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.
145 JH/01 Rework error string handling in TLS interface so that the caller in
146 more cases is responsible for logging. This permits library-sourced
147 string to be attached to addresses during delivery, and collapses
148 pairs of long lines into single ones.
150 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
151 during configuration. Wildcards are allowed and expanded.
153 JH/02 Rework error string handling in DKIM to pass more info back to callers.
154 This permits better logging.
156 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
157 do not close it down and have the child transport start it up again on
158 the passed-on TCP connection. Instead, proxy the child (and any
159 subsequent ones) for TLS via a unix-domain socket channel. Logging is
160 affected: the continued delivery log lines do not have any DNSSEC, TLS
161 Certificate or OCSP information. TLS cipher information is still logged.
163 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
164 identical IP addresses on different listening ports. Will also affect
167 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
168 add noisy ifdef guards to special-case this sillyness.
169 Patch from Bernd Kuhls.
171 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
172 than 255 are no longer allowed.
174 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
175 Disable the session-cache too, which might reduce our load. Since we
176 currrectly use a new context for every connection, both as server and
177 client, there is no benefit for these.
178 GnuTLS appears to not support tickets server-side by default (we don't
179 call gnutls_session_ticket_enable_server()) but client side is enabled
180 by default on recent versions (3.1.3 +) unless the PFS priority string
183 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
184 <https://reproducible-builds.org/specs/source-date-epoch/>.
186 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
187 the check for any unsuccessful recipients did not notice the limit, and
188 erroneously found still-pending ones.
190 JH/08 Pipeline CHUNKING command and data together, on kernels that support
191 MSG_MORE. Only in-clear (not on TLS connections).
193 JH/09 Avoid using a temporary file during transport using dkim. Unless a
194 transport-filter is involved we can buffer the headers in memory for
195 creating the signature, and read the spool data file once for the
196 signature and again for transmission.
198 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
199 4.77 as the kernel support then wasn't solid, having issues in 64bit
200 mode. Now, it's been long enough. Add support for FreeBSD also.
202 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
203 case where the routing stage had gathered several addresses to send to
204 a host before calling the transport for the first, we previously failed
205 to close down TLS in the old transport process before passing the TCP
206 connection to the new process. The new one sent a STARTTLS command
207 which naturally failed, giving a failed delivery and bloating the retry
208 database. Investigation and fix prototype from Wolfgang Breyha.
210 JH/12 Fix check on SMTP command input synchronisation. Previously there were
211 false-negatives in the check that the sender had not preempted a response
212 or prompt from Exim (running as a server), due to that code's lack of
213 awareness of the SMTP input buffering.
215 PP/04 Add commandline_checks_require_admin option.
216 Exim drops privileges sanely, various checks such as -be aren't a
217 security problem, as long as you trust local users with access to their
218 own account. When invoked by services which pass untrusted data to
219 Exim, this might be an issue. Set this option in main configuration
220 AND make fixes to the calling application, such as using `--` to stop
223 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
224 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
225 responses to those, into a single TLS record each way (this usually means
226 a single packet). As a side issue, smtp_enforce_sync now works on TLS
229 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
230 affects you only if you're dancing at the edge of the param size limits.
231 If you are, and this message makes sense to you, then: raise the
232 configured limit or use OpenSSL 1.1. Nothing we can do for older
235 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
236 cmdline element to get the documented default one. Previously it was
239 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
240 are unsuable for non-message-specific reasons - eg. connection timeout,
241 banner-time rejection.
243 JH/16 Fix logging of delivery remote port, when specified by router, under
246 PP/06 Repair manualroute's ability to take options in any order, even if one
247 is the name of a transport.
250 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
252 JH/17 Change the list-building routines interface to use the expanding-string
253 triplet model, for better allocation and copying behaviour.
255 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
256 Previously it was constructed the first time a possibly-matching string
257 was met in the configuration file input during startup; now it is done
260 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
261 compatible one, to avoid the (poorly documented) possibility of a config
262 file in the working directory redirecting the DB files, possibly correpting
263 some existing file. CVE-2017-10140 assigned for BDB.
265 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
266 cache-hot. Previously, although the result was properly cached, the
267 initial verify call returned a defer.
269 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
270 the main verify for receipient in uncached-mode.
272 JH/22 Retire historical build files to an "unsupported" subdir. These are
273 defined as "ones for which we have no current evidence of testing".
275 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
276 if present. Previously it was ignored.
278 JH/24 Start using specified-initialisers in C structure init coding. This is
279 a C99 feature (it's 2017, so now considered safe).
281 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
282 if was a fixed-sized field and bitmask ops via macros; it is now more
285 PP/07 GitHub PR 56: Apply MariaDB build fix.
286 Patch provided by Jaroslav Škarvada.
288 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
289 during Coverity cleanups [4.87 JH/47]
290 Diagnosis and fix provided by Michael Fischer v. Mollard.
292 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
293 the right size to place the terminating semicolon on its own folded
294 line, the header hash was calculated to an incorrect value thanks to
295 the (relaxed) space the fold became.
297 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
298 and confused the parent.
300 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
301 which could crash as a result. This could lead to undeliverable messages.
303 JH/28 Logging: "next input sent too soon" now shows where input was truncated
306 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
307 matters on fast-turnover and PID-randomising systems, which were getting
308 out-of-order delivery.
310 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
311 a possibly-overlapping copy. The symptom was that "Remote host closed
312 connection in response to HELO" was logged instead of the actual 4xx
315 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
316 Previously only that bufferd was discarded, resulting in SYMTP command
319 JH/32 DKIM: when a message has multiple signatures matching an identity given
320 in dkim_verify_signers, run the dkim acl once for each. Previously only
321 one run was done. Bug 2189.
