiple envelope elements, the last element determined the
3628 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3629 Electronic Mail") by including:
3631 Auto-submitted: auto-generated
3633 in the messages that it generates (bounce messages and others, such as
3634 warnings). In the case of bounce messages for non-SMTP messages, there was
3635 also a typo: it was using "Auto_submitted" (underscore instead of
3636 hyphen). Since every message generated by Exim is necessarily in response
3637 to another message, thes have all been changed to:
3639 Auto-Submitted: auto-replied
3641 in accordance with these statements in the RFC:
3643 The auto-replied keyword:
3645 - SHOULD be used on messages sent in direct response to another
3646 message by an automatic process,
3648 - MUST NOT be used on manually-generated messages,
3650 - MAY be used on Delivery Status Notifications (DSNs) and Message
3651 Disposition Notifications (MDNs),
3653 - MUST NOT be used on messages generated by automatic or periodic
3654 processes, except for messages which are automatic responses to
3657 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3658 to the default Received: header definition.
3660 PH/12 Added log selector acl_warn_skipped (default on).
3662 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3663 variables because (a) they are in the wrong storage pool and (b) even if
3664 they were copied, it wouldn't work properly because of the caching.
3666 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3667 checking when decoding. Apparently there are clients that generate
3668 overlong encoded strings. Why am I not surprised?
3670 PH/15 If the first argument of "${if match_address" was not empty, but did not
3671 contain an "@" character, Exim crashed. Now it writes a panic log message
3672 and treats the condition as false.
3674 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3676 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3677 extension "envelope-auth". The code is finished and in agreement with
3678 other implementations, but there is no documentation so far and in fact,
3679 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3680 not changing the active code.
3682 Print executed "if" and "elsif" statements when debugging is used. This
3683 helps a great deal to understand what a filter does.
3685 Document more things not specified clearly in RFC3028. I had all this
3686 sorted out, when out of a sudden new issues came to my mind. Oops."
3688 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3691 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3692 canonical form (as documented). However, after a host name lookup from
3693 the IP address, check_host() was doing a simple string comparison with
3694 addresses acquired from the DNS when checking that the found name did
3695 have the original IP as one of its addresses. Since any found IPv6
3696 addresses are likely to be in abbreviated form, the comparison could
3697 fail. Luckily, there already exists a function for doing the comparison
3698 by converting both addresses to binary, so now that is used instead of
3699 the text comparison.
3701 PH/20 There was another similar case to PH/19, when a complete host name was
3702 given in a host list; looking up its IP address could give an abbreviated
3703 form, whereas the current host's name might or might not be abbreviated.
3704 The same fix has been applied.
3710 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3711 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3714 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3715 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3717 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3719 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3720 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3721 Also, for "Auto-Submitted", treat anything other than "no" as
3722 non-personal, in accordance with RFC 3834. (Previously it treated
3723 anything starting "auto-" as non-personal.)
3725 TF/01 The control=submission/name=... option had a problem with syntax
3726 errors if the name included a slash character. The /name= option
3727 now slurps the rest of the string, so it can include any characters
3728 but it must come last in the list of options (after /sender_retain
3731 PH/05 Some modifications to the interface to the fake nameserver for the new
3739 TK/01 Added the "success_on_redirect" address verification option. See
3740 NewStuff for rationale and an example.
3742 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3744 PH/02 Patch to exigrep to allow it to work on syslog lines.
3746 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3747 fread() to skip over the body file's header line, because in Cygwin the
3748 header line is locked and is inaccessible.
3750 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3751 co-exist for some time) to make it clear that it is the Exim ID that is
3752 referenced, not the Message-ID: header line.
3754 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3755 string_format() function, because snprintf() does not exist on all
3758 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3759 ${stat: expansion item.
3761 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3762 protocol synchronization error", to keep the pedants happy.
3764 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3765 well as for IRIX systems, when gcc is being used. See the host.c source
3768 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3770 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3773 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3774 transport and to the smtp transport in LMTP mode.
3776 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3778 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3779 run a filter in a subprocess. This could lead to confusion in subsequent
3780 lookups in the parent process. There should also be a search_tidyup() at
3781 the end of the subprocess.
3783 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3784 only if the host matched helo_try_verify_hosts, which caused the
3785 verification to occur when the EHLO/HELO command was issued. The ACL just
3786 tested the remembered result. Now, if a previous verification attempt has
3787 not happened, "verify = helo" does it there and then.
3789 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3791 TK/03 Fix log output including CR from clamd.
3793 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3794 exist provoked a memory error which could cause a segfault.
3796 PH/15 Installed PCRE 6.2
3798 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3800 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3801 of the problem. Specifically, suggested +O2 rather than +O1 for the
3804 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3806 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3807 record, the port was not correctly logged when the outgoing_port log
3808 selector was set (it logged the transort's default port).
3810 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3811 fallback_hosts, and "hosts" in the smtp transport.
3813 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3814 host errors such as "Connection refused".
3816 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3817 authentication with radiusclient 0.4.9:
3819 - Error returned from rc_read_config was caught wrongly
3820 - Username/password not passed on to radius server due to wrong length.
3822 The presumption is that some radiusclient API changes for 4.51/PH/17
3823 were not taken care of correctly. The code is still untested by me (my
3824 Linux distribution still has 0.3.2 of radiusclient), but it was
3825 contributed by a Radius user.
3827 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3828 expanding the "port" option of the smtp transport.
3830 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3831 while reading a MIME header. Thanks to Tom Hughes for a patch.
3833 PH/24 Include config.h inside local_scan.h so that configuration settings are
3836 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3837 that in an ACL for RCPT (for example), you can examine exactly what was
3840 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3841 commands, but it was not correctly comparing the address with the actual
3842 client host address. Thus, it would show the EHLO address in Received:
3843 header lines when this was not necessary.
3845 PH/27 Added the % operator to ${eval:}.
3847 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3848 it should be ignoring failures (because with -C, for example, it has lost
3849 privilege). It wasn't ignoring creation failures other than "already
3852 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3855 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3856 because an input line was too long, either on its own, or by virtue of
3857 too many continuations, the temporary file was not being removed, and the
3858 return code was incorrect.
3860 PH/31 Missing "BOOL" in function definition in filtertest.c.
3862 PH/32 Applied Sieve patches from the maintainer.
3864 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3866 PH/33 Added "verify = not_blind".
3868 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3869 Local/Makefile (with some defaults set). These are used in built scripts
3870 such as exicyclog, but they have never been used in the exim_install
3871 script (though there are many overriding facilities there). I have
3872 arranged that the exim_install script now takes note of these two
3875 PH/35 Installed configuration files for Dragonfly.
3877 PH/36 When a locally submitted message by a trusted user did not contain a
3878 From: header, and the sender address was obtained from -f or from an SMTP
3879 MAIL command, and the trusted user did not use -F to supply a sender
3880 name, $originator_name was incorrectly used when constructing a From:
3881 header. Furthermore, $originator_name was used for submission mode
3882 messages from external hosts without From: headers in a similar way,
3883 which is clearly wrong.
3885 PH/37 Added control=suppress_local_fixups.
3887 PH/38 When log_selector = +received_sender was set, and the addition of the
3888 sender made the log line's construction buffer exactly full, or one byte
3889 less than full, an overflow happened when the terminating "\n" was
3892 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3893 when the result of a list match is failure because a DNS lookup failed.
3895 PH/40 RM_COMMAND is now used in the building process.
3897 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3898 the "build-* directories that it finds.
3900 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3901 address in a domain literal was a prefix of an interface address.
3903 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3904 when verifying a sender address, unless rewrite_headers is false.
3906 PH/44 Wrote a long comment about why errors_to addresses are verified as
3907 recipients, not senders.
3909 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3910 the ratelimit ACL was added.
3912 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3914 PH/46 Added extra information about PostgreSQL errors to the error string.
3916 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3917 avoiding the need to install special zones in a real server. This is
3918 backwards compatible; if it can't find the fake resolver, it drops back.
3919 Thus, both old and new test suites can be run.
3921 TF/02 Added util/ratelimit.pl
3923 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3924 clock is set back in time.
3926 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3927 Candler <B.Candler@pobox.com>.
3929 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3930 OK for addresses that are the result of redirections.
3932 PH/48 A number of further additions for the benefit of the new test suite,
3933 including a fake gethostbyname() that interfaces to the fake DNS resolver
3936 TF/06 The fix for widen_domains has also been applied to qualify_single and
3937 search_parents which are the other dnslookup options that can cause
3940 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3943 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3945 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3946 service" if it encountered an MX record with an empty target hostname.
3947 The message is now "an MX or SRV record indicated no SMTP service".
3949 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3950 if the DNS of the sending site is misconfigured. This is quite a
3951 common situation. This change restores the behaviour of treating a
3952 helo verification defer as a failure.
3954 PH/51 If self=fail was set on a router, the bounce message did not include the
3955 actual error message.
3961 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3963 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3964 command itself ended in error, the underlying message about the transport
3965 filter timeout was being overwritten with the pipe command error. Now the
3966 underlying error message should be appended to the second error message.
3968 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3970 PH/02 Reduce the amount of output that "make" produces by default. Full output
3971 can still be requested.
3973 PH/03 The warning log line about a condition test deferring for a "warn" verb
3974 was being output only once per connection, rather than after each
3975 occurrence (because it was using the same function as for successful
3976 "warn" verbs). This seems wrong, so I have changed it.
3978 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3979 it should not have, which might have caused a crash in the right
3980 circumstances, but probably never did.
3982 PH/04 Installed a modified version of Tony Finch's patch to make submission
3983 mode fix the return path as well as the Sender: header line, and to
3984 add a /name= option so that you can make the user's friendly name appear
3987 TF/03 Added the control = fakedefer ACL modifier.
3989 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3990 Mark Lowes for thorough testing.
3992 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3994 TK/03 Merged latest SRS patch from Miles Wilton.
3996 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3997 with the definition in sysexits.h (which is #included earlier).
3998 Fortunately, Exim does not actually use EX_OK. The code used to try to
3999 preserve the sysexits.h value, by assuming that macro definitions were
4000 scanned for macro replacements. I have been disabused of this notion,
4001 so now the code just undefines EX_OK before #including unistd.h.
4003 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4004 in the smtp transport. When a block could not be written in a single
4005 write() function, the timeout was being re-applied to each part-write.
4006 This seems wrong - if the receiver was accepting one byte at a time it
4007 would take for ever. The timeout is now adjusted when this happens. It
4008 doesn't have to be particularly precise.
4010 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4011 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4013 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4014 without a domain if the check to <postmaster@domain> fails.
4016 SC/01 Eximstats: added -xls and the ability to specify output files
4017 (patch written by Frank Heydlauf).
4019 SC/02 Eximstats: use FileHandles for outputting results.
4021 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4023 SC/04 Eximstats: fixed display of large numbers with -nvr option
4025 SC/05 Eximstats: fixed merging of reports with empty tables.
4027 SC/06 Eximstats: added the -include_original_destination flag
4029 SC/07 Eximstats: removed tabs and trailing whitespace.
4031 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4033 TK/06 MBOX spool code: Add real "From " MBOX separator line
4034 so the .eml file is really in mbox format (even though
4035 most programs do not really care). Patch from Alex Miller.
4037 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4038 The latter is generated from $received_to and is only set if the
4039 message has one envelope recipient. SA can use these headers,
4040 obviously out-of-the-box. Patch from Alex Miller.
4042 PH/08 The ${def test on a variable was returning false if the variable's
4043 value was "0", contrary to what the specification has always said!
4044 The result should be true unless the variable is empty.
4046 PH/09 The syntax error of a character other than { following "${if
4047 def:variable_name" (after optional whitespace) was not being diagnosed.
4048 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4049 accidental colon was present, for example, could give incorrect results.
4051 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4052 result is used (not including appendfile, where other changes are about
4055 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4056 This involved changing a lot of size variables from int to off_t. It
4057 should work with maildirs and everything.
4059 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4060 spamd dying while we are connected to it.
4062 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4065 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4066 function for BDB 4.3.
4068 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4070 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4071 the sources for PCRE have been re-arranged and more files are now
4074 PH/15 The code I had for printing potentially long long variables in PH/11
4075 above was not the best (it lost precision). The length of off_t variables
4076 is now inspected at build time, and an appropriate printing format (%ld
4077 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4078 to be "long long int" or "long int". This is needed for the internal
4079 formatting function string_vformat().
4081 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4082 the configuration file to be ":syslog", then the script "guesses" where
4083 the logs files are, rather than using the compiled in default. In our
4084 case the guess is not the same as the compiled default, so the script
4085 suddenly stopped working when I started to use syslog. The patch checks
4086 to see if log_file_path is "". If so, it attempts to read it from exim
4087 with no configuration file to get the compiled in version, before it
4088 falls back to the previous guessing code."
4090 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4091 implementing BATV in an Exim configuration. See NewStuff for the gory
4094 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4095 Makefile that are specific to HP-UX.
4097 PH/18 If the "use_postmaster" option was set for a recipient callout together
4098 with the "random" option, the postmaster address was used as the MAIL
4099 FROM address for the random test, but not for the subsequent recipient
4100 test. It is now used for both.
4102 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4103 patch removes a few documentation additions to RFC 3028, because the
4104 latest draft now contains them. It adds the new en;ascii-case comparator
4105 and a new error check for 8bit text in MIME parts. Comparator and
4106 require names are now matched exactly. I enabled the subaddress
4107 extension, but it is not well tested yet (read: it works for me)."
4109 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4110 rework some of the code of TK/09 above to avoid the hardwired use of
4111 "%lld" and "long long". Replaced the call to snprintf() with a call to
4114 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4115 records point to non-existent hosts", "retry timeout exceeded", and
4116 "retry time not reached for any host after a long failure period".
4118 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4119 experimental DomainKeys support:
4121 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4122 (2) On an error such as an illegally used "control", the wrong name for
4123 the control was given.
4125 These problems did NOT occur unless DomainKeys support was compiled.
4127 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4129 PH/24 Added ${if match_ip condition.
4131 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4132 databases so that it will be absolutely obvious if a crash occurs in the
4133 DB library. This is a regular occurrence (often caused by mis-matched
4136 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4137 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4138 picked up on a user's system that detects such things. There doesn't seem
4139 to be a gcc warning option for this - only an attribute that has to be
4140 put on the function's prototype. It seems that in Fedora Core 4 they have
4141 set this on a number of new functions. No doubt there will be more in due
4144 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4145 specify a transport. However, if an address that was verified by such a
4146 router was the subject of a callout, Exim crashed because it tried to
4147 read the rcpt_include_affixes from the non-existent transport. Now it
4148 just assumes that the setting of that option is false. This bug was
4149 introduced by 4.51/PH/31.