323 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
324 "panic the current process" to "deliberately defer". The panic log is
325 still written with the problem list name; the mail and reject logs now
326 get a temp-reject line for the message that was being handled, saying
327 something like "domains check lookup or other defer". The SMTP 451
328 message is still "Temporary local problem".
330 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
331 A crafted sequence of BDAT commands could result in in-use memory beeing
332 freed. CVE-2017-16943.
334 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
335 from SMTP input. Previously it was always done; now only done for DATA
336 and not BDAT commands. CVE-2017-16944.
338 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
339 to the message (such as an overlong header line). Previously this was
340 not done and we did not exit BDAT mode. Followon from the previous item
341 though a different problem.
347 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
348 than -2003 did; needs libidn2 in addition to libidn.
350 JH/02 The path option on a pipe transport is now expanded before use.
352 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
353 Patch provided by "Björn", documentation fix added too.
355 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
356 missing a wire-to-host endian conversion.
358 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
359 close after a BDAT command line could be taken as a following command,
360 giving a synch failure. Fix by only checking for synch immediately
361 before acknowledging the chunk.
363 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
364 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
365 macro. Patches provided by Josh Soref.
367 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
368 Previously we did not; the RFC seems ambiguous and VRFY is not listed
369 by IANA as a service extension. However, John Klensin suggests that we
372 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
373 the dkim code may be unix-mode line endings rather than smtp wire-format
374 CRLF, so prepend a CR to any bare LF.
376 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
377 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
379 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
380 the first were themselves being wrongly included in the feed into dkim
381 processing; with most chunk sizes in use this resulted in an incorrect
382 body hash calculated value.
384 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
385 DKIM signature block, for verification. Although advised against by
386 standards it is specifically not ruled illegal.
388 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
390 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
391 missing a body hash (the bh= tag).
393 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
394 It seems that HAProxy sends the Proxy Protocol information in clear and
395 only then does a TLS startup, so do the same.
397 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
398 TCP connections (such as for Spamd) unless the daemon successfully set
399 Fast Open mode on its listening sockets. This fixes breakage seen on
400 too-old kernels or those not configured for Fast Open, at the cost of
401 requiring both directions being enabled for TFO, and TFO never being used
402 by non-daemon-related Exim processes.
404 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
405 endings, at least on the first header line. Try to canonify any that get
406 past that check, despite the cost.
408 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
409 now limited to an arbitrary five deep, while parsing addresses with the
410 strip_excess_angle_brackets option enabled.
412 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
413 instead leave the unprompted TLS handshake in socket buffer for the
414 TLS library to consume.
416 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
418 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
420 JH/16 Drop variables when they go out of scope. Memory management drops a whole
421 region in one operation, for speed, and this leaves assigned pointers
422 dangling. Add checks run only under the testsuite which checks all
423 variables at a store-reset and panics on a dangling pointer; add code
424 explicitly nulling out all the variables discovered. Fixes one known
425 bug: a transport crash, where a dangling pointer for $sending_ip_address
426 originally assigned in a verify callout, is re-used.
428 PP/06 Drop '.' from @INC in various Perl scripts.
430 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
432 PP/08 Reduce a number of compilation warnings under clang; building with
433 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
434 should be warning-free.
436 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
438 HS/01 Fix portability problems introduced by PP/08 for platforms where
439 realloc(NULL) is not equivalent to malloc() [SunOS et al].
441 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
442 chunk. This allows us to accept broken chunked messages. We need a more
443 general solution here.
445 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
446 already-broken messages in the queue.
448 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
450 JH/19 Fix reference counting bug in routing-generated-address tracking.
456 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
457 supports it and a size is available (ie. the sending peer gave us one).
459 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
460 years of being deprecated). The replacements are the ACLs
461 acl_smtp_mime and acl_not_smtp_mime.
463 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
464 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
465 or even an in-clear connection were permitted. Now, if the host lookup
466 was dnssec and dane was requested then the host is only used if the
467 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
468 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
469 if one fails this test.
470 This means that a poorly-configured remote DNS will make it incommunicado;
471 but it protects against a DNS-interception attack on it.
473 JH/04 Bug 1810: make continued-use of an open smtp transport connection
474 non-noisy when a race steals the message being considered.
476 JH/05 If main configuration option tls_certificate is unset, generate a
477 self-signed certificate for inbound TLS connections.
479 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
480 in rewrites and routers.
482 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
483 and logged a warning sing 4.83; now they are a configuration file error.
485 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
486 (lacking @domain). Apply the same qualification processing as RCPT.
488 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
490 JH/10 Support ${sha256:} applied to a string (as well as the previous
493 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
494 a cutthrough deliver is pending, as we always want to make a connection.
495 This also avoids re-routing the message when later placing the cutthrough
496 connection after a verify cache hit.
497 Do not update it with the verify result either.
499 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
500 when routing results in more than one destination address.
502 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
503 signing (which inhibits the cutthrough capability). Previously only
504 the presence of an option was tested; now an expansion evaluating as
505 empty is permissible (obviously it should depend only on data available
506 when the cutthrough connection is made).
508 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
509 the relevant preceding SMTP command did not note the pipelining mode.
511 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
512 Previously they were not counted.
514 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
515 as one having no matching records. Previously we deferred the message
516 that needed the lookup.
518 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
519 distinguished as "(=".
521 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
522 for missing MX records. Previously it only worked for missing A records.
524 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
526 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
527 after the data-go-ahead and data-ack. Patch from Jason Betts.
529 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
530 even for a "none" policy. Patch from Tony Meyer.
532 JH/22 Fix continued use of a connection for further deliveries. If a port was
533 specified by a router, it must also match for the delivery to be
536 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
537 When one of the recipients of a message was unsuitable for the connection
538 (has no matching addresses), we lost track of needing to mark it
539 deferred. As a result mail would be lost.
541 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
543 JH/25 Decoding ACL controls is now done using a binary search; the source code
544 takes up less space and should be simpler to maintain. Merge the ACL
545 condition decode tables also, with similar effect.
547 JH/26 Fix problem with one_time used on a redirect router which returned the
548 parent address unchanged. A retry would see the parent address marked as
549 delivered, so not attempt the (identical) child. As a result mail would
552 JH/27 Fix a possible security hole, wherein a process operating with the Exim
553 UID can gain a root shell. Credit to http://www.halfdog.net/ for
554 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
557 JH/28 Enable {spool,log} filesystem space and inode checks as default.
558 Main config options check_{log,spool}_{inodes,space} are now
559 100 inodes, 10MB unless set otherwise in the configuration.
561 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
562 Previously it only applied to the main-section connection policy
565 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
567 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
568 by me. Added RFC7919 DH primes as an alternative.
570 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
571 are not in the system include path.
573 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
574 GnuTLS, when a session startup failed (eg because the client disconnected)
575 Exim did stdio operations after fclose. This was exposed by a recent
576 change which nulled out the file handle after the fclose.
578 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
579 signed directly by the cert-signing cert, rather than an intermediate
580 OCSP-signing cert. This is the model used by LetsEncrypt.
582 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
584 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
585 an incoming connection.
587 HS/02 Bug 1802: Do not half-close the connection after sending a request
590 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
591 fallback to "prime256v1".
593 JH/34 SECURITY: Use proper copy of DATA command in error message.
594 Could leak key material. Remotely exploitable. CVE-2016-9963.
600 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
601 and 3.4.4 - once the server is enabled to respond to an OCSP request
602 it does even when not requested, resulting in a stapling non-aware
603 client dropping the TLS connection.
605 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
606 support variable-length bit vectors. No functional change.
608 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
609 The I= interface field on outgoing lines is now after the H= remote
610 host field, same as incoming lines. There is a separate
611 outgoing_interface log selector which allows you to disable the
614 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
615 If not running log_selector +smtp_connection the mainlog would be held
616 open indefinitely after a "too many connections" event, including to a
617 deleted file after a log rotate. Leave the per net connection logging
618 leaving it open for efficiency as that will be quickly detected by the
619 check on the next write.
621 HS/01 Bug 1671: Fix post transport crash.
622 Processing the wait-<transport> messages could crash the delivery
623 process if the message IDs didn't exist for some reason. When
624 using 'split_spool_directory=yes' the construction of the spool
625 file name failed already, exposing the same netto behaviour.
627 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
628 mime_regex ACL conditions.
630 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
631 to DSN fail messages (bounces): remote IP, remote greeting, remote response
632 to HELO, local diagnostic string.
634 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
635 log line to debug. Even when configured with a tls authenticator many
636 client connections are expected to not authenticate in this way, so
637 an authenticate fail is not an error.
639 HS/02 Add the Exim version string to the process info. This way exiwhat
640 gives some more detail about the running daemon.
642 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
643 matter for fast-change records such as DNSBLs.
645 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
646 retry record, even if constant. There may be multiple transports with
647 different interface settings and the retry behaviour needs to be kept
650 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
652 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
654 JH/10 Bug 840: fix log_defer_output option of pipe transport
656 JH/11 Bug 830: use same host for all RCPTS of a message, even under
657 hosts_randomize. This matters a lot when combined with mua_wrapper.
659 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
660 ${quote_pgsql:<string>} operator.
662 JH/13 Bug 1708: avoid misaligned access in cached lookup.
664 JH/14 Change header file name for freeradius-client. Relevant if compiling
665 with Radius support; from the Gentoo tree and checked under Fedora.
667 JH/15 Bug 1712: Introduce $prdr_requested flag variable
669 JH/16 Bug 1714: Permit an empty string as expansion result for transport
670 option transport_filter, meaning no filtering.
672 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
674 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
675 defaults to "*" (all hosts). The variable is now available when not built
676 with TLS, default unset, mainly to enable keeping the testsuite sane.
677 If a server certificate is not supplied (via tls_certificate) an error is
678 logged, and clients will find TLS connections fail on startup. Presumably
679 they will retry in-clear.
680 Packagers of Exim are strongly encouraged to create a server certificate
681 at installation time.
683 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
684 with the $config_file variable.
686 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
687 in transport context, after the attempt, and per-recipient. The latter type
688 is per host attempted. The event data is the error message, and the errno
689 information encodes the lookup type (A vs. MX) used for the (first) host,
690 and the trailing two digits of the smtp 4xx response.
692 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
693 to write to mainlog (or rejectlog, paniclog) in the window between file
694 creation and permissions/ownership being changed. Particularly affects
695 installations where exicyclog is run as root, rather than exim user;
696 result is that the running daemon panics and dies.
698 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
700 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
701 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
702 "pri" and "weight". Note that the previous implicit priority given by the
703 list order is no longer honoured.
705 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
708 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
709 by defining SUPPORT_SOCKS.
711 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
712 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
713 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
714 variables are renamed to proxy_{local,external}_{address,port}.
716 JH/27 Move Internationalisation support from Experimental to mainline, enabled
717 for a build by defining SUPPORT_I18N
719 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
720 of the query string, and make ${quote_redis:} do that quoting.
722 JH/29 Move Events support from Experimental to mainline, enabled by default
723 and removable for a build by defining DISABLE_EVENT.