4151 PH/28 Changed -d+all to exclude +memory, because that information is very
4152 rarely of interest, but it makes the output a lot bigger. People tend to
4153 do -d+all out of habit.
4155 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4156 code in os-type was giving problems when libc.so lives in lib64, like on
4159 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4160 aren't the modern standard, and it seems that some systems' include files
4161 don't always have them. Exim was already checking for some of the newer
4162 ones like T_AAAA, and defining it itself. I've added checks for all the
4163 record types that Exim uses.
4165 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4166 not automatically generating a new one, as it is supposed to. This
4167 prevented TLS from working. If the file did exist, but contained invalid
4168 data, a new version was generated, as expected. It was only the case of a
4169 non-existent file that was broken.
4171 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4172 with a change in libdomainkeys > 0.64.
4174 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4175 from DNS. If the selector record carries the flag, it now has
4176 precedence over the domain-wide flag.
4178 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4180 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4181 the use of a port name that isn't defined in /etc/services) occurred, the
4182 message was deferred as in a normal delivery, and thus remained on the
4183 spool, instead of being failed because of the mua_wrapper setting. This
4184 is now fixed, and I tidied up some of the mua_wrapper messages at the
4187 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4188 the messages in a hash of arrays rather than using individual hashes.
4189 This is a bit cleaner and results in dramatic memory savings, albeit
4190 at a slight CPU cost.
4192 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4193 as requested by Marc Sherman.
4195 SC/10 Eximstats: added histograms for user specified patterns as requested
4198 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4200 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4201 fopen() in the content-scanning modules that did not already have it.
4207 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4208 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4210 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4212 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4214 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4215 to test Sieve filters that use "vacation".
4217 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4218 that changes the way the GnuTLS parameters are stored in the cache file.
4219 The new format can be generated externally. For backward compatibility,
4220 if the data in the cache doesn't make sense, Exim assumes it has read an
4221 old-format file, and it generates new data and writes a new file. This
4222 means that you can't go back to an older release without removing the
4225 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4226 work if there are any delivery delays because "one_time" forces the
4227 parent to be marked "delivered", so its unseen clone is never tried
4228 again. For this reason, Exim now forbids the simultaneous setting of
4231 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4232 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4233 entry if you want to know the details. The fix, however, neglected to
4234 consider the case when local delivery batching is involved. The test for
4235 "previously delivered" was not happening when checking to see if an
4236 address could be batched with a previous (undelivered) one; under
4237 certain circumstances this could lead to multiple deliveries to the same
4240 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4241 in its include files, and this causes problems building Exim.
4243 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4244 header_syntax) but Exim was just ignoring anything given after a slash.
4245 In particular, this caused confusion with an attempt to use "verify =
4246 reverse_host_lookup/defer_ok". An error is now given when options are
4247 supplied for verify items that do not have them. (Maybe reverse_host_
4248 lookup should have a defer_ok option, but that's a different point.)
4250 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4251 defined by RFC 821) to 2048, because there were problems with some AUTH
4252 commands, and RFC 1869 says the size should be increased for extended
4253 SMTP commands that take arguments.
4255 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4258 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4259 "unknown" error; now it says that the functionality isn't in the binary.
4261 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4262 an address' error message when a string expansion fails (syntax or
4263 whatever). Otherwise the password may appear in the log. Following change
4264 PH/42 below, there is no longer a chance of it appearing in a bounce
4267 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4269 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4270 of its addresses was ever tried. (Bugzilla bug #2.)
4272 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4273 the result incorrectly in the debug output. (It correctly added a newline
4274 to what was transported.)
4276 TF/01 Added $received_time.
4278 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4279 commented out examples of how to interface to a virus scanner and to
4280 SpamAssassin. Also added commented examples of av_scanner and
4281 spamd_address settings.
4283 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4284 and controls are allowed in which ACLs. There were a couple of minor
4285 errors. Some of the entries in the conditions table (which is a table of
4286 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4287 negation of where the condition IS allowed.
4289 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4291 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4292 header file does not have a version number, so I've had to invent a new
4293 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4294 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4295 radiusclient), but it was contributed by a Radius user.
4297 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4298 files or directories, for OpenSSL.
4300 PH/19 When an Exim process that is running as root has to create an Exim log
4301 file, it does so in a subprocess that runs as exim:exim so as to get the
4302 ownership right at creation (otherwise, other Exim processes might see
4303 the file with the wrong ownership). There was no test for failure of this
4304 fork() call, which would lead to the process getting stuck as it waited
4305 for a non-existent subprocess. Forks do occasionally fail when resources
4306 run out. I reviewed all the other calls to fork(); they all seem to check
4309 PH/20 When checking for unexpected SMTP input at connect time (before writing
4310 the banner), Exim was not dealing correctly with a non-positive return
4311 from the read() function. If the client had disconnected by this time,
4312 the result was a log entry for a synchronization error with an empty
4313 string after "input=" when read() returned zero. If read() returned -1
4314 (an event I could not check), uninitialized data bytes were printed.
4315 There were reports of junk text (parts of files, etc) appearing after
4318 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4320 PH/22 Added support for macro redefinition, and (re)definition in between
4321 driver and ACL definitions.
4323 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4324 forgetting to use the resulting value; it was using the unexpanded value.
4326 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4327 hadn't been configured. The fix is from Juergen Kreileder, who
4328 understands it better than I do:
4330 "Here's what I see happening with three configured cyrus_sasl
4331 authenticators configured (plain, login, cram-md5):
4333 On startup auth_cyrus_sasl_init() gets called for each of these.
4334 This means three calls to sasl_listmech() without a specified mech_list.
4335 => SASL tests which mechs of all available mechs actually work
4336 => three warnings about OTP not working
4337 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4339 With the patch, sasl_listmech() also gets called three times. But now
4340 SASL's mech_list option is set to the server_mech specified in the the
4341 authenticator. Or in other words, the answer from sasl_listmech()
4342 gets limited to just the mech you're testing for (which is different
4344 => the return list contains just 'plain' or 'login', 'cram-md5' or
4345 nothing depending on the value of ob->server_mech.
4347 I've just tested the patch: Authentication still works fine,
4348 unavailable mechs specified in the exim configuration are still
4349 caught, and the auth.log warnings about OTP are gone."
4351 PH/25 When debugging is enabled, the contents of the command line are added
4352 to the debugging output, even when log_selector=+arguments is not
4355 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4356 answer is "GNU", and only if the return is "GNU/something" is the answer
4359 PH/27 $acl_verify_message is now set immediately after the failure of a
4360 verification in an ACL, and so is available in subsequent modifiers. In
4361 particular, the message can be preserved by coding like this:
4363 warn !verify = sender
4364 set acl_m0 = $acl_verify_message
4366 Previously, $acl_verify_message was set only while expanding "message"
4367 and "log_message" when a very denied access.
4369 PH/28 Modified OS/os.c-Linux with
4371 -#ifndef OS_LOAD_AVERAGE
4372 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4374 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4375 nomenclature these days.)
4377 PH/29 Installed patch from the Sieve maintainer that adds the options
4378 sieve_useraddress and sieve_subaddress to the redirect router.
4380 PH/30 In these circumstances:
4381 . Two addresses routed to the same list of hosts;
4382 . First host does not offer TLS;
4383 . First host accepts first address;
4384 . First host gives temporary error to second address;
4385 . Second host offers TLS and a TLS session is established;
4386 . Second host accepts second address.
4387 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4388 and peerdn, if requested) that were in fact used only for the second
4391 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4392 attention to any local part prefix or suffix that was matched by the
4393 router that accepted the address. It now behaves in the same way as it
4394 does for delivery: the affixes are removed from the local part unless
4395 rcpt_include_affixes is set on the transport.
4397 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4398 timeout during the DATA phase of an incoming message.
4400 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4401 applied a patch sanctioned by the Sieve maintainer.
4403 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4404 the uid or gid is negative. A case of a negative gid caused this to be
4405 noticed. The fix allows for either to be negative.
4407 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4408 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4411 PH/36 The change PH/12 above was broken. Fixed it.
4413 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4414 the grounds that routing the same address twice would always produce the
4415 same answer. This might have been true once, but it is certainly no
4416 longer true now. Routing a child address may depend on the previous
4417 routing that produced that child. Some complicated redirection strategies
4418 went wrong when messages had multiple recipients, and made Exim's
4419 behaviour dependent on the order in which the addresses were given.
4421 I have moved the duplicate checking until after the routing is complete.
4422 Exim scans the addresses that are assigned to local and remote
4423 transports, and removes any duplicates. This means that more work will be
4424 done, as duplicates will always all be routed, but duplicates are
4425 presumably rare, so I don't expect this is of any significance.
4427 For deliveries to pipes, files, and autoreplies, the duplicate checking
4428 still happens during the routing process, since they are not going to be
4431 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4432 It corrects a timeout issue with spamd. This is Ian's comment: "The
4433 background is that sometimes spamd either never reads data from a
4434 connection it has accepted, or it never writes response data. The exiscan
4435 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4436 blindly assumes that writes won't block so it may never time out."
4438 PH/39 Allow G after quota size as well as K and M.
4440 PH/40 The value set for $authenticated_id in an authenticator may not contain
4441 binary zeroes or newlines because the value is written to log lines and
4442 to spool files. There was no check on this. Now the value is run through
4443 the string_printing() function so that such characters are converted to
4444 printable escape sequences.
4446 PH/41 $message_linecount is a new variable that contains the total number of
4447 lines in the message. Compare $body_linecount, which is the count for the
4450 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4451 bounce and delay warning messages, except in certain special cases, which
4454 (a) An SMTP error message from a remote host;
4455 (b) A message specified in a :fail: redirection;
4456 (c) A message specified in a "fail" command in a system filter;
4457 (d) A message specified in a FAIL return from the queryprogram router;
4458 (e) A message specified by the cannot_route_message router option.
4460 In these cases only, Exim does include the error details in bounce and
4461 warning messages. There are also a few cases where bland messages such
4462 as "unrouteable address" or "local delivery error" are given.
4464 PH/43 $value is now also set for the "else" part of a ${run expansion.
4466 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4467 being worked on, but at least Exim now implements the latest version to
4470 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4471 process to complete was treated as a delivery failure, a timeout while
4472 writing the message to the pipe was logged, but erroneously treated as a
4473 successful delivery. Such timeouts include transport filter timeouts. For
4474 consistency with the overall process timeout, these timeouts are now
4475 treated as errors, giving rise to delivery failures by default. However,
4476 there is now a new Boolean option for the pipe transport called
4477 timeout_defer, which, if set TRUE, converts the failures into defers for
4478 both kinds of timeout. A transport filter timeout is now identified in
4481 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4482 systems where "make" and "gmake" are different, calling "gmake" at top
4483 level broke things. I've arranged for the value of $(MAKE) to be passed
4484 from the Makefile to this script so that it can call the same version of
4488 A note about Exim versions 4.44 and 4.50
4489 ----------------------------------------
4491 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4492 changes of various kinds. As a consequence, a big documentation update was
4493 needed. This delayed the release for rather longer than seemed good, especially
4494 in the light of a couple of (minor) security issues. Therefore, the changes
4495 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4496 release. So 4.44 and 4.50 are in effect two different branches that both start
4499 I have left the 4.50 change log unchanged; it contains all the changes since
4500 4.43. The change log for 4.44 is below; many of its items are identical to
4501 those for 4.50. This seems to be the most sensible way to preserve the
4502 historical information.
4508 1. Minor wording change to the doc/README.SIEVE file.
4510 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4511 computation of the current number of files was incorrect.
4513 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4514 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4517 4. Give more explanation in the error message when the command for a transport
4518 filter fails to execute.
4520 5. There are several places where Exim runs a non-Exim command in a
4521 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4522 was being done only for the command run by the queryprogram router. It is
4523 now done for all such subprocesses. The other cases are: ${run, transport
4524 filters, and the commands run by the lmtp and pipe transports.
4526 6. Added CONFIGURE_GROUP build-time option.
4528 7. Some older OS have a limit of 256 on the maximum number of file
4529 descriptors. Exim was using setrlimit() to set 1000 as a large value
4530 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4531 systems. I've change it so that if it can't get 1000, it tries for 256.
4533 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4534 was an oversight, and furthermore, ever since the addition of extra
4535 controls (e.g. 4.43/32), the checks on when to allow different forms of
4536 "control" were broken. There should now be diagnostics for all cases when a
4537 control that does not make sense is encountered.
4539 9. Added the /retain_sender option to "control=submission".
4541 10. $recipients is now available in the predata ACL (oversight).
4543 11. Tidy the search cache before the fork to do a delivery from a message
4544 received from the command line. Otherwise the child will trigger a lookup
4545 failure and thereby defer the delivery if it tries to use (for example) a
4546 cached ldap connection that the parent has called unbind on.
4548 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4549 of $address_data from the recipient verification was clobbered by the
4550 sender verification.
4552 13. The value of address_data from a sender verification is now available in
4553 $sender_address_data in subsequent conditions in the ACL statement.
4555 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4557 15. Added a new option "connect=<time>" to callout options, to set a different
4560 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4561 was its contents. (It was OK if the option was not defined at all.)
4563 17. A "Completed" log line is now written for messages that are removed from
4564 the spool by the -Mrm option.
4566 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4567 information about exactly what failed.
4569 19. Added -dd to debug only the daemon process.
4571 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4572 handles timeouts, both on the server side and network timeouts. Renamed the
4573 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4575 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4576 It is now set to "smtps".
4578 22. $host_address is now set to the target address during the checking of
4579 ignore_target_hosts.
4581 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4582 being passed; this would have caused $sender_host_name to have been used if
4583 matching the list had actually called for a host name (not very likely,
4584 since this list is usually IP addresses). A host name is now passed as
4587 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4588 code that specifies a non-restarting handler (typically sigaction() in
4589 modern systems) in an attempt to fix a rare and obscure crash bug.
4591 25. Narrowed the window for a race in the daemon that could cause it to ignore
4592 SIGCHLD signals. This is not a major problem, because they are used only to
4593 wake it up if nothing else does.
4595 26. A malformed maildirsize file could cause Exim to calculate negative values
4596 for the mailbox size or file count. Odd effects could occur as a result.
4597 The maildirsize information is now recalculated if the size or filecount
4600 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4601 support for a long time. Removed HAVE_SYS_VFS_H.
4603 28. Installed the latest version of exipick from John Jetmore.
4605 29. In an address list, if the pattern was not a regular expression, an empty
4606 subject address (from a bounce message) matched only if the pattern was an
4607 empty string. Non-empty patterns were not even tested. This was the wrong
4608 because it is perfectly reasonable to use an empty address as part of a
4609 database query. An empty address is now tested by patterns that are
4610 lookups. However, all the other forms of pattern expect the subject to
4611 contain a local part and a domain, and therefore, for them, an empty
4612 address still always fails if the pattern is not itself empty.
4614 30. Exim went into a mad DNS loop when attempting to do a callout where the
4615 host was specified on an smtp transport, and looking it up yielded more
4616 than one IP address.
4618 31. Re-factored the code for checking spool and log partition space into a
4619 function that finds that data and another that does the check. The former
4620 is then used to implement four new variables: $spool_space, $log_space,
4621 $spool_inodes, and $log_inodes.
4623 32. The RFC2047 encoding function was originally intended for short strings
4624 such as real names; it was not keeping to the 75-character limit for
4625 encoded words that the RFC imposes. It now respects the limit, and
4626 generates multiple encoded words if necessary. To be on the safe side, I
4627 have increased the buffer size for the ${rfc2047: expansion operator from
4630 33. It is now permitted to omit both strings after an "if" condition; if the
4631 condition is true, the result is "true". As before, when the second string
4632 is omitted, a false condition yields an empty string. This makes it less
4633 cumbersome to write custom ACL and router conditions.
4635 34. Failure to deliver a bounce message always caused it to be frozen, even if
4636 there was an errors_to setting on the router. The errors_to setting is now
4639 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4640 canonical form (fully expanded) before being placed in
4641 $sender_host_address.
4643 36. The table in the code that translates DNS record types into text (T_A to
4644 "A" for instance) was missing entries for NS and CNAME. It is just possible
4645 that this could have caused confusion if both these types were looked up
4646 for the same domain, because the text type is used as part of Exim's
4647 per-process caching. But the chance of anyone hitting this buglet seems
4650 37. The dnsdb lookup has been extended in a number of ways.
4652 (1) There is a new type, "zns", which walks up the domain tree until it
4653 finds some nameserver records. It should be used with care.
4655 (2) There is a new type, "mxh", which is like "mx" except that it returns
4656 just the host names, not the priorities.
4658 (3) It is now possible to give a list of domains (or IP addresses) to be
4659 looked up. The behaviour when one of the lookups defers can be
4660 controlled by a keyword.
4662 (4) It is now possible to specify the separator character for use when
4663 multiple records are returned.
4665 38. The dnslists ACL condition has been extended: it is now possible to supply
4666 a list of IP addresses and/or domains to be looked up in a particular DNS
4669 39. Added log_selector=+queue_time_overall.
4671 40. When running the queue in the test harness, wait just a tad after forking a
4672 delivery process, to get repeatability of debugging output.
4674 41. Include certificate and key file names in error message when GnuTLS fails
4675 to set them up, because the GnuTLS error message doesn't include the name
4676 of the failing file when there is a problem reading it.
4678 42. Allow both -bf and -bF in the same test run.
4680 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4682 44. The "Exiscan patch" is now merged into the mainline Exim source.
4684 45. Sometimes the final signoff response after QUIT could fail to get
4685 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4686 < 0 before doing a fflush(). This bug looks as though it goes back to the
4687 introduction of TLS in release 3.20, but "sometimes" must have been rare
4688 because the tests only now provoked it.
4690 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4691 (this can affect the format of dates).
4693 47. exim_tidydb, when checking for the continued existence of a message for
4694 which it has found a message-specific retry record, was not finding
4695 messages that were in split spool directories. Consequently, it was
4696 deleting retry records that should have stayed in existence.
4698 48. Steve fixed some bugs in eximstats.
4700 49. The SPA authentication driver was not abandoning authentication and moving
4701 on to the next authenticator when an expansion was forced to fail,
4702 contradicting the general specification for all authenticators. Instead it
4703 was generating a temporary error. It now behaves as specified.
4705 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4706 (the order specifies the preference for clients). The order is now AES256,
4707 AES128, 3DES, ARCFOUR128.
4709 51. Small patch to Sieve code - explicitly set From: when generating an
4712 52. Exim crashed if a remote delivery caused a very long error message to be
4713 recorded - for instance if somebody sent an entire SpamAssassin report back
4714 as a large number of 550 error lines. This bug was coincidentally fixed by
4715 increasing the size of one of Exim's internal buffers (big_buffer) that
4716 happened as part of the Exiscan merge. However, to be on the safe side, I
4717 have made the code more robust (and fixed the comments that describe what
4720 53. Now that there can be additional text after "Completed" in log lines (if
4721 the queue_time_overall log selector is set), a one-byte patch to exigrep
4722 was needed to allow it to recognize "Completed" as not the last thing in
4725 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4726 patch that reportedly fixes this has been added. I am not expert enough to
4727 create a test for it. This is what the patch creator wrote:
4729 "I found a little strange behaviour of ldap code when working with
4730 Windows 2003 AD Domain, where users was placed in more than one
4731 Organization Units. When I tried to give exim partial DN, the exit code
4732 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4733 But simultaneously result of request was absolutely normal ldap result,
4734 so I produce this patch..."
4736 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4737 REFERENCE, so I have modified the code to exclude the patch when that macro
4740 55. Some experimental protocols are using DNS PTR records for new purposes. The
4741 keys for these records are domain names, not reversed IP addresses. The
4742 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4743 leaves it alone. Component reversal etc. now happens only for IP addresses.
4746 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4748 57. Double the size of the debug message buffer (to 2048) so that more of very
4749 long debug lines gets shown.
4751 58. The exicyclog utility now does better if the number of log files to keep
4752 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4754 59. Two changes related to the smtp_active_hostname option:
4756 (1) $smtp_active_hostname is now available as a variable.
4757 (2) The default for smtp_banner uses $smtp_active_hostname instead
4758 of $primary_hostname.
4760 60. The host_aton() function is supposed to be passed a string that is known
4761 to be a valid IP address. However, in the case of IPv6 addresses, it was
4762 not checking this. This is a hostage to fortune. Exim now panics and dies
4763 if the condition is not met. A case was found where this could be provoked
4764 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4765 components; fortuitously, this particular loophole had already been fixed
4766 by change 4.50/55 above.
4768 If there are any other similar loopholes, the new check in host_aton()
4769 itself should stop them being exploited. The report I received stated that
4770 data on the command line could provoke the exploit when Exim was running as
4771 exim, but did not say which command line option was involved. All I could
4772 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4773 running as the user.
4776 61. There was a buffer overflow vulnerability in the SPA authentication code
4777 (which came originally from the Samba project). I have added a test to the
4778 spa_base64_to_bits() function which I hope fixes it.
4781 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4782 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4784 63. The daemon start-up calls getloadavg() while still root for those OS that
4785 need the first call to be done as root, but it missed one case: when
4786 deliver_queue_load_max is set with deliver_drop_privilege. This is
4787 necessary for the benefit of the queue runner, because there is no re-exec
4788 when deliver_drop_privilege is set.
4790 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4791 This has been fixed.
4793 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4794 was in use, was not putting the data itself into the right store pool;
4795 consequently, it could be overwritten for a subsequent message in the same
4796 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4799 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4801 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4802 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4803 the function were treating the return as a boolean value, which happened to
4804 work because 0=false and not-0=true, but is not correct code.
4806 68. The host_aton() function was not handling scoped IPv6 addresses (those
4807 with, for example, "%eth0" on the end) correctly.
4809 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4810 negated items (that is, ~something) in unsigned ints. Some compilers
4811 apparently mutter when there is no cast.
4813 70. If an address verification called from an ACL failed, and did not produce a
4814 user-specific message (i.e. there was only a "system" message), nothing was
4815 put in $acl_verify_message. In this situation, it now puts the system
4818 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4819 session; change 4.31/43 added the unwanted input to the log line - except
4820 that it did not do this in the start of session case. It now does.
4822 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4823 This is wrong and can cause the other end to generate a synchronization
4824 error if it is another Exim or anything else that does the synchronization
4825 check. A QUIT command is no longer sent after a timeout.
4827 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4828 during host lookups.
4830 74. The defer_ok option of callout verification was not working if it was used
4831 when verifying addresses in header lines, that is, for this case:
4833 verify = header_sender/callout=defer_ok
4835 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4836 those file descriptors could be used for SMTP connections. If anything
4837 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4838 could be sent to the SMTP client, causing chaos. The daemon now opens
4839 stdin, stdout, and stderr to /dev/null when it puts itself into the
4842 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4843 log by default. The user can override this with suitable Perl magic.
4845 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4846 the log message when discard triggers, was not working for the DATA ACL or
4847 for the non-SMTP ACL.
4849 78. Error message wording change in sieve.c.
4851 79. If smtp_accept_max_per_host was set, the number of connections could be
4852 restricted to fewer than expected, because the daemon was trying to set up
4853 a new connection before checking whether the processes handling previous
4854 connections had finished. The check for completed processes is now done
4855 earlier. On busy systems, this bug wouldn't be noticed because something
4856 else would have woken the daemon, and it would have reaped the completed
4859 80. If a message was submitted locally by a user whose login name contained one
4860 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4861 It caused a spool format error. I have fixed the spool reading code. A
4862 related problem was that the "from" clause in the Received: line became
4863 illegal because of the space(s). It is now covered by ${quote_local_part.
4865 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4868 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4870 83. Patch from Sieve maintainer to fix unterminated string problem in
4871 "vacation" handling.
4873 84. Some minor changes to the Linux configuration files to help with other
4874 OS variants using glibc.
4876 85. One more patch for Sieve to update vacation handling to latest spec.
4879 ----------------------------------------------------
4880 See the note above about the 4.44 and 4.50 releases.
4881 ----------------------------------------------------
4887 1. Change 4.43/35 introduced a bug that caused file counts to be
4888 incorrectly computed when quota_filecount was set in an appendfile
4891 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4892 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4895 3. Give more explanation in the error message when the command for a transport
4896 filter fails to execute.
4898 4. There are several places where Exim runs a non-Exim command in a
4899 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4900 was being done only for the command run by the queryprogram router. It is
4901 now done for all such subprocesses. The other cases are: ${run, transport
4902 filters, and the commands run by the lmtp and pipe transports.
4904 5. Some older OS have a limit of 256 on the maximum number of file
4905 descriptors. Exim was using setrlimit() to set 1000 as a large value
4906 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4907 systems. I've change it so that if it can't get 1000, it tries for 256.
4909 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4910 was an oversight, and furthermore, ever since the addition of extra
4911 controls (e.g. 4.43/32), the checks on when to allow different forms of
4912 "control" were broken. There should now be diagnostics for all cases when a
4913 control that does not make sense is encountered.
4915 7. $recipients is now available in the predata ACL (oversight).
4917 8. Tidy the search cache before the fork to do a delivery from a message
4918 received from the command line. Otherwise the child will trigger a lookup
4919 failure and thereby defer the delivery if it tries to use (for example) a
4920 cached ldap connection that the parent has called unbind on.
4922 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4923 of $address_data from the recipient verification was clobbered by the
4924 sender verification.
4926 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4927 was its contents. (It was OK if the option was not defined at all.)
4929 11. A "Completed" log line is now written for messages that are removed from
4930 the spool by the -Mrm option.
4932 12. $host_address is now set to the target address during the checking of
4933 ignore_target_hosts.
4935 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4936 being passed; this would have caused $sender_host_name to have been used if
4937 matching the list had actually called for a host name (not very likely,
4938 since this list is usually IP addresses). A host name is now passed as
4941 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4942 code that specifies a non-restarting handler (typically sigaction() in
4943 modern systems) in an attempt to fix a rare and obscure crash bug.
4945 15. Narrowed the window for a race in the daemon that could cause it to ignore
4946 SIGCHLD signals. This is not a major problem, because they are used only to
4947 wake it up if nothing else does.
4949 16. A malformed maildirsize file could cause Exim to calculate negative values
4950 for the mailbox size or file count. Odd effects could occur as a result.
4951 The maildirsize information is now recalculated if the size or filecount
4954 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4955 support for a long time. Removed HAVE_SYS_VFS_H.
4957 18. Updated exipick to current release from John Jetmore.
4959 19. Allow an empty sender to be matched against a lookup in an address list.
4960 Previously the only cases considered were a regular expression, or an
4963 20. Exim went into a mad DNS lookup loop when doing a callout where the
4964 host was specified on the transport, if the DNS lookup yielded more than
4967 21. The RFC2047 encoding function was originally intended for short strings
4968 such as real names; it was not keeping to the 75-character limit for
4969 encoded words that the RFC imposes. It now respects the limit, and
4970 generates multiple encoded words if necessary. To be on the safe side, I
4971 have increased the buffer size for the ${rfc2047: expansion operator from
4974 22. Failure to deliver a bounce message always caused it to be frozen, even if
4975 there was an errors_to setting on the router. The errors_to setting is now
4978 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4979 canonical form (fully expanded) before being placed in
4980 $sender_host_address.
4982 24. Updated eximstats to version 1.33
4984 25. Include certificate and key file names in error message when GnuTLS fails
4985 to set them up, because the GnuTLS error message doesn't include the name
4986 of the failing file when there is a problem reading it.
4988 26. Expand error message when OpenSSL has problems setting up cert/key files.
4991 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4992 (this can affect the format of dates).
4994 28. exim_tidydb, when checking for the continued existence of a message for
4995 which it has found a message-specific retry record, was not finding
4996 messages that were in split spool directories. Consequently, it was
4997 deleting retry records that should have stayed in existence.
4999 29. eximstats updated to version 1.35
5000 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5001 1.35 - bugfix such that pie charts by volume are generated correctly
5003 30. The SPA authentication driver was not abandoning authentication and moving
5004 on to the next authenticator when an expansion was forced to fail,
5005 contradicting the general specification for all authenticators. Instead it
5006 was generating a temporary error. It now behaves as specified.
5008 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5009 (the order specifies the preference for clients). The order is now AES256,
5010 AES128, 3DES, ARCFOUR128.
5012 31. Small patch to Sieve code - explicitly set From: when generating an
5015 32. Exim crashed if a remote delivery caused a very long error message to be
5016 recorded - for instance if somebody sent an entire SpamAssassin report back
5017 as a large number of 550 error lines. This bug was coincidentally fixed by
5018 increasing the size of one of Exim's internal buffers (big_buffer) that
5019 happened as part of the Exiscan merge. However, to be on the safe side, I
5020 have made the code more robust (and fixed the comments that describe what
5023 33. Some experimental protocols are using DNS PTR records for new purposes. The
5024 keys for these records are domain names, not reversed IP addresses. The
5025 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5026 leaves it alone. Component reversal etc. now happens only for IP addresses.