725 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
727 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
728 cached by the daemon.
730 JH/32 Move Redis support from Experimental to mainline, enabled for a build
731 by defining LOOKUP_REDIS. The libhiredis library is required.
733 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
734 keys are given for lookup.
736 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
737 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
738 only supported when built with TLS support. The PolarSSL SHA routines
739 are still used when the TLS library is too old for convenient support.
741 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
742 openssl_options), for security. OpenSSL forces this from version 1.1.0
743 server-side so match that on older versions.
745 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
746 allocation for $value could be released as the expansion processing
747 concluded, but leaving the global pointer active for it.
749 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
750 and to use the domains and local_parts ACL conditions.
752 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
753 incorrectly not doubled on cutthrough transmission, hence seen as a
754 body-termination at the receiving system - resulting in truncated mails.
755 Commonly the sender saw a TCP-level error, and retransmitted the message
756 via the normal store-and-forward channel. This could result in duplicates
757 received - but deduplicating mailstores were liable to retain only the
758 initial truncated version.
760 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
762 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
764 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
765 we're in there, support oversigning also; bug 1309.
767 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
769 HS/04 Add support for keep_environment and add_environment options.
771 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
772 either intentional arithmetic overflow during PRNG, or testing config-
775 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
776 delivery resulted in actual delivery. Cancel cutthrough before DATA
779 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
780 rejects a recipient: pass the reject to the originator.
782 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
783 Many were false-positives and ignorable, but it's worth fixing the
786 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
787 for the new environment-manipulation done at startup. Move the routines
788 from being local to tls.c to being global via the os.c file.
790 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
791 an extract embedded as result-arg for a map, the first arg for extract
792 is unavailable so we cannot tell if this is a numbered or keyed
793 extraction. Accept either.
799 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
802 JH/02 The smtp transport option "multi_domain" is now expanded.
804 JH/03 The smtp transport now requests PRDR by default, if the server offers
807 JH/04 Certificate name checking on server certificates, when exim is a client,
808 is now done by default. The transport option tls_verify_cert_hostnames
809 can be used to disable this per-host. The build option
810 EXPERIMENTAL_CERTNAMES is withdrawn.
812 JH/05 The value of the tls_verify_certificates smtp transport and main options
813 default to the word "system" to access the system default CA bundle.
814 For GnuTLS, only version 3.0.20 or later.
816 JH/06 Verification of the server certificate for a TLS connection is now tried
817 (but not required) by default. The verification status is now logged by
818 default, for both outbound TLS and client-certificate supplying inbound
821 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
824 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
825 Status Notification (bounce) messages are now MIME format per RFC 3464.
826 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
827 under the control of the dsn_advertise_hosts option, and routers may
828 have a dsn_lasthop option.
830 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
831 default, modifiable by a malware= option. The list separator for
832 the options can now be changed in the usual way. Bug 68.
834 JH/10 The smtp_receive_timeout main option is now expanded before use.
836 JH/11 The incoming_interface log option now also enables logging of the
837 local interface on delivery outgoing connections.
839 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
840 if the interface and destination host and port all match.
842 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
845 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
846 Patch from Andrew Lewis.
848 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
849 now supports optional time-restrictions, weighting, and priority
850 modifiers per server. Patch originally by <rommer@active.by>.
852 JH/16 The spamd_address main option now supports a mixed list of local
853 and remote servers. Remote servers can be IPv6 addresses, and
854 specify a port-range.
856 JH/17 Bug 68: The spamd_address main option now supports an optional
857 timeout value per server.
859 JH/18 Bug 1581: Router and transport options headers_add/remove can
860 now have the list separator specified.
862 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
865 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
868 JH/21 Support for the A6 type of dns record is withdrawn.
870 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
871 rather than the verbs used.
873 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
874 from 255 to 1024 chars.
876 JH/24 Verification callouts now attempt to use TLS by default.
878 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
879 are generic router options now. The defaults didn't change.
881 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
882 Original patch from Alexander Shikoff, worked over by JH.
884 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
885 files if ZCAT_COMMAND is not executable.
887 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
889 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
891 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
892 Normally benign, it bites when the pair was led to by a CNAME;
893 modern usage is to not canonicalize the domain to a CNAME target
894 (and we were inconsistent anyway for A-only vs AAAA+A).
896 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
898 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
899 when evaluating $sender_host_dnssec.
901 JH/31 Check the HELO verification lookup for DNSSEC, adding new
902 $sender_helo_dnssec variable.
904 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
906 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
908 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
910 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
911 documented as working, but never had. Support all but $spam_report.
913 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
914 added for tls authenticator.
916 HS/03 Add perl_taintmode main config option
922 TL/01 When running the test suite, the README says that variables such as
923 no_msglog_check are global and can be placed anywhere in a specific
924 test's script, however it was observed that placement needed to be near
925 the beginning for it to behave that way. Changed the runtest perl
926 script to read through the entire script once to detect and set these
927 variables, reset to the beginning of the script, and then run through
928 the script parsing/test process like normal.
930 TL/02 The BSD's have an arc4random API. One of the functions to induce
931 adding randomness was arc4random_stir(), but it has been removed in
932 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
933 function when detected.
935 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
936 cause callback expansion.
938 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
939 syntax errors in an expansion can be treated as a string instead of
940 logging or causing an error, due to the internal use of bool_lax
941 instead of bool when processing it.
943 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
944 server certificates when making smtp deliveries.
946 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
948 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
950 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
952 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
953 Merged patch from Sebastian Wiedenroth.
955 JH/05 Fix results-pipe from transport process. Several recipients, combined
956 with certificate use, exposed issues where response data items split
957 over buffer boundaries were not parsed properly. This eventually
958 resulted in duplicates being sent. This issue only became common enough
959 to notice due to the introduction of connection certificate information,
960 the item size being so much larger. Found and fixed by Wolfgang Breyha.
962 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
963 size buffer was used, resulting in syntax errors when an expansion
966 JH/07 Add support for directories of certificates when compiled with a GnuTLS
967 version 3.3.6 or later.
969 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
970 is EXPERIMENTAL_EVENT, the main-configuration and transport options
971 both become "event_action", the variables become $event_name, $event_data
972 and $event_defer_errno. There is a new variable $verify_mode, usable in
973 routers, transports and related events. The tls:cert event is now also
974 raised for inbound connections, if the main configuration event_action
977 TL/06 In test suite, disable OCSP for old versions of openssl which contained
978 early OCSP support, but no stapling (appears to be less than 1.0.0).
980 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
981 server certificate names available under the smtp transport option
982 "tls_verify_cert_hostname" now do not permit multi-component wildcard
985 JH/10 Time-related extraction expansions from certificates now use the main
986 option "timezone" setting for output formatting, and are consistent
987 between OpenSSL and GnuTLS compilations. Bug 1541.
989 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
990 encoded parameter in the incoming message. Bug 1558.
992 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
993 include certificate info, eximon was claiming there were spoolfile
996 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
998 JH/14 Log delivery-related information more consistently, using the sequence
999 "H=<name> [<ip>]" wherever possible.
1001 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1002 are problematic for Debian distribution, omit them from the release
1005 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1007 JH/16 Fix string representation of time values on 64bit time_t architectures.
1010 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1011 output list separator was used.
1016 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1017 checkers that were complaining about end of non-void function with no
1020 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1021 This was a regression introduced in 4.83 by another bugfix.
1023 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1025 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1026 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1032 TF/01 Correctly close the server side of TLS when forking for delivery.
1034 When a message was received over SMTP with TLS, Exim failed to clear up
1035 the incoming connection properly after forking off the child process to
1036 deliver the message. In some situations the subsequent outgoing
1037 delivery connection happened to have the same fd number as the incoming
1038 connection previously had. Exim would try to use TLS and fail, logging
1039 a "Bad file descriptor" error.
1041 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1042 utilities have not been installed.
1044 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1045 temporary space as the ACL may create new global variables.
1047 TL/01 LDAP support uses per connection or global context settings, depending
1048 upon the detected version of the libraries at build time.
1050 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1051 to extract and use the src ip:port in logging and expansions as if it
1052 were a direct connection from the outside internet. PPv2 support was
1053 updated based on HAProxy spec change in May 2014.
1055 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1057 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1058 Properly escape header and check for NULL return.
1060 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1063 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1065 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1066 characters in header names, implemented as a verify condition.
1067 Contributed by Michael Fischer v. Mollard.
1069 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1070 results permerror and temperror. Previous values are deprecated but
1071 still accepted. In a future release, err_perm and err_temp will be
1072 completely removed, which will be a backward incompatibility if the
1073 ACL tests for either of these two old results. Patch contributed by
1074 user bes-internal on the mailing list.
1076 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1078 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1079 selectors, in both main and reject logs.
1081 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1084 JH/07 Add malware type "sock" for talking to simple daemon.
1086 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1088 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1089 routers/transports under cutthrough routing.
1091 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1092 numbers. Touch up "bool" conditional to keep the same definition.
1094 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1096 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1098 JH/12 Expand items in router/transport headers_add or headers_remove lists
1099 individually rather than the list as a whole. Bug 1452.
1101 Required for reasonable handling of multiple headers_ options when
1102 they may be empty; requires that headers_remove items with embedded
1103 colons must have them doubled (or the list-separator changed).
1105 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1106 view the policy declared in the DMARC record. Currently, $dmarc_status
1107 is a combined value of both the record presence and the result of the
1110 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1112 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1113 dnslookup router and the smtp transport (applying to the forward
1116 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1117 of ldap servers used for a specific lookup. Patch provided by Heiko
1120 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1121 New variable $lookup_dnssec_authenticated for observability.
1123 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1124 Patch submitted by Lars Timman.
1126 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1128 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1129 Requires trusted mode and valid format message id, aborts otherwise.
1130 Patch contributed by Heiko Schlichting.
1132 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1133 certextract with support for various fields. Bug 1358.
1135 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1136 is requested by default, modifiable by smtp transport option
1139 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1140 operate on certificate variables to give certificate fingerprints
1141 Also new ${sha256:cert_variable}.
1143 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1145 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1146 Christian Aistleitner.
1148 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1150 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1151 file. Patch from Wolfgang Breyha.
1153 JH/25 Expand the coverage of the delivery $host and $host_address to
1154 client authenticators run in verify callout. Bug 1476.
1156 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1157 align with daemon_smtp_ports. Bug 72.
1159 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1160 support and error reporting did not work properly.
1162 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1163 and is readable. Patch from Andrew Colin Kissa.
1165 TL/14 Enhance documentation of ${run expansion and how it parses the
1166 commandline after expansion, particularly in the case when an
1167 unquoted variable expansion results in an empty value.
1169 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1171 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1172 by renaming away from it. Addresses GH issue 15, reported by
1175 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1177 TL/15 SECURITY: prevent double expansion in math comparison functions
1178 (can expand unsanitized data). Not remotely exploitable.
1185 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1187 PP/02 Make -n do something, by making it not do something.
1188 When combined with -bP, the name of an option is not output.
1190 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1193 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1194 $sender_host_name and config options to manage this, and basic check
1197 PP/05 DSCP support for outbound connections and control modifier for inbound.
1199 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1200 (Only plugin which currently uses this is kerberos4, which nobody should
1201 be using, but we should make it available and other future plugins might
1202 conceivably use it, even though it would break NAT; stuff *should* be
1203 using channel bindings instead).