5029 34. The host_aton() function is supposed to be passed a string that is known
5030 to be a valid IP address. However, in the case of IPv6 addresses, it was
5031 not checking this. This is a hostage to fortune. Exim now panics and dies
5032 if the condition is not met. A case was found where this could be provoked
5033 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5034 components; fortuitously, this particular loophole had already been fixed
5035 by change 4.50/55 or 4.44/33 above.
5037 If there are any other similar loopholes, the new check in host_aton()
5038 itself should stop them being exploited. The report I received stated that
5039 data on the command line could provoke the exploit when Exim was running as
5040 exim, but did not say which command line option was involved. All I could
5041 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5042 running as the user.
5045 35. There was a buffer overflow vulnerability in the SPA authentication code
5046 (which came originally from the Samba project). I have added a test to the
5047 spa_base64_to_bits() function which I hope fixes it.
5050 36. The daemon start-up calls getloadavg() while still root for those OS that
5051 need the first call to be done as root, but it missed one case: when
5052 deliver_queue_load_max is set with deliver_drop_privilege. This is
5053 necessary for the benefit of the queue runner, because there is no re-exec
5054 when deliver_drop_privilege is set.
5056 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5057 was in use, was not putting the data itself into the right store pool;
5058 consequently, it could be overwritten for a subsequent message in the same
5059 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5062 38. Sometimes the final signoff response after QUIT could fail to get
5063 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5064 < 0 before doing a fflush(). This bug looks as though it goes back to the
5065 introduction of TLS in release 3.20, but "sometimes" must have been rare
5066 because the tests only now provoked it.
5072 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5073 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5074 Now it returns an integer. A number of calls were still expecting a T/F
5075 return. Fortuitously, in all cases, the tests worked in OK situations,
5076 which is the norm. However, things would have gone wrong on any write
5077 failures on the smtp file descriptor. This function is used when sending
5078 messages over SMTP and also when doing verify callouts.
5080 2. When Exim is called to do synchronous delivery of a locally submitted
5081 message (the -odf or -odi options), it no longer closes stderr before doing
5084 3. Implemented the mua_wrapper option.
5086 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5088 5. Implemented the functions header_remove(), header_testname(),
5089 header_add_at_position(), and receive_remove_recipient(), and exported them
5092 6. If an ACL "warn" statement specified the addition of headers, Exim already
5093 inserted X-ACL-Warn: at the start if there was no header name. However, it
5094 was not making this test for the second and subsequent header lines if
5095 there were newlines in the string. This meant that an invalid header could
5096 be inserted if Exim was badly configured.
5098 7. Allow an ACL "warn" statement to add header lines at the start or after all
5099 the Received: headers, as well as at the end.
5101 8. Added the rcpt_4xx retry error code.
5103 9. Added postmaster_mailfrom=xxx to callout verification option.
5105 10. Added mailfrom=xxxx to the callout verification option, for verify=
5108 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5109 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5111 12. Inserted some casts to stop certain compilers warning when using pointer
5112 differences as field lengths or precisions in printf-type calls (mostly
5113 affecting debugging statements).
5115 13. Added optional readline() support for -be (dynamically loaded).
5117 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5118 same clock tick as a message's arrival, so that its received time was the
5119 same as the "first fail" time on the retry record, and that message
5120 remained on the queue past the ultimate address timeout, every queue runner
5121 would try a delivery (because it was past the ultimate address timeout) but
5122 after another failure, the ultimate address timeout, which should have then
5123 bounced the address, did not kick in. This was a "< instead of <=" error;
5124 in most cases the first failure would have been in the next clock tick
5125 after the received time, and all would be well.
5127 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5128 being recognized when the domain list was tested by the match_domain
5129 condition in an expansion string.
5131 16. Added the ${str2b64: operator.
5133 17. Exim was always calling setrlimit() to set a large limit for the number of
5134 processes, without checking whether the existing limit was already
5135 adequate. (It did check for the limit on file descriptors.) Furthermore,
5136 errors from getrlimit() and setrlimit() were being ignored. Now they are
5137 logged to the main and panic logs, but Exim does carry on, to try to do its
5138 job under whatever limits there are.
5140 18. Imported PCRE 5.0.
5142 19. Trivial typo in log message " temporarily refused connection" (the leading
5145 20. If the log selector return_path_on_delivery was set and an address was
5146 redirected to /dev/null, the delivery process crashed because it assumed
5147 that a return path would always be set for a "successful" delivery. In this
5148 case, the whole delivery is bypassed as an optimization, and therefore no
5151 21. Internal re-arrangement: the function for sending a challenge and reading
5152 a response while authentication was assuming a zero-terminated challenge
5153 string. It's now changed to take a pointer and a length, to allow for
5154 binary data in such strings.
5156 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5158 23. Exim was not respecting finduser_retries when seeking the login of the
5159 uid under which it was called; it was always trying 10 times. (The default
5160 setting of finduser_retries is zero.) Also, it was sleeping after the final
5161 failure, which is pointless.
5163 24. Implemented tls_on_connect_ports.
5165 25. Implemented acl_smtp_predata.
5167 26. If the domain in control=submission is set empty, Exim assumes that the
5168 authenticated id is a complete email address when it generates From: or
5169 Sender: header lines.
5171 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5172 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5173 chown and chgrp in /bin and hostname in /usr/bin.
5175 28. Exim was keeping the "process log" file open after each use, just as it
5176 does for the main log. This opens the possibility of it remaining open for
5177 long periods when the USR1 signal hits a daemon. Occasional processlog
5178 errors were reported, that could have been caused by this. Anyway, it seems
5179 much more sensible not to leave this file open at all, so that is what now
5182 29. The long-running daemon process does not normally write to the log once it
5183 has entered its main loop, and it closes the log before doing so. This is
5184 so that log files can straightforwardly be renamed and moved. However,
5185 there are a couple of unusual error situations where the daemon does write
5186 log entries, and I had neglected to close the log afterwards.
5188 30. The text of an SMTP error response that was received during a remote
5189 delivery was being truncated at 512 bytes. This is too short for some of
5190 the long messages that one sometimes sees. I've increased the limit to
5193 31. It is now possible to make retry rules that apply only when a message has a
5194 specific sender, in particular, an empty sender.
5196 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5197 it possible to be selective about when SMTP synchronization is enforced.
5199 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5201 32. Implemented hosts_connection_nolog.
5203 33. Added an ACL for QUIT.
5205 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5208 35. Added mailbox_size and mailbox_filecount to appendfile.
5210 36. Added control = no_multiline_responses to ACLs.
5212 37. There was a bug in the logic of the code that waits for the clock to tick
5213 in the case where the clock went backwards by a substantial amount such
5214 that the microsecond fraction of "now" was more than the microsecond
5215 fraction of "then" (but the whole seconds number was less).
5217 38. Added support for the libradius Radius client library this is found on
5218 FreeBSD (previously only the radiusclient library was supported).
5224 1. When certain lookups returned multiple values in the form name=value, the
5225 quoting of the values was not always being done properly. Specifically:
5226 (a) If the value started with a double quote, but contained no whitespace,
5228 (b) If the value contained whitespace other than a space character (i.e.
5229 tabs or newlines or carriage returns) it was not quoted.
5230 This fix has been applied to the mysql and pgsql lookups by writing a
5231 separate quoting function and calling it from the lookup code. The fix
5232 should probably also be applied to nisplus, ibase and oracle lookups, but
5233 since I cannot test any of those, I have not disturbed their existing code.
5235 2. A hit in the callout cache for a specific address caused a log line with no
5236 reason for rejecting RCPT. Now it says "Previous (cached) callout
5237 verification failure".
5239 3. There was an off-by-one bug in the queryprogram router. An over-long
5240 return line was truncated at 256 instead of 255 characters, thereby
5241 overflowing its buffer with the terminating zero. As well as fixing this, I
5242 have increased the buffer size to 1024 (and made a note to document this).
5244 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5245 when Exim is waiting for an SMTP response from a remote server, Exim
5246 restarts its select() call on the socket, thereby resetting its timeout.
5247 This is not a problem when such interrupts are rare. Somebody set up a cron
5248 job to run exiwhat every 2 minutes, which is less than the normal select()
5249 timeout (5 or 10 minutes). This meant that the select() timeout never
5250 kicked in because it was always reset. I have fixed this by comparing the
5251 time when an interrupt arrives with the time at the start of the first call
5252 to select(). If more time than the timeout has elapsed, the interrupt is
5253 treated as a timeout.
5255 5. Some internal re-factoring in preparation for the addition of Sieve
5256 extensions (by MH). In particular, the "personal" test is moved to a
5257 separate function, and given an option for scanning Cc: and Bcc: (which is
5258 not set for Exim filters).
5260 6. When Exim created an email address using the login of the caller as the
5261 local part (e.g. when creating a From: or Sender: header line), it was not
5262 quoting the local part when it contained special characters such as @.
5264 7. Installed new OpenBSD configuration files.
5266 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5267 try to make them clearer.
5269 9. Callout options, other than the timeout value, were being ignored when
5270 verifying sender addresses in header lines. For example, when using
5272 verify = header_sender/callout=no_cache
5274 the cache was (incorrectly) being used.
5276 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5277 only the Cygwin environment.
5279 11. When return_path_on_delivery was set as a log selector, if different remote
5280 addresses in the same message used different return paths and parallel
5281 remote delivery occurred, the wrong values would sometimes be logged.
5282 (Whenever a remote delivery process finished, the return path value from
5283 the most recently started remote delivery process was logged.)
5285 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5286 lines when AUTH and/or TLS are in use. This is the "received protocol"
5287 field. Exim used to use "asmtp" for authenticated SMTP, without any
5288 indication (in the protocol name) for TLS use. Now it follows the RFC and
5289 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5290 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5291 names appear in log lines as well as in Received: header lines.
5293 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5294 extensions, and comparison tests, and to fix some bugs.
5296 14. Changes to the "personal" filter test:
5298 (1) The test was buggy in that it was just doing the equivalent of
5299 "contains" tests on header lines. For example, if a user's address was
5300 anne@some.where, the "personal" test would incorrectly be true for
5302 To: susanne@some.where
5304 This test is now done by extracting each address from the header in turn,
5305 and checking the entire address. Other tests that are part of "personal"
5306 are now done using regular expressions (for example, to check local parts
5307 of addresses in From: header lines).
5309 (2) The list of non-personal local parts in From: addresses has been
5310 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5311 taken from the Sieve specification recommendations.
5313 (3) If the message contains any header line starting with "List-" it is
5314 treated as non-personal.
5316 (4) The test for "circular" in the Subject: header line has been removed
5317 because it now seems ill-conceived.
5319 15. Minor typos in src/EDITME comments corrected.
5321 16. Installed latest exipick from John Jetmore.
5323 17. If headers_add on a router specified a text string that was too long for
5324 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5325 of string_sprintf() is now avoided.
5327 18. $message_body_size was not set (it was always zero) when running the DATA
5328 ACL and the local_scan() function.
5330 19. For the "mail" command in an Exim filter, no default was being set for
5331 the once_repeat time, causing a random time value to be used if "once" was
5332 specified. (If the value happened to be <= 0, no repeat happened.) The
5333 default is now 0s, meaning "never repeat". The "vacation" command was OK
5334 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5335 (I found it when inspecting the code).
5337 20. There is now an overall timeout for performing a callout verification. It
5338 defaults to 4 times the callout timeout, which applies to individual SMTP
5339 commands during the callout. The overall timeout applies when there is more
5340 than one host that can be tried. The timeout is checked before trying the
5341 next host. This prevents very long delays if there are a large number of
5342 hosts and all are timing out (e.g. when the network connections are timing
5343 out). The value of the overall timeout can be changed by specifying an
5344 additional sub-option for "callout", called "maxwait". For example:
5346 verify = sender/callout=5s,maxwait=20s
5348 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5349 to the end before writing, but this should make it even safer).
5351 22. Exim was forgetting that it had advertised PIPELINING for the second and
5352 subsequent messages on an SMTP connection. It was also not resetting its
5353 memory on STARTTLS and an internal HELO.
5355 23. When Exim logs an SMTP synchronization error within a session, it now
5356 records whether PIPELINING has been advertised or not.
5358 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5359 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5360 rather than long int.
5362 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5364 26. Added the never_mail option to autoreply.
5370 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5371 crash if the getsockname() call failed; this can happen if a connection is
5372 closed very soon after it is established. The problem was simply in the
5373 order in which certain operations were done, causing Exim to try to write
5374 to the SMTP stream before it had set up the file descriptor. The bug has
5375 been fixed by making things happen in the correct order.
5381 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5382 before the connection was closed, thus losing the rejection response.
5384 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5385 some early Solaris releases, but causes trouble in current releases where
5386 socklen_t is defined.
5388 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5391 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5394 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5395 that releasing the top of it at the end releases what was used for sub-
5396 expansions (unless the block got too big). However, discard this block if
5397 the first thing is a variable or header, so that we can use its block when
5398 it is dynamic (useful for very large $message_headers, for example).
5400 6. Lookups now cache *every* query, not just the most recent. A new, separate
5401 store pool is used for this. It can be recovered when all lookup caches are
5402 flushed. Lookups now release memory at the end of their result strings.
5403 This has involved some general refactoring of the lookup sources.
5405 7. Some code has been added to the store_xxx() functions to reduce the amount
5406 of flapping under certain conditions.
5408 8. log_incoming_interface used to affect only the <= reception log lines. Now
5409 it causes the local interface and port to be added to several more SMTP log
5410 lines, for example "SMTP connection from", and rejection lines.
5412 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5414 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5416 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5417 could be overwritten at the end of the current message (or the start of a
5418 new message if it was set in a HELO ACL). The value is now preserved for
5419 the duration of the SMTP connection.
5421 12. If a transport had a headers_rewrite setting, and a matching header line
5422 contained an unqualified address, that address was qualified, even if it
5423 did not match any rewriting rules. The underlying bug was that the values
5424 of the flags that permit the existence of unqualified sender and recipient
5425 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5426 non-local messages, and by -bnq for local messages) were not being
5427 preserved with the message after it was received.
5429 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5430 "next input=" as part of the text comprising the host identity instead of
5431 the correct text. The code was using the same buffer for two different
5432 strings. However, depending on which order the printing function evaluated
5433 its arguments, the bug did not always show up. Under Linux, for example, my
5434 test suite worked just fine.