1205 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1206 name; added for Sendmail compatibility; requires admin caller.
1207 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1208 just ignore it); requires trusted caller.
1209 Also parse but ignore: -Ac -Am -X<logfile>
1212 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1214 TL/02 Add +smtp_confirmation as a default logging option.
1216 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1217 Patch by Magnus Holmgren from 2007-02-20.
1219 TL/04 Bugzilla 1281 - Spec typo.
1220 Bugzilla 1283 - Spec typo.
1221 Bugzilla 1290 - Spec grammar fixes.
1223 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1225 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1227 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1228 mailing list by by Dmitry Isaikin.
1230 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1232 JH/02 Support "G" suffix to numbers in ${if comparisons.
1234 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1236 NM/01 Bugzilla 1197 - Spec typo
1237 Bugzilla 1196 - Spec examples corrections
1239 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1241 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1242 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1245 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1246 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1248 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1249 "acl {{name}{arg}...}", and optional args on acl condition
1252 JH/05 Permit multiple router/transport headers_add/remove lines.
1254 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1256 JH/07 Avoid using a waiting database for a single-message-only transport.
1257 Performance patch from Paul Fisher. Bugzilla 1262.
1259 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1262 JH/09 Add $headers_added variable, with content from use of ACL modifier
1263 add_header (but not yet added to the message). Bugzilla 199.
1265 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1266 Pulled from Bugzilla 817 by Wolfgang Breyha.
1268 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1270 (nb: this is the same fix as in Exim 4.80.1)
1272 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1275 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1278 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1280 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1281 advertise SMTP AUTH mechanism to us, instead of a generic
1282 protocol violation error. Also, make Exim more robust to bad
1283 data from the Dovecot auth socket.
1285 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1287 When a queue runner is handling a message, Exim first routes the
1288 recipient addresses, during which it prunes them based on the retry
1289 hints database. After that it attempts to deliver the message to
1290 any remaining recipients. It then updates the hints database using
1293 So if a recipient address works intermittently, it can get repeatedly
1294 deferred at routing time. The retry hints record remains fresh so the
1295 address never reaches the final cutoff time.
1297 This is a fairly common occurrence when a user is bumping up against
1298 their storage quota. Exim had some logic in its local delivery code
1299 to deal with this. However it did not apply to per-recipient defers
1300 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1302 This change adds a proper retry rule check during routing so that the
1303 final cutoff time is checked against the message's age. We only do
1304 this check if there is an address retry record and there is not a
1305 domain retry record; this implies that previous attempts to handle
1306 the address had the retry_use_local_parts option turned on. We use
1307 this as an approximation for the destination being like a local
1308 delivery, as in LMTP.
1310 I suspect this new check makes the old local delivery cutoff check
1311 redundant, but I have not verified this so I left the code in place.
1313 TF/02 Correct gecos expansion when From: is a prefix of the username.
1315 Test 0254 submits a message to Exim with the header
1319 When I ran the test suite under the user fanf2, Exim expanded
1320 the header to contain my full name, whereas it should have added
1321 a Resent-Sender: header. It erroneously treats any prefix of the
1322 username as equal to the username.
1324 This change corrects that bug.
1326 GF/01 DCC debug and logging tidyup
1327 Error conditions log to paniclog rather than rejectlog.
1328 Debug lines prefixed by "DCC: " to remove any ambiguity.
1330 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1332 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1333 Bug spotted by Jeremy Harris; was flawed since initial commit.
1334 Would have resulted in OCSP responses post-SNI triggering an Exim
1335 NULL dereference and crash.
1337 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1339 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1340 Bug detection, analysis and fix by Samuel Thibault.
1341 Bugzilla 1331, Debian bug #698092.
1343 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1345 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1346 Server implementation by Todd Lyons, client by JH.
1347 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1348 config variable "prdr_enable" controls whether the server
1349 advertises the facility. If the client requests PRDR a new
1350 acl_data_smtp_prdr ACL is called once for each recipient, after
1351 the body content is received and before the acl_smtp_data ACL.
1352 The client is controlled by both of: a hosts_try_prdr option
1353 on the smtp transport, and the server advertisement.
1354 Default client logging of deliveries and rejections involving
1355 PRDR are flagged with the string "PRDR".
1357 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1358 fclose(). Diagnosis by Todd Lyons.
1360 PP/17 Update configure.default to handle IPv6 localhost better.
1361 Patch by Alain Williams (plus minor tweaks).
1364 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1365 This is now consistent with GnuTLS, and is now documented: the
1366 previous undocumented portable approach to treating the option as
1367 unset was to force an expansion failure. That still works, and
1368 an empty string is now equivalent.
1370 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1371 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1372 not performing validation itself.
1374 PP/20 Added force_command boolean option to pipe transport.
1375 Patch from Nick Koston, of cPanel Inc.
1377 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1380 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1382 PP/21 Fix eximon continuous updating with timestamped log-files.
1383 Broken in a format-string cleanup in 4.80, missed when I repaired the
1384 other false fix of the same issue.
1385 Report and fix from Heiko Schlichting.
1388 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1389 Report from Prashanth Katuri.
1391 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1392 It's SecureTransport, so affects any MacOS clients which use the
1393 system-integrated TLS libraries, including email clients.
1395 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1396 using a MIME ACL for non-SMTP local injection.
1397 Report and assistance in diagnosis by Warren Baker.
1399 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1401 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1403 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1404 last id that failed so it may be referenced in subsequent ACL's.