5436 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5437 doesn't work with some older versions of Perl. It has been changed to "my",
5438 which in any case is probably the better facility to use.
5440 15. A really picky compiler found some instances of statements for creating
5441 error messages that either had too many or two few arguments for the format
5444 16. The size of the buffer for calls to the DNS resolver has been increased
5445 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5446 for addresses that have a lot of PTR records. This alleviates a problem; it
5447 does not fully solve it.
5449 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5450 buffer now truncates the list and logs the incident, which is the same
5451 action as happens when Exim is looking up a host name and its aliases.
5452 Previously in this situation something unpredictable would happen;
5453 sometimes it was "internal error: store_reset failed".
5455 18. If a server dropped the connection unexpectedly when an Exim client was
5456 using GnuTLS and trying to read a response, the client delivery process
5457 crashed while trying to generate an error log message.
5459 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5460 string, for example:
5462 warn message = H1: something\nH2: something
5464 the text was added as a single header line from Exim's point of view
5465 though it ended up OK in the delivered message. However, searching for the
5466 second and subsequent header lines using $h_h2: did not work. This has been
5467 fixed. Similarly, if a system filter added multiple headers in this way,
5468 the routers could not see them.
5470 20. Expanded the error message when iplsearch is called with an invalid key to
5471 suggest using net-iplsearch in a host list.
5473 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5474 ACLs are no longer actually imposed (and a message to that effect is
5477 22. If a "gecos" field in a passwd entry contained escaped characters, in
5478 particular, if it contained a \" sequence, Exim got it wrong when building
5479 a From: or a Sender: header from that name. A second bug also caused
5480 incorrect handling when an unquoted " was present following a character
5481 that needed quoting.
5483 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5484 was not being matched caselessly.
5486 24. Arranged for all hyphens in the exim.8 source to be escaped with
5489 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5490 properly. Recipient callout cache records were still being keyed to include
5491 the sender, even when use_sender was set false. This led to far more
5492 callouts that were necessary. The sender is no longer included in the key
5493 when use_sender is false.
5495 26. Added "control = submission" modifier to ACLs.
5497 27. Added the ${base62d: operator to decode base 62 numbers.
5499 28. dnsdb lookups can now access SRV records.
5501 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5502 the configuration file.
5504 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5505 (-v) mode. This makes the output for a verbose queue run more intelligible.
5507 31. Added a use_postmaster feature to recipient callouts.
5509 32. Added the $body_zerocount variable, containing the number of binary zero
5510 bytes in the message body.
5512 33. The time of last modification of the "new" subdirectory is now used as the
5513 "mailbox time last read" when there is a quota error for a maildir
5516 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5518 35. Added +ignore_unknown as a special item in host lists.
5520 36. Code for decoding IPv6 addresses in host lists is now included, even if
5521 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5522 address was recognized as an IP address, but was then not correctly decoded
5523 into binary, causing unexpected and incorrect effects when compared with
5530 1. Very minor rewording of debugging text in manualroute to say "list of
5531 hosts" instead of "hostlist".
5533 2. If verify=header_syntax was set, and a header line with an unqualified
5534 address (no domain) and a large number of spaces between the end of the
5535 name and the colon was received, the reception process suffered a buffer
5536 overflow, and (when I tested it) crashed. This was caused by some obsolete
5537 code that should have been removed. The fix is to remove it!
5539 3. When running in the test harness, delay a bit after writing a bounce
5540 message to get a bit more predictability in the log output.
5542 4. Added a call to search_tidyup() just before forking a reception process. In
5543 theory, someone could use a lookup in the expansion of smtp_accept_max_
5544 per_host which, without the tidyup, could leave open a database connection.
5546 5. Added the variables $recipient_data and $sender_data which get set from a
5547 lookup success in an ACL "recipients" or "senders" condition, or a router
5548 "senders" option, similar to $domain_data and $local_part_data.
5550 6. Moved the writing of debug_print from before to after the "senders" test
5553 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5554 problems for message scanning, either using a data ACL, or using
5555 local_scan() because the Received: header was not generated till after they
5556 were called (in order to set the time as the time of reception completion).
5557 I have revised the way this works. The header is now generated after the
5558 body is received, but before the ACL or local_scan() are called. After they
5559 are run, the timestamp in the header is updated.
5565 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5566 before starting a queue runner without re-exec. This happened only when
5567 deliver_drop_privilege was set or when the Exim user was set to root. The
5568 effect of the bug was that timeouts during subsequent deliveries caused
5569 crashes instead of being properly handled. The handler is now left at its
5570 default (and expected) setting.
5572 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5573 message, again when deliver_drop_privilege is set or Exim is run as root.
5574 The bug described in (1) was not present in this case, but the tidying up
5575 of the other signals was missing. I have made the two cases consistent.
5577 3. The ignore_target_hosts setting on a manualroute router was being ignored
5578 for hosts that were looked up using the /MX notation.
5580 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5583 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5584 operated on the sender address. After changing the $sender_address to <>
5585 for the sender address verify, Exim was re-instated it as the original
5586 (before rewriting) address, but remembering that it had rewritten it, so it
5587 wasn't rewriting it again. This bug also had the effect of breaking the
5588 sender address verification caching when the sender address was rewritten.
5590 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5591 This has been changed so that if the ip literal address matches
5592 ignore_target_hosts, the router declines.
5594 7. Added expansion conditions match_domain, match_address, and match_local_
5595 part (NOT match_host).
5597 8. The placeholder for the Received: header didn't have a length field set.
5599 9. Added code to Exim itself and to exim_lock to test for a specific race
5600 condition that could lead to file corruption when using MBX delivery. The
5601 issue is with the lockfile that is created in /tmp. If this file is removed
5602 after a process has opened it but before that process has acquired a lock,
5603 there is the potential for a second process to recreate the file and also
5604 acquire a lock. This could lead to two Exim processes writing to the file
5605 at the same time. The added code performs the same test as UW imapd; it
5606 checks after acquiring the lock that its file descriptor still refers to
5607 the same named file.
5609 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5610 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5613 11. Added the smtp_active_hostname option. If used, this will typically be made
5614 to depend on the incoming interface address. Because $interface_address is
5615 not set up until the daemon has forked a reception process, error responses
5616 that can happen earlier (such as "too many connections") no longer contain
5619 12. If an expansion in a condition on a "warn" statement fails because a lookup
5620 defers, the "warn" statement is abandoned, and the next ACL statement is
5621 processed. Previously this caused the whole ACL to be aborted.
5623 13. Added the iplsearch lookup type.
5625 14. Added ident_timeout as a log selector.
5627 15. Added tls_certificate_verified as a log selector.
5629 16. Added a global option tls_require_ciphers (compare the smtp transport
5630 option of the same name). This controls incoming TLS connections.
5632 17. I finally figured out how to make tls_require_ciphers do a similar thing
5633 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5634 before starting the TLS session.
5636 18. Tabs are now shown as \t in -bP output.
5638 19. If the log selector return_path_on_delivery was set, Exim crashed when
5639 bouncing a message because it had too many Received: header lines.
5641 20. If two routers both had headers_remove settings, and the first one included
5642 a superfluous trailing colon, the final name in the first list and the
5643 first name in the second list were incorrectly joined into one item (with a
5644 colon in the middle).
5650 1. Added -C and -D options to the exinext utility, mainly to make it easier
5651 to include in the automated testing, but these could be helpful when
5652 multiple configurations are in use.
5654 2. The exinext utility was not formatting the output nicely when there was
5655 an alternate port involved in the retry record key, nor when there was a
5656 message id as well (for retries that were specific to a specific message
5657 and a specific host). It was also confused by IPv6 addresses, because of
5658 the additional colons they contain. I have fixed the IPv4 problem, and
5659 patched it up to do a reasonable job for IPv6.
5661 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5662 delivery, the log line now contains "pipelined" if PIPELINING was used.
5664 4. An SMTP transport process used to panic and die if the bind() call to set
5665 an explicit outgoing interface failed. This has been changed; it is now
5666 treated in the same way as a connect() failure.
5668 5. A reference to $sender_host_name in the part of a conditional expansion
5669 that was being skipped was still causing a DNS lookup. This no longer
5672 6. The def: expansion condition was not recognizing references to header lines
5673 that used bh_ and bheader_.
5675 7. Added the _cache feature to named lists.
5677 8. The code for checking quota_filecount in the appendfile transport was
5678 allowing one more file than it should have been.
5680 9. For compatibility with Sendmail, the command line option
5688 and sets the incoming protocol and host name (for trusted callers). The
5689 host name and its colon can be omitted when only the protocol is to be set.
5690 Note the Exim already has two private options, -pd and -ps, that refer to
5691 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5692 "s", but I don't think that's a major issue.
5694 10. A number of refactoring changes to the code, none of which should affect
5697 (a) The number of logging options was getting close to filling up the
5698 32-bit word that was used as a bit map. I have split them into two classes:
5699 those that are passed in the argument to log_write(), and those that are
5700 only ever tested independently outside of that function. These are now in
5701 separate 32-bit words, so there is plenty of room for expansion again.
5702 There is no change in the user interface or the logging behaviour.
5704 (b) When building, for example, log lines, the code previously used a
5705 macro that called string_cat() twice, in order to add two strings. This is
5706 not really sufficiently general. Furthermore, there was one instance where
5707 it was actually wrong because one of the argument was used twice, and in
5708 one call a function was used. (As it happened, calling the function twice
5709 did not affect the overall behaviour.) The macro has been replaced by a
5710 function that can join an arbitrary number of extra strings onto a growing
5713 (c) The code for expansion conditions now uses a table and a binary chop
5714 instead of a serial search (which was left over from when there were very
5715 few conditions). Also, it now recognizes conditions like "pam" even when
5716 the relevant support is not compiled in: a suitably worded error message is
5717 given if an attempt is made to use such a condition.
5719 11. Added ${time_interval:xxxxx}.
5721 12. A bug was causing one of the ddress fields not to be passed back correctly
5722 from remote delivery subprocesses. The field in question was not being
5723 subsequently used, so this caused to problems in practice.
5725 13. Added new log selectors queue_time and deliver_time.
5727 14. Might have fixed a bug in maildirsizefile handling that threw up
5728 "unexpected character" debug warnings, and recalculated the data
5729 unnecessarily. In any case, I expanded the warning message to give more
5732 15. Added the message "Restricted characters in address" to the statements in
5733 the default ACL that block characters like @ and % in local parts.
5735 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5736 Three changes have been made:
5738 (a) There was a serious bug; a negative response to MAIL caused the whole
5739 recipient domain to be cached as invalid, thereby blocking all messages
5740 to all local parts at the same domain, from all senders. This bug has
5741 been fixed. The domain is no longer cached after a negative response to
5742 MAIL if the sender used is not empty.
5744 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5747 (c) A new callout option, "use_sender" has been added for people who want
5748 the modified behaviour.
5754 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5757 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5758 indeed breaks things for older releases.
5760 3. Added additional logging to the case where there is a problem reading data
5761 from a filter that is running in a subprocess using a pipe, in order to
5762 try to track down a specific problem.
5764 4. Testing facility fudge: when running in the test harness and attempting
5765 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5766 getting "No route to host". Convert this to a timeout.
5768 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5771 6. Some OS don't have socklen_t but use size_t instead. This affects the
5772 fifth argument of getsockopt() amongst other things. This is now
5773 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5774 can be set for individual OS. I have set it for SunOS5, OSF1, and
5775 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5776 some earlier ones do not.
5778 7. Change 4.30/15 was not doing the test caselessly.
5780 8. The standard form for an IPv6 address literal was being rejected by address
5781 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5782 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5783 this, as well as the form without the "IPv6" on the front (but only when
5784 address literals are enabled, of course).
5786 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5788 10. Exim crashed if a message with an empty sender address specified by -f
5789 encountered a router with an errors_to setting. This could be provoked only
5790 by a command such as
5794 where an empty string was supplied; "<>" did not hit this bug.
5796 11. Installed PCRE release 4.5.
5798 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5799 remained set. It is now erased.
5801 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5802 times from message ids (which are base 36 rather than the normal 62).
5804 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5805 were being counted as actual protocol errors, and logged if the log
5806 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5807 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5808 and DATA following a set of rejected RCPTs do not count as protocol errors.
5809 In other words, Exim assumes they were pipelined, though this may not
5810 actually be the case. Of course, in all cases the client gets an
5811 appropriate error code.
5813 15. If a lookup fails in an ACL condition, a message about the failure may
5814 be available; it is used if testing the ACL cannot continue, because most
5815 such messages specify what the cause of the deferral is. However, some
5816 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5817 that caused an old message to be retained and used if a later statement
5818 caused a defer, replacing the real cause of the deferral.
5820 16. If an IP address had so many PTR records that the DNS lookup buffer
5821 was not large enough to hold them, Exim could crash while trying to process
5822 the truncated data. It now detects and logs this case.
5824 17. Further to 4.21/58, another change has been made: if (and only if) the
5825 first line of a message (the first header line) ends with CRLF, a bare LF
5826 in a subsequent header line has a space inserted after it, so as not to
5827 terminate the header.
5829 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5830 unnecessarily, used atoi() instead of strtol(), and didn't check the
5831 termination when getting file sizes from file names by regex.
5833 19. Completely re-implemented the support for maildirsize files, in the light
5834 of a number of problems with the previous contributed implementation
5835 (4.30/29). In particular:
5837 . If the quota is zero, the maildirsize file is maintained, but no quota is
5840 . If the maildir directory does not exist, it is created before any attempt
5841 to write a maildirsize file.
5843 . The quota value in the file is just a cache; if the quota is changed in
5844 the transport, the new value overrides.
5846 . A regular expression is available for excluding directories from the
5849 20. The autoreply transport checks the characters in options that define the
5850 message's headers; it allows continued headers, but it was checking with
5851 isspace() after an embedded newline instead of explicitly looking for a
5854 21. If all the "regular" hosts to which an address was routed had passed their
5855 expiry times, and had not reached their retry times, the address was
5856 bounced, even if fallback hosts were defined. Now Exim should go on to try
5859 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5860 equivalent code in the SMTP transport. Some hosts send humungous responses
5861 to HELO/EHLO, more than 1024 it seems.
5863 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5864 gives compiler warnings in some environments. I've now done it "properly",
5867 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5868 (because of problems with the built-in one) was declared to return uschar *
5869 instead of char *, causing compiler failure.
5871 25. Fixed a file descriptor leak when processing alias/forward files.
5873 26. Fixed a minor format string issue in dbfn.c.
5875 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5877 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5878 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5881 29. When a sender address is verified, it is cached, to save repeating the test
5882 when there is more than one recipient in a message. However, when the
5883 verification involves a callout, it is possible for different callout
5884 options to be set for different recipients. It is too complicated to keep
5885 track of this in the cache, so now Exim always runs a verification when a
5886 callout is required, relying on the callout cache for the optimization.
5887 The overhead is duplication of the address routing, but this should not be
5890 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5891 to be verified with callout=postmaster, and the main callout worked but the
5892 postmaster check failed, the verification correctly failed. However, if a
5893 subsequent RCPT command asked for sender verification *without* the
5894 postmaster check, incorrect caching caused this verification also to fail,
5897 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5898 it was not caching the DNS options (qualify_single, search_parents) that
5899 were used when the lookup failed. A subsequent lookup with different
5900 options therefore always gave the same answer, though there were cases
5901 where it should not have. (Example: a "domains = !$mx_any" option on a
5902 dnslookup router: the "domains" option is always processed without any
5903 widening, but the router might have qualify_single set.) Now Exim uses the
5904 cached value only when the same options are set.
5906 32. Added John Jetmore's "exipick" utility to the distribution.
5908 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5909 than a timeout (e.g. a certificate is required, and is not provided), an
5910 Exim server now closes the connection immediately. Previously it waited for
5911 the client to close - but if the client is SSL, it seems that they each
5912 wait for each other, leading to a delay before one of them times out.
5914 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5915 maintained 0.8.x compatibility because I don't think many are using it, and
5916 it is clearly obsolete.
5918 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5921 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5922 client certificate was expired. A simple patch fixes this, though I don't
5923 understand the full logic of why the verify callback is called multiple
5926 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5927 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5928 which causes problems with some clients (such as the Certicom SSL Plus
5929 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5930 disables the coutermeasure allowing Eudora to connect."
5932 38. Exim was not checking that a write() to a log file succeeded. This could
5933 lead to Bad Things if a log got too big, in particular if it hit a file
5934 size limit. Exim now panics and dies if it cannot write to a log file, just
5935 as it does if it cannot open a log file.
5937 39. Modified OS/Makefile-Linux so that it now contains
5939 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5941 The two -D definitions ensure that Exim is compiled with large file
5942 support, which makes it possible to handle log files that are bigger than
5945 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5946 instance) a domain was checked against a named list that involved a lookup,
5947 causing $domain_data to be set, then another domain was checked against the
5948 same list, then the first domain was re-checked, the value of $domain_data
5949 after the final check could be wrong. In particular, if the second check
5950 failed, it could be set empty. This bug probably also applied to
5953 41. The strip_trailing_dot option was not being applied to the address given
5954 with the -f command-line option.
5956 42. The code for reading a message's header from the spool was incrementing
5957 $received_count, but never initializing it. This meant that the value was
5958 incorrect (doubled) while delivering a message in the same process in which
5959 it was received. In the most common configuration of Exim, this never
5960 happens - a fresh exec is done - but it can happen when
5961 deliver_drop_privilege is set.
5963 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5964 it now includes up to 150 characters of the unexpected data in the log
5967 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5968 and building data strings. The size of both of these buffers was 10 000
5969 bytes - far larger than anybody would *ever* want, thought I. Needless to
5970 say, somebody hit the limit. I have increased the maximum line length to
5971 20 000 and the maximum data length of concatenated lines to 100 000. I have
5972 also fixed two bugs, because there was no checking on these buffers. Tsk,
5973 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5974 buffer is too small.
5976 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5977 lsearch lookups. Now it does.
5979 46. When parsing a route_list item in a manualroute router, a fixed-length
5980 buffer was used for the list of hosts. I made this 1024 bytes long,
5981 thinking that nobody would ever have a list of hosts that long. Wrong.
5982 Somebody had a whole pile of complicated expansion conditions, and the
5983 string was silently truncated, leading to an expansion error. It turns out
5984 that it is easier to change to an unlimited length (owing to other changes
5985 that have happened since this code was originally written) than to build
5986 structure for giving a limitation error. The length of the item that
5987 expands into the list of hosts is now unlimited.
5989 47. The lsearch lookup could not handle data where the length of text line was
5990 more than 4095 characters. Such lines were truncated, leading to shortened
5991 data being returned. It should now handle lines of any length.
5993 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5994 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5997 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5998 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5999 no longer happens in comment lines. A list of replacements is now placed
6000 at the head of all of the source files, except those whose only change is
6001 to replace PERL_COMMAND in the very first #! line.
6003 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6004 messages on the queue, with a bottom-up merge sort, using code contributed
6005 by Michael Haardt. This should make operations like -bp somewhat faster on
6006 large queues. It won't affect queue runners, except when queue_run_in_order
6009 51. Installed eximstats 1.31 in the distribution.
6011 52. Added support for SRV lookups to the dnslookup router.
6013 53. If an ACL referred to $message_body or $message_body_end, the value was not
6014 reset for any messages that followed in the same SMTP session.
6016 54. The store-handling optimization for building very long strings was not
6017 differentiating between the different store pools. I don't think this
6018 actually made any difference in practice, but I've tidied it.
6020 55. While running the routers to verify a sender address, $sender_address
6021 was still set to the sender address. This is wrong, because when routing to
6022 send a bounce to the sender, it would be empty. Therefore, I have changed
6023 it so that, while verifying a sender address, $sender_address is set to <>.
6024 (There is no change to what happens when verifying a recipient address.)
6026 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6027 target A or AAAA records (if not already returned) without resetting the
6028 qualify_single or search_parents options of the DNS resolver. These are
6029 inappropriate in this case because the targets of MX and SRV records must
6030 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6031 target that, when qualified, matched something in the local domain. These
6032 two options are now turned off when doing these lookups.
6034 57. It seems that at least some releases of Reiserfs (which does not have the
6035 concept of a fixed number of inodes) returns zero and not -1 for the
6036 number of available inodes. This interacted badly with check_spool_inodes,
6037 which assumed that -1 was the "no such thing" setting. What I have done is
6038 to check that the total number of inodes is greater than zero before doing
6039 the test of how many are available.
6041 58. When a "warn" ACL statement has a log_message modifier, the message is
6042 remembered, and not repeated. This is to avoid a lot of repetition when a
6043 message has many recipients that cause the same warning to be written.
6044 However, Exim was preserving the list of already written lines for an
6045 entire SMTP session, which doesn't seem right. The memory is now reset if a
6046 new message is started.
6048 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6049 debugging output unless log_rewrite was also set.
6051 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6052 of (int)(handle) when we know that handle contains (void *)(-1).
6054 61. The Exim daemon panic-logs an error return when it closes the incoming
6055 connection. However "connection reset by peer" seems to be common, and
6056 isn't really an error worthy of noting specially, so that particular error
6059 62. When Exim is trying to find all the local interfaces, it used to panic and
6060 die if the ioctl to get the interface flags failed. However, it seems that
6061 on at least one OS (Solaris 9) it is possible to have an interface that is
6062 included in the list of interfaces, but for which you get a failure error
6063 for this call. This happens when the interface is not "plumbed" into a
6064 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6065 failure of the "get flags" call assumes that the interface is down.
6067 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6068 makes life easier for people who are doing arithmetic on fields extracted
6069 from dates, where you often get leading zeros that should not be
6070 interpreted as octal.
6072 64. Added qualify_domain to the redirect router, to override the global
6075 65. If a pathologically long header line contained very many addresses (the
6076 report of this problem mentioned 10 000) and each of them was rewritten,
6077 Exim could use up a very large amount of memory. (It kept on making new
6078 copies of the header line as it rewrote, and never released the old ones.)
6079 At the expense of a bit more processing, the header rewriting function has
6080 been changed so that it no longer eats memory in this way.
6082 66. The generation of the Received: header has been moved from the time that a
6083 message starts to be received, to the time that it finishes. The timestamp
6084 in the Received: header should now be very close to that of the <= log
6085 line. There are two side-effects of this change:
6087 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6088 logged header lines no longer include the local Received: line, because
6089 it has not yet been created. The same applies to a copy of the message
6090 that is returned to a non-SMTP sender when a message is rejected.
6092 (b) When a filter file is tested using -bf, no additional Received: header
6093 is added to the test message. After some thought, I decided that this
6096 This change does not affect the value of $received_for. It is still set
6097 after address rewriting, but before local_scan() is called.
6099 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6101 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6102 gave an unhelpful panic error message, and a defer error. I have managed to
6103 change this behaviour so that it now rejects any supplied certificate,
6104 which seems right, as the list of acceptable certificates is empty.
6106 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6107 gave an unhelpful defer error. I have not managed to make this reject any
6108 supplied certificates, but the error message it gives is "no certificate
6109 supplied", which is not helpful.
6111 70. exigrep's output now also includes lines that are not associated with any
6112 message, but which match the given pattern. Implemented by a patch from
6113 Martin Sluka, which also tidied up the Perl a bit.
6115 71. Recipient callout verification, like sender verification, was using <> in
6116 the MAIL FROM command. This isn't really the right thing, since the actual
6117 sender may affect whether the remote host accepts the recipient or not. I
6118 have changed it to use the actual sender in the callout; this means that
6119 the cache record is now keyed on a recipient/sender pair, not just the
6120 recipient address. There doesn't seem to be a real danger of callout loops,
6121 since a callout by the remote host to check the sender would use <>.
6122 [SEE ABOVE: changed after hitting problems.]
6124 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6125 temporary errors. However, in the case of such a code being given after
6126 the end of a data transmission (i.e. after ".") Exim was failing to write
6127 a retry record for the message. (Yes, there was some broken host that was
6128 actually sending 8xx at this point.)
6130 73. An unknown lookup type in a host list could cause Exim to panic-die when
6131 the list was checked. (An example that provoked this was putting <; in the
6132 middle of a list instead of at the start.) If this happened during a DATA
6133 ACL check, a -D file could be left lying around. This kind of configuration
6134 error no longer causes Exim to die; instead it causes a defer error. The
6135 incident is still logged to the main and panic logs.
6137 74. Buglet left over from Exim 3 conversion. The message "too many messages
6138 in one connection" was written to the rejectlog but not the mainlog, except
6139 when address rewriting (yes!) was being logged.
6141 75. Added write_rejectlog option.
6143 76. When a system filter was run not as root (that is, when system_filter_user
6144 was set), the values of the $n variables were not being returned to the
6145 main process; thus, they were not subsequently available in the $sn
6148 77. Added +return_path_on_delivery log selector.
6150 78. A connection timeout was being treated differently from recipients deferred
6151 when testing hosts_max_try with a message that was older than the host's
6152 retry timeout. (The host should not be counted, thus allowing all hosts to
6153 be tried at least once before bouncing.) This may have been the cause of an
6154 occasionally reported bug whereby a message would remain on the queue
6155 longer than the retry timeout, but would be bounced if a delivery was
6156 forced. I say "may" because I never totally pinned down the problem;
6157 setting up timeout/retry tests is difficult. See also the next item.
6159 79. The ultimate address timeout was not being applied to errors that involved
6160 a combination of host plus message (for example, a timeout on a MAIL
6161 command). When an address resolved to a number of possible hosts, and they
6162 were not all tried for each delivery (e.g. because of hosts_max_try), a
6163 message could remain on the queue longer than the retry timeout.
6165 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6168 81. Fixed an obscure SMTP outgoing bug which required at least the following
6169 conditions: (a) there was another message waiting for the same server;
6170 (b) the server returned 5xx to all RCPT commands in the first message so
6171 that the message was not completed; (c) the server dropped the connection
6172 or gave a negative response to the RSET that Exim sends to abort the
6173 transaction. The observed case was a dropped connection after DATA that had
6174 been sent in pipelining mode. That is, the server had advertised PIPELINING
6175 but was not implementing it correctly. The effect of the bug was incorrect
6176 behaviour, such as trying another host, and this could lead to a crash.
6182 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6183 and daemon.c were passed as pointers to ints; they should have been
6184 pointers to socklen_t variables (which are typically unsigned ints).
6186 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6189 3. Fixed a really odd bug that affected only the testing scheme; patching a
6190 certain fixed string in the binary changed the value of another string that
6191 happened to be identical to the end of the original first string.
6193 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6194 name", it returns that address as the IP address. On some operating
6195 systems (e.g. Solaris), it also passes back the IP address string as the
6196 "host name". However, on others (e.g. Linux), it passes back an empty
6197 string. Exim wasn't checking for this, and was changing the host name to an
6198 empty string, assuming it had been canonicalized.
6200 5. Although rare, it is permitted to have more than one PTR record for a given
6201 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6202 all the names associated with an address, because they do in Solaris.
6203 However, it seems that they do not in Linux for data that comes from the
6204 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6205 I found this out when I moved to a new Linux workstation and tried to run
6206 the Exim test suite.
6208 To get round this problem I have changed the code so that it now does its
6209 own call to the DNS to look up PTR records when searching for a host name.
6210 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6211 addresses that are only in /etc/hosts are still found.
6213 This behaviour is, however, controlled by an option called host_lookup_
6214 order, which defaults to "bydns:byaddr". If people want to use the other
6215 order, or indeed, just use one or the other means of lookup, they can
6216 specify it in this variable.
6218 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6219 some operating systems, this comes back from gethostbyaddr() as an empty
6220 string, and this is what Exim used to test for. However, it seems that in
6221 other systems, "." is yielded. Exim now tests for this case too.
6223 7. The values of check_spool_space and check_log_space are now held internally
6224 as a number of kilobytes instead of an absolute number of bytes. If a
6225 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6226 kilobyte. This means that much larger values can be stored.
6228 8. Exim monitor: an attempt to get the action menu when not actually pointing
6229 at a message produces an empty menu entitled "No message selected". This
6230 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6231 no entries in it ("Shell widget menu has zero width and/or height"). So I
6232 have added a single, blank menu entry in this case.
6234 9. Added ${quote_local_part.
6236 10. MIME decoding is now applied to the contents of Subject: header lines when
6239 11. Now that a reference to $sender_host_address automatically causes a reverse
6240 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6241 host lookup before query-style lookups in lists that might use this
6242 variable. This has therefore been abolished, and the "net-" prefix is no
6243 longer necessary for query-style lookups.
6245 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6246 have been a typo for LFLAGS, so it has been changed.
6248 13. The install script calls Exim with "-C /dev/null" in order to find the
6249 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6250 to be output. However, since Exim outputs its version number before the
6251 error, it didn't break the script. It just looked ugly. I fixed this by
6252 always allowing "-C /dev/null" if the caller is root.