1406 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1409 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1410 ldap library initialization, allowing self-signed CA's to be
1411 used. Also properly sets require_cert option later in code by
1412 using NULL (global ldap config) instead of ldap handle (per
1413 session). Bug diagnosis and testing by alxgomz.
1415 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1416 the src/util/ subdirectory.
1418 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1419 renamed to Transport Post Delivery Action by Jeremy Harris, as
1422 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1423 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1424 redis_servers = needs to be configured which will be used by the redis
1425 lookup. Patch from Warren Baker, of The Packet Hub.
1427 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1429 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1430 hostname or reverse DNS when processing a host list. Used suggestions
1431 from multiple comments on this bug.
1433 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1435 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1436 Missed a few lines, added it to make the runtest require no keyboard
1439 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1440 contains upper case chars. Make router use caseful_local_part.
1442 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1443 support when GnuTLS has been built with p11-kit.
1449 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1451 This, or similar/improved, will also be change PP/11 of 4.82.
1457 PP/01 Handle short writes when writing local log-files.
1458 In practice, only affects FreeBSD (8 onwards).
1459 Bugzilla 1053, with thanks to Dmitry Isaikin.
1461 NM/01 Bugzilla 949 - Documentation tweak
1463 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1466 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1468 PP/02 Implemented gsasl authenticator.
1470 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1472 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1473 `pkg-config foo` for cflags/libs.
1475 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1476 with rest of GSASL and with heimdal_gssapi.
1478 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1479 `pkg-config foo` for cflags/libs for the TLS implementation.
1481 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1482 properties get this fed in as external SSF. A number of robustness
1483 and debugging improvements to the cyrus_sasl authenticator.
1485 PP/08 cyrus_sasl server now expands the server_realm option.
1487 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1488 Patch by Jeremy Harris.
1490 PP/10 Added dbmjz lookup type.
1492 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1494 PP/12 MAIL args handles TAB as well as SP, for better interop with
1495 non-compliant senders.
1496 Analysis and variant patch by Todd Lyons.
1498 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1499 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1500 Patch from Dirk Mueller <dmueller@suse.com>
1502 PP/13 tls_peerdn now print-escaped for spool files.
1503 Observed some $tls_peerdn in wild which contained \n, which resulted
1504 in spool file corruption.
1506 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1507 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1508 or write after TLS renegotiation, which otherwise led to messages
1511 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1512 as a tracking header (ie: a signed header comes before the signature).
1513 Patch from Wolfgang Breyha.
1515 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1516 comma-sep list; embedded commas doubled.
1518 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1520 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1522 Report and patch from Dmitry Banschikov.
1524 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1525 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1526 failures. We appear to now support TLS1.1+ with Exim.
1528 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1529 lets Exim select keys and certificates based upon TLS SNI from client.
1530 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1531 before an outbound SMTP session. New log_selector, +tls_sni.
1533 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1534 NULL dereference. Report and patch from Alun Jones.
1536 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1537 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1538 Not seeing resolver debug output on NetBSD, but suspect this is a
1539 resolver implementation change.
1541 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1542 Left warnings. Added "eximon gdb" invocation mode.
1544 PP/21 Defaulting "accept_8bitmime" to true, not false.
1546 PP/22 Added -bw for inetd wait mode support.
1548 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1549 locate the relevant includes and libraries. Made this the default.
1551 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1552 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1554 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1555 This may cause build issues on older platforms.
1557 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1558 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1559 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1560 Added SNI support via GnuTLS too.
1561 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1563 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1565 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1566 Applied second patch from Janne, implementing suggestion to default
1567 multiple-strings-in-record handling to match SPF spec.
1569 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1571 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1572 read-only, out of scope).
1573 Patch from Wolfgang Breyha, report from Stuart Northfield.
1575 PP/29 Fix three issues highlighted by clang analyser static analysis.
1576 Only crash-plausible issue would require the Cambridge-specific
1577 iplookup router and a misconfiguration.
1578 Report from Marcin Mirosław.
1580 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1582 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1583 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1584 As part of this, removing so much warning spew let me fix some minor
1585 real issues in debug logging.
1587 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1588 assignment on my part. Fixed.
1590 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1591 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1592 Janne Snabb (who went above and beyond: thank you).
1594 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1595 string otherwise requires a connection and a bunch more work and it's
1596 relatively easy to get wrong. Should also expose TLS library linkage
1599 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1600 64-bit ${eval} (JH/03).
1602 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1603 GNU libc to support some of the 64-bit stuff, should not lead to
1604 conflicts. Defined before os.h is pulled in, so if a given platform
1605 needs to override this, it can.
1607 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1608 protection layer was required, which is not implemented.
1609 Bugzilla 1254, patch from Wolfgang Breyha.
1611 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1612 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1613 tls_dhparam take prime identifiers. Also unbreak combination of
1614 OpenSSL+DH_params+TLSSNI.
1616 PP/39 Disable SSLv2 by default in OpenSSL support.
1622 PP/01 Solaris build fix for Oracle's LDAP libraries.
1623 Bugzilla 1109, patch from Stephen Usher.
1625 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1627 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1630 TF/02 Fix a couple more cases where we did not log the error message
1631 when unlink() failed. See also change 4.74-TF/03.
1633 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1634 lock up or crash if it happened to be inside a call to libc when it
1635 got a SIGUSR1 from exiwhat.
1637 The SIGUSR1 handler appends the current process status to the process
1638 log which is later printed by exiwhat. It used to use the general
1639 purpose logging code to do this, but several functions it calls are
1640 not safe for signals.
1642 The new output code in the SIGUSR1 handler is specific to the process
1643 log, and simple enough that it's easy to inspect for signal safety.