6254 14. Ignore overlarge ACL variable number when reading spool file - insurance
6255 against a later release with more variables having written the file.
6257 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6258 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6259 this, as well as the form without the "IPv6" on the front.
6261 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6262 OS/Makefile-Darwin file.
6264 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6265 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6267 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6268 variable. However, some OS use ints for inodes. I've added cast to long int
6269 to get rid of the compiler warning.
6271 19. I had forgotten to lock out "/../" in configuration file names when
6272 ALT_CONFIG_PREFIX was set.
6274 20. Routers used for verification do not need to specify transports. However,
6275 if such a router generated a host list, and callout was configured, Exim
6276 crashed, because it could not find a port number from the (non-existent)
6277 transport. It now assumes port 25 in this circumstance.
6279 21. Added the -t option to exigrep.
6281 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6282 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6283 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6284 them is set, LOOKUP_LSEARCH is forced.
6286 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6288 24. Added sender and host information to the "rejected by local_scan()" log
6289 line; previously there was no indication of these.
6291 25. Added .include_if_exists.
6293 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6294 Linux. It turns out that not all file systems support this. Apparently some
6295 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6296 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6297 not supported on the file descriptor, is now ignored when Exim is trying to
6298 sync a directory. This applies only to Linux.
6300 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6302 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6303 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6306 29. Added support for maildirsize files from supplied patch (modified a bit).
6308 30. The use of :fail: followed by an empty string could lead Exim to respond to
6309 sender verification failures with (e.g.):
6311 550 Verification failed for <xxx>
6312 550 Sender verify failed
6314 where the first response line was missing the '-' that indicates it is not
6315 the final line of the response.
6317 31. The loop for finding the name of the user that called Exim had a hardwired
6318 limit of 10; it now uses the value of finduser_retries, which is used for
6319 all other user lookups.
6321 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6324 33. Exim was neglecting to zero errno before one call of strtol() when
6325 expanding a string and expecting an integer value. On some systems this
6326 resulted in spurious "integer overflow" errors. Also, it was casting the
6327 result into an int without checking.
6329 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6330 did not work. The code looks as if it has *never* worked, though it appears
6331 to have been documented since at least release 1.62. I have made it work.
6333 35. The "timeout_DNS" error in retry rules, also documented since at least
6334 1.62, also never worked. As it isn't clear exactly what this means, and
6335 clearly it isn't a major issue, I have abolished the feature by treating it
6336 as "timeout", and writing a warning to the main and panic logs.
6338 36. The display of retry rules for -brt wasn't always showing the error code
6341 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6342 timeout_connect_A, timeout_connect_MX.
6344 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6345 to the empty sender.
6347 39. The daemon was not analysing the content of -oX till after it had closed
6348 stderr and disconnected from the controlling terminal. This meant that any
6349 syntax errors were only noted on the panic log, and the return code from
6350 the command was 0. By re-arranging the code a little, I've made the
6351 decoding happen first, so such errors now appear on stderr, and the return
6352 code is 1. However, the actual setting up of the sockets still happens in
6353 the disconnected process, so errors there are still only recorded on the
6356 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6357 connections (as happens on some IP stacks) was logged at start up time as
6358 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6359 it from "IPv6 and IPv4", which means that two separate sockets are being
6362 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6363 says whether AF_INET or AF_INET6 was passed as an argument.
6365 42. Exiwhat output was messed up when time zones were included in log
6368 43. Exiwhat now gives more information about the daemon's listening ports,
6369 and whether -tls-on-connect was used.
6371 44. The "port" option of the smtp transport is now expanded.
6373 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6374 silently ignored. Now an error message is written to the main and panic
6377 46. There's a new ACL modifier called "logwrite" which writes to a log file
6378 as soon as it is encountered.
6380 47. Added $local_user_uid and $local_user_gid at routing time.
6382 48. Exim crashed when trying to verify a sender address that was being
6385 49. Exim was recognizing only a space character after ".include". It now also
6386 recognizes a tab character.
6388 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6389 extracting the relevant information from the specification. The man page no
6390 longer contains scrambled data for the -d option, and I've added a section
6391 at the front about calling Exim under different names.
6393 51. Added "extra_headers" argument to the "mail" command in filter files.
6395 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6398 53. Installed eximstats 1.29.
6400 54. Added transport_filter_timeout as a generic transport option.
6402 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6403 Cc: header lines. This was required by RFC 822, but it not required by RFC
6406 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6407 incoming messages that did not have them. Now it does so only if the
6408 message originates locally, that is, if there is no associated remote host
6409 address. When Resent- header lines are present, this applies to the Resent-
6410 lines rather than the non-Resent- lines.
6412 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6413 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6415 58. Messages for configuration errors now include the name of the main
6416 configuration file - useful now that there may be more than one file in a
6417 list (.included file names were always shown).
6419 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6420 for those rare installations that do not start the daemon as root or run it
6421 setuid root. I've cut out the call to initgroups() if the daemon is not
6424 60. The Exim user and group can now be bound into the binary as text strings
6425 that are looked up at the start of Exim's processing.
6427 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6429 62. Added $mailstore_basename variable.
6431 63. Installed patch to sieve.c from Michael Haardt.
6433 64. When Exim failed to open the panic log after failing to open the main log,
6434 the original message it was trying to log was written to stderr and debug
6435 output, but if they were not available (the usual case in production), it
6436 was lost. Now it is written to syslog before the two lines that record the
6437 failures to open the logs.
6439 65. Users' Exim filters run in subprocesses under the user's uid. It is
6440 possible for a "deliver" command or an alias in a "personal" command to
6441 provoke an address rewrite. If logging of address rewriting is configured,
6442 this fails because the process is not running as root or exim. There may be
6443 a better way of dealing with this, but for the moment (because 4.30 needs
6444 to be released), I have disabled address rewrite logging when running a
6445 filter in a non-root, non-exim process.
6451 1. The buildconfig auxiliary program wasn't quoting the value set for
6452 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6453 not defined. This bug was masked in 4.22 by the effect that was fixed in
6456 2. Some messages that were rejected after a message id was allocated were
6457 shown as "incomplete" by exigrep. It no longer does this for messages that
6458 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6460 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6461 have allow_domain_literals set, the ID did not get logged in the <= line.
6462 Domain literals are now always recognized in Message-ID: header lines.
6464 4. The first argument for a ${extract expansion item is the key name or field
6465 number. Leading and trailing spaces in this item were not being ignored,
6466 causing some misleading effects.
6468 5. When deliver_drop_privilege was set, single queue runner processes started
6469 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6470 same command in the process it spins off) were not dropping privilege.
6472 6. When the daemon running as "exim" started a queue runner, it always
6473 re-executed Exim in the spun-off process. This is a waste of effort when
6474 deliver_drop_privilege is set. The new process now just calls the
6475 queue-runner function directly.
6481 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6484 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6485 ignored. Though the use of -oP was forcing the writing of a pid file, it
6486 was always written to the default place.
6488 3. If the message "no IP address found for host xxxx" is generated during
6489 incoming verification, it is now followed by identification of the incoming
6490 connection (so you can more easily find what provoked it).
6492 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6494 5. Added some features to "harden" Exim a bit more against certain attacks:
6496 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6497 be put in Local/Makefile. This is like the never_users runtime option,
6498 but it cannot be overridden. The default setting is "root".
6500 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6501 prefix string with which any file named in a -C command line option
6504 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6505 is retained for -C and -D only if the caller of Exim is root. Without
6506 it, the exim user may also use -C and -D and retain privilege.
6508 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6509 command line option is disabled.
6511 6. Macro names set by the -D option must start with an upper case letter, just
6512 like macro names defined in the configuration file.
6514 7. Added "dereference=" facility to LDAP.
6516 8. Two instances of the typo "uknown" in the source files are fixed.
6518 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6519 the Configure-Makefile script screwed up while processing it.
6521 10. Incorporated PCRE 4.4.
6523 11. The SMTP synchronization check was not operating right at the start of an
6524 SMTP session. For example, it could not catch a HELO sent before the client
6525 waited for the greeting. There is now a check for outstanding input at the
6526 point when the greeting is written. Because of the duplex, asynchronous
6527 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6528 way, but not yet received, when the check is performed.
6530 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6531 on TCP/IP sockets, because this apparently causes some broken clients to
6534 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6535 unchanged) from the Cygwin maintainer.
6537 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6538 was supported instead of testing for mbx. Effectively a typo.
6540 15. The spa authenticator server code was not checking that the input it
6541 received was valid base64.
6543 16. The debug output line for the "set" modifier in ACLs was not showing the
6544 name of the variable that was being set.
6546 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6548 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6549 lookup on its own. Something else was needed to trigger the lookup. For
6550 example, a match in host_lookup or the need for a host name in a host list.
6551 Now, if $sender_host_name is referenced and the host name has not yet been
6552 looked up, a lookup is performed. If the lookup fails, the variable remains
6553 empty, and $host_lookup_failed is set to "1".
6555 19. Added "eqi" as a case-independent comparison operator.
6557 20. The saslauthd authentication condition could segfault if neither service
6558 nor realm was specified.
6560 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6561 error message that was logged was misleading, and incoming SMTP
6562 connections were dropped. The message is now more accurate, and temporary
6563 errors are given to SMTP connections.
6565 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6566 (except RSET) with a 421 error, until QUIT is received. However, it was
6567 failing to send a response to QUIT.
6569 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6570 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6571 ACL is now run after the helo_try_verify_hosts code.
6573 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6574 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6575 case-independent, so other case variants are also recognized). Apparently
6576 some systems use these upper case variants.
6578 25. If more than two messages were waiting for the same host, and a transport
6579 filter was specified for the transport, Exim sent two messages over the
6580 same TCP/IP connection, and then failed with "socket operation on non-
6581 socket" when it tried to send the third.
6583 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6585 27. The extern definition of crypt16() in expand.c was not being excluded when
6586 the OS had its own crypt16() function.
6588 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6589 as a preferred synonym for return_size_limit, both as an option and as an
6592 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6594 30. Changed the default configuration ACL to relax the local part checking rule
6595 for addresses that are not in any local domains. For these addresses,
6596 slashes and pipe symbols are allowed within local parts, but the sequence
6597 /../ is explicitly forbidden.
6599 31. SPA server authentication was not clearing the challenge buffer before
6602 32. log_message in a "warn" ACL statement was writing to the reject log as
6603 well as to the main log, which contradicts the documentation and doesn't
6604 seem right (because no rejection is happening). So I have stopped it.
6606 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6607 However, I am unable to do any testing of this.
6609 34. Fixed an infelicity in the appendfile transport. When checking directories
6610 for a mailbox, to see if any needed to be created, it was accidentally
6611 using path names with one or more superfluous leading slashes; tracing
6612 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6614 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6615 contents are added to the log line that is written for every discarded
6616 recipient. (Previously a log_message setting was ignored.)
6618 36. The ${quote: operator now quotes the string if it is empty.
6620 37. The install script runs exim in order to find its version number. If for
6621 some reason other than non-existence or emptiness, which it checks, it
6622 could not run './exim', it was installing it with an empty version number,
6623 i.e. as "exim-". This error state is now caught, and the installation is
6626 38. An argument was missing from the function that creates an error message
6627 when Exim fails to connect to the socket for saslauthd authentication.
6628 This could cause Exim to crash, or give a corrupted message.
6630 39. Added isip, isip4, and isip6 to ${if conditions.
6632 40. The ACL variables $acl_xx are now saved with the message, and can be
6633 accessed later in routers, transports, and filters.
6635 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6636 strings in the file are not string-expanded.
6638 42. If a MAIL command specified a SIZE value that was too large to fit into an
6639 int variable, the check against message_size_limit failed. Such values are
6640 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6641 day this will have to be increased, but I don't think I want to be around
6642 when emails are that large.
6649 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6650 iconv() is not standard in FreeBSD.
6652 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6653 IPv6 enabled. The observed symptom was a segmentation fault on return from
6654 the function os_common_find_running_interfaces() in src/os.c.
6656 3. In the check_special_case() function in daemon.c I had used "errno" as an
6657 argument name, which causes warnings on some systems. This was basically a
6658 typo, since it was named "eno" in the comments!
6660 4. The code that waits for the clock to tick (at a resolution of some fraction
6661 of a second) so as to ensure message-id uniqueness was always waiting for
6662 at least one whole tick, when it could have waited for less. [This is
6663 almost certainly not relevant at current processor speeds, where it is
6664 unlikely to ever wait at all. But we try to future-proof.]
6666 5. The function that sleeps for a time interval that includes fractions of a
6667 second contained a race. It did not block SIGALRM between setting the
6668 timer, and suspending (a couple of lines later). If the interval was short
6669 and the sigsuspend() was delayed until after it had expired, the suspension
6670 never ended. On busy systems this could lead to processes getting stuck for
6673 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6674 process, before it forks any delivery processes. The open lookup caching
6675 mechanism meant that the open file or database connection was passed into
6676 the delivery process. The problem was that delivery processes always tidy
6677 up cached lookup data. This could cause a problem for the next delivery
6678 process started by the queue runner, because the external queue runner
6679 process does not know about the closure. So the next delivery process
6680 still has data in the lookup cache. In the case of a file lookup, there was
6681 no problem because closing a file descriptor in a subprocess doesn't affect
6682 the parent. However, if the lookup was caching a connection to a database,
6683 the connection was closed, and the second delivery process was likely to
6684 see errors such as "PGSQL: query failed: server closed the connection
6685 unexpectedly". The problem has been fixed by closing all cached lookups
6686 in a queue runner before running a delivery process.
6688 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6689 seem to have the "const" qualifier which it has on other OS. I've
6692 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6693 *of the same type* (client or server) with the same public name that an
6694 error should be diagnosed.
6696 9. When Exim looked up a host name for an IP address, but failed to find the
6697 original IP address when looking up the host name (a safety check), it
6698 output the message "<ip address> does not match any IP for NULL", which was
6699 confusing, to say the least. The bug was that the host name should have
6700 appeared instead of "NULL".
6702 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6703 user that is built into the binary, and the -C or -D options is used, root
6704 privilege is dropped before the configuration file is read. In addition,
6705 logging is switched to stderr instead of the normal log files. If the
6706 configuration then re-defines the Exim user, the unprivileged environment
6707 is probably not what is expected, so Exim logs a panic warning message (but
6710 However, if deliver_drop_privilege is set, the unprivileged state may well
6711 be exactly what is intended, so the warning has been cut out in that case,
6712 and Exim is allowed to try to write to its normal log files.