1644 Removing some special cases also simplifies the general logging code.
1645 Removing the spurious timestamps from the process log simplifies
1648 TF/04 Improved ratelimit ACL condition.
1650 The /noupdate option has been deprecated in favour of /readonly which
1651 has clearer semantics. The /leaky, /strict, and /readonly update modes
1652 are mutually exclusive. The update mode is no longer included in the
1653 database key; it just determines when the database is updated. (This
1654 means that when you upgrade Exim will forget old rate measurements.)
1656 Exim now checks that the per_* options are used with an update mode that
1657 makes sense for the current ACL. For example, when Exim is processing a
1658 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1659 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1660 must specify per_mail/readonly. If you omit the update mode it defaults to
1661 /leaky where that makes sense (as before) or /readonly where required.
1663 The /noupdate option is now undocumented but still supported for
1664 backwards compatibility. It is equivalent to /readonly except that in
1665 ACLs where /readonly is required you may specify /leaky/noupdate or
1666 /strict/noupdate which are treated the same as /readonly.
1668 A useful new feature is the /count= option. This is a generalization
1669 of the per_byte option, so that you can measure the throughput of other
1670 aggregate values. For example, the per_byte option is now equivalent
1671 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1673 The per_rcpt option has been generalized using the /count= mechanism
1674 (though it's more complicated than the per_byte equivalence). When it is
1675 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1676 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1677 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1678 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1679 Note that using per_rcpt with a non-readonly update mode in more than
1680 one ACL will cause the recipients to be double-counted. (The per_mail
1681 and per_byte options don't have this problem.)
1683 The handling of very low rates has changed slightly. If the computed rate
1684 is less than the event's count (usually one) then this event is the first
1685 after a long gap. In this case the rate is set to the same as this event's
1686 count, so that the first message of a spam run is counted properly.
1688 The major new feature is a mechanism for counting the rate of unique
1689 events. The new per_addr option counts the number of different
1690 recipients that someone has sent messages to in the last time period. It
1691 behaves like per_rcpt if all the recipient addresses are different, but
1692 duplicate recipient addresses do not increase the measured rate. Like
1693 the /count= option this is a general mechanism, so the per_addr option
1694 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1695 example, measure the rate that a client uses different sender addresses
1696 with the options per_mail/unique=$sender_address. There are further
1697 details in the main documentation.
1699 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1701 TF/06 Removed a few PCRE remnants.
1703 TF/07 Automatically extract Exim's version number from tags in the git
1704 repository when doing development or release builds.
1706 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1707 Bugzilla 879. Patch from Paul Fisher.
1709 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1710 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1713 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1715 PP/05 Variable $av_failed, true if the AV scanner deferred.
1716 Bugzilla 1078. Patch from John Horne.
1718 PP/06 Stop make process more reliably on build failure.
1719 Bugzilla 1087. Patch from Heiko Schlittermann.
1721 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1722 Bugzilla 1089. Patch from Heiko Schlittermann.
1724 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1725 Bugzilla 1131. Patch from Holger Weiß.
1727 PP/09 Handle IPv6 addresses with SPF.
1728 Bugzilla 860. Patch from Wolfgang Breyha.
1730 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1732 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1735 PP/11 match_* no longer expand right-hand-side by default.
1736 New compile-time build option, EXPAND_LISTMATCH_RHS.
1737 New expansion conditions, "inlist", "inlisti".
1739 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1741 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1743 PP/14 fix log_write() format string regression from TF/03.
1744 Bugzilla 1152. Patch from Dmitry Isaikin.
1750 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1752 PP/02 Harmonised TLS library version reporting; only show if debugging.
1753 Layout now matches that introduced for other libraries in 4.74 PP/03.
1755 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1757 PP/04 New "dns_use_edns0" global option.
1759 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1762 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1763 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1765 TK/01 Updated PolarSSL code to 0.14.2.
1766 Bugzilla 1097. Patch from Andreas Metzler.
1768 PP/07 Catch divide-by-zero in ${eval:...}.
1769 Fixes bugzilla 1102.
1771 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1774 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1775 format-string attack -- SECURITY: remote arbitrary code execution.
1777 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1778 time unintentionally subject to list matching rules, letting the header
1779 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1780 arbitrary string expansion). This allowed for information disclosure.
1782 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1783 INT_MIN/-1 -- value coerced to INT_MAX.
1789 NM/01 Workaround for PCRE version dependency in version reporting
1792 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1793 This fixes portability to compilers other than gcc, notably
1794 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1796 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1797 makefiles for portability to HP-UX and POSIX correctness.
1799 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1800 Also via indented variable definition in the Makefile.
1801 (Debugging by Oliver Heesakkers).
1803 PP/02 Restore caching of spamd results with expanded spamd_address.
1804 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1806 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1807 Improves build reliability. Fix from: Frank Elsner
1809 NM/02 Fix wide character breakage in the rfc2047 coding
1810 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1812 NM/03 Allow underscore in dnslist lookups
1813 Fixes bug 1026. Patch from Graeme Fowler
1815 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1816 Code patches from Adam Ciarcinski of NetBSD.
1818 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1821 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1822 is logged, to avoid truncation. Patch from John Horne.
1824 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1825 Patch from Jakob Hirsch.
1827 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1828 SQL string expansion failure details.
1829 Patch from Andrey Oktyabrski.
1831 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1832 Patch from Simon Arlott.
1834 PP/09 New lookups functionality failed to compile on old gcc which rejects
1835 extern declarations in function scope.
1836 Patch from Oliver Fleischmann
1838 PP/10 Use sig_atomic_t for flags set from signal handlers.
1