6718 1. smtp_return_error_details was not giving details for temporary sender
6719 or receiver verification errors.
6721 2. Diagnose a configuration error if two authenticators have the same public
6724 3. Exim used not to create the message log file for a message until the first
6725 delivery attempt. This could be confusing when incoming messages were held
6726 for policy or load reasons. The message log file is now created at the time
6727 the message is received, and an initial "Received" line is written to it.
6729 4. The automatically generated man page for command line options had a minor
6730 bug that caused no ill effects; however, a more serious problem was that
6731 the procedure for building the man page automatically didn't always
6732 operate. Consequently, release 4.20 contains an out-of-date version. This
6733 shouldn't happen again.
6735 5. When building Exim with embedded Perl support, the script that builds the
6736 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6737 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6739 6. The freeze_tell option was not being used for messages that were frozen on
6740 arrival, either by an ACL or by local_scan().
6742 7. Added the smtp_incomplete_transaction log selector.
6744 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6745 was accepting AUTH without a new EHLO.
6747 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6748 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6751 10. Logging of TCP/IP connections (when configured) now happens in the main
6752 daemon process instead of the child process, so that the TCP/IP connection
6753 count is more accurate (but it can never be perfect).
6755 11. The use of "drop" in a nested ACL was not being handled correctly in the
6756 outer ACL. Now, if condition failure induced by the nested "drop" causes
6757 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6758 or "require"), the connection is dropped.
6760 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6761 that yield "discard" can now be used with an "accept" or a "discard" verb,
6762 but an error is generated for any others (because I can't see a useful way
6763 to define what should happen).
6765 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6766 are now processed in the same way as lines in the Exim configuration file.
6767 In particular, continuation lines are supported.
6769 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6771 15. Added -ti meaning -t -i.
6773 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6774 domains, and warn by logging if others are found.
6776 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6777 returned from GIFCONF when Exim is trying to find the list of interfaces on
6778 a host. The code in os.c has been modified to copy each ifreq to an aligned
6779 structure in all cases.
6781 Also, in some cases, the returned ifreq's were being copied to a 'struct
6782 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6783 means the last couple of bytes of an IPv6 address could be chopped if the
6784 ifreq contained only a normal sockaddr (14 bytes storage).
6786 18. Named domain lists were not supported in the hosts_treat_as_local option.
6787 An entry such as +xxxx was not recognized, and was treated as a literal
6790 19. Ensure that header lines added by a DATA ACL are included in the reject log
6791 if the ACL subsequently rejects the message.
6793 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6794 MD5 (which is deprecated).
6796 21. When testing a filter file using -bf, Exim was writing a message when it
6797 took the sender from a "From " line in the message, but it was not doing so
6798 when it took $return_path from a Return-Path: header line. It now does.
6800 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6801 with a valid header line field name (a series of printing characters
6802 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6804 23. Changed "disc" in the source to "disk" to conform to the documentation and
6805 the book and for uniformity.
6807 24. Ignore Sendmail's -Ooption=value command line item.
6809 25. When execve() failed while trying to run a command in a pipe transport,
6810 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6811 could be confused with a return value of 69 from the command itself. This
6812 has been changed to 127, the value the shell returns if it is asked to run
6813 a non-existent command. The wording for the related log line suggests a
6814 non-existent command as the problem.
6816 26. If received_header_text expands to an empty string, do not add a Received:
6817 header line to the message. (Well, it adds a token one on the spool, but
6818 marks it "old" so that it doesn't get used or transmitted.)
6820 27. Installed eximstats 1.28 (addition of -nt option).
6822 28. There was no check for failure on the call to getsockname() in the daemon
6823 code. This can fail if there is a shortage of resources on the system, with
6824 ENOMEM, for example. A temporary error is now given on failure.
6826 29. Contrary to the C standard, it seems that in some environments, the
6827 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6828 program. Exim now does this explicitly; it affects the formatting of
6829 timestamps using strftime().
6831 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6832 complaints. I've now initialized all the variables, to avoid this.
6834 32. Header lines added by a system filter were not being "seen" during
6835 transport-time rewrites.
6837 33. The info_callback() function passed to OpenSSL is set up with type void
6838 (*)(SSL *, int, int), as described somewhere. However, when calling the
6839 function (actually a macro) that sets it up, the type void(*)() is
6840 expected. I've put in a cast to prevent warnings from picky compilers.
6842 34. If a DNS black list lookup found a CNAME record, but there were no A
6843 records associated with the domain it pointed at, Exim crashed.
6845 35. If a DNS black list lookup returned more than one A record, Exim ignored
6846 all but the first. It now scans all returned addresses if a particular IP
6847 value is being sought. In this situation, the contents of the
6848 $dnslist_value variable are a list of all the addresses, separated by a
6851 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6852 to accept a host name and all its aliases if the forward lookup for any of
6853 them yielded the IP address of the incoming connection. Now it accepts only
6854 those names whose forward lookup yields the correct IP address. Any other
6855 names are discarded. This closes a loophole whereby a rogue DNS
6856 administrator could create reverse DNS records to break through a
6857 wildcarded host restriction in an ACL.
6859 37. If a user filter or a system filter that ran in a subprocess used any of
6860 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6861 the wrong values were passed to the pipe command ($thisaddress had the
6862 value of $0, $0 had the value of $1, etc). This bug was introduced by
6863 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6865 38. Improved the line breaking for long SMTP error messages from ACLs.
6866 Previously, if there was no break point between 40 and 75 characters, Exim
6867 left the rest of the message alone. Two changes have been made: (a) I've
6868 reduced the minimum length to 35 characters; (b) if it can't find a break
6869 point between 35 and 75 characters, it looks ahead and uses the first one
6870 that it finds. This may give the occasional overlong line, but at least the
6871 remaining text gets split now.
6873 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6874 file descriptors might be low, and that setting 1000 would always raise it.
6875 It turns out that in some environments, the limit is already over 1000 and
6876 that lowering it causes trouble. So now Exim takes care not to decrease it.
6878 40. When delivering a message, the value of $return_path is set to $sender_
6879 address at the start of routing (routers may change the value). By an
6880 oversight, this default was not being set up when an address was tested by
6881 -bt or -bv, which affected the outcome if any router or filter referred to
6884 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6885 log only when "message" or "log_message" are set. However, if one of the
6886 conditions was an address verification, or a call to a nested ACL, the
6887 messages generated by the underlying test were being passed through. This
6888 no longer happens. The underlying message is available in $acl_verify_
6889 message for both "message" and "log_message" expansions, so it can be
6890 passed through if needed.
6892 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6893 new expansion $bh_ to give the encoded byte string without charset
6894 translation. Translation happens only if iconv() is available; HAVE_ICONV
6895 indicates this at build time. HEADERS_CHARSET gives the charset to
6896 translate to; headers_charset can change it in the configuration, and
6897 "headers charset" can change it in an individual filter file.
6899 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6900 that creates RFC 2047 encoded "words" labels them as that charset instead
6901 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6902 expansion operator; (ii) when Exim creates a From: line for a local
6903 message; (iii) when a header line is rewritten to include a "phrase" part.
6905 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6906 buggy, causing it to skip the first lines of messages whose message ID
6907 ended in 'D'. This would not have bitten before Exim release 4.14, because
6908 message IDs were unlikely to end in 'D' before then. The effect was to have
6909 incorrect size information for certain domains.
6911 45. #include "config.h" was missing at the start of the crypt16.c module. This
6912 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6915 46. If there was a timeout during a "random" callout check, Exim treated it as
6916 a failure of the random address, and carried on sending RSET and the real
6917 address. If the delay was just some slowness somewhere, the response to the
6918 original RCPT would be taken as a response to RSET and so on, causing
6919 mayhem of various kinds.
6921 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6922 when I implemented it. It didn't allow for the fact that some option values
6923 may legitimately be negative (e.g. size_addition), and it didn't even do
6924 the right test for positive values.
6926 48. Domain names in DNS records are case-independent. Exim always looks them up
6927 in lower case. Some resolvers return domain names in exactly the case they
6928 appear in the zone file, that is, they may contain uppercase letters. Not
6929 all resolvers do this - some return always lower case. Exim was treating a
6930 change of case by a resolver as a change of domain, similar to a widening
6931 of a domain abbreviation. This triggered its re-routing code and so it was
6932 trying to route what was effectively the same domain again. This normally
6933 caused routing to fail (because the router wouldn't handle the domain
6934 twice). Now Exim checks for this case specially, and just changes the
6935 casing of the domain that it ultimately uses when it transmits the message
6938 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6941 50. If a filter generated a file delivery with a non-absolute name (possible if
6942 no home directory exists for the router), the forbid_file option was not
6945 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6946 the existing equality matching.
6948 52. Exim was using ints instead of ino_t variables in some places where it was
6949 dealing with inode numbers.
6951 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6952 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6953 called TMPDIR, and if it finds it is different, it changes its value.
6955 54. The smtp_printf() function is now made available to local_scan() so
6956 additional output lines can be written before returning. There is also an
6957 smtp_fflush() function to enable the detection of a dropped connection.
6958 The variables smtp_input and smtp_batched_input are exported to
6961 55. Changed the default runtime configuration: the message "Unknown user"
6962 has been removed from the ACL, and instead placed on the localuser router,
6963 using the cannot_route_message feature. This means that any verification
6964 failures that generate their own messages won't get overridden. Similarly,
6965 the "Unrouteable address" message that was in the ACL for unverifiable
6966 relay addresses has also been removed.
6968 56. Added hosts_avoid_esmtp to the smtp transport.
6970 57. The exicyclog script was not checking for the esoteric option
6971 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6972 will work only if exicyclog is run under the appropriate euid.
6974 58. Following a discussion on the list, the rules by which Exim recognises line
6975 endings on incoming messages have been changed. The -dropcr and drop_cr
6976 options are now no-ops, retained only for backwards compatibility. The
6977 following line terminators are recognized: LF CRLF CR. However, special
6978 processing applies to CR:
6980 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6981 nor a local message in the state where . is a terminator.
6983 (ii) If a bare CR is encountered in a header line, an extra space is added
6984 after the line terminator so as not to end the header. The reasoning
6985 behind this is that bare CRs in header lines are most likely either
6986 to be mistakes, or people trying to play silly games.
6988 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6989 was being incorrectly given as 18 bytes larger than it should have been.
6990 This is a VOB (very old bug).
6992 60. This may never have affected anything current, but just in case it has:
6993 When the local host is found other than at the start of a list of hosts,
6994 the local host, those with the same MX, and any that follow, are discarded.
6995 When the list in question was part of a longer list of hosts, the following
6996 hosts (not currently being processed) were also being discarded. This no
6997 longer happens. I'm not sure if this situation could ever has previously
7000 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7003 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7004 header. This is something that is recommended in a new Internet Draft, and
7005 is something that is documented as being done by Sendmail. There are two
7006 possible values. For messages generated by the autoreply transport, Exim
7009 Auto-Submitted: auto-replied
7011 whereas for all other generated messages (e.g. bounces) it adds
7013 Auto-Submitted: auto-generated
7015 63. The "personal" condition in filters now includes a test for the
7016 Auto-Submitted: header. If it contains the string "auto-" the message it
7017 not considered personal.
7019 64. Added rcpt_include_affixes as a generic transport option.
7021 65. Added queue_only_override (default true).
7023 66. Added the syslog_duplication option.
7025 67. If what should have been the first header line of a message consisted of
7026 a space followed by a colon, Exim was mis-interpreting it as a header line.
7027 It isn't of course - it is syntactically invalid and should therefore be
7028 treated as the start of the message body. The misbehaviour could have
7029 caused a number of strange effects, including loss of data in subsequent
7030 header lines, and spool format errors.
7032 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7033 client host had authenticated. This control can now be exercised by an ACL
7034 for more flexibility.
7036 69. By default, callouts do not happen when testing with -bh. There is now a
7037 variant, -bhc, which does actually run the callout code, including
7038 consulting and updating the callout cache.
7040 70. Added support for saslauthd authentication, courtesy of Alexander
7043 71. If statvfs() failed on the spool or log directories while checking their
7044 size for availability, Exim confusingly gave the error "space shortage".
7045 Furthermore, in debugging mode it crashed with a floating point exception.
7046 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7047 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7048 serious problem, Exim now writes to the main and panic logs when this
7049 happens, with details of the failure. It then refuses to accept the
7050 incoming message, giving the message "spool directory problem" or "log
7051 directory problem" with a 421 code for SMTP messages.
7053 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7054 0, 1, and 2 exist, because some OS complain for execs without them (see
7055 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7056 descriptors. However, the code omitted to check that the open succeeded,
7057 causing mysterious errors if for some reason the permissions on /dev/null
7058 got screwed. Now Exim writes a message to the main and panic logs, and
7059 bombs out if it can't open /dev/null.
7061 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7062 interact so that it is all more flexible. It is supposed to remain
7063 backwards compatible. Also added extra_local_interfaces.
7065 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7066 to bomb out with an assertion failure - to the client this appears as a
7067 connection drop. This problem occurs in the part of the code that was taken
7068 from the Samba project. Fortunately, the assertion is in a very simple
7069 function, so I have fixed this by reproducing the function inline in the
7070 one place where it is called, and arranging for authentication to fail
7071 instead of killing the process with assert().
7073 75. The SPA client code was not working when the server requested OEM rather
7074 than Unicode encoding.
7076 76. Added code to make require_files with a specific uid setting more usable in
7077 the case where statting the file as root fails - usually a non-root-mounted
7078 NFS file system. When this happens and the failure is EACCES, Exim now
7079 forks a subprocess and does the per-uid checking as the relevant uid.
7081 77. Added process_log_path.
7083 78. If log_file_path was not explicitly set, a setting of check_log_space or
7084 check_log_inodes was ignored.
7086 79. If a space check for the spool or log partitions fails, the incident is now
7087 logged. Of course, in the latter case the data may get lost...
7089 80. Added the %p formatting code to string_format() so that it can be used to
7090 print addresses in debug_print(). Adjusted all the address printing in the
7091 debugging in store.c to use %p rather than %d.
7093 81. There was a concern that a line of code in smtp_in.c could overflow a
7094 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7095 initially expressed, the concern was not well-founded, because trailing
7096 spaces are removed early. However, if the trailing spaces were followed by
7097 a NULL, they did not get removed, so the overflow was possible. Two fixes
7100 (a) I re-wrote the offending code in a cleaner fashion.
7101 (b) If an incoming SMTP command contains a NULL character, it is rejected
7104 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7105 runs initgroups(), so that if the Exim user is in any additional groups,
7106 they will be used during message reception.
7112 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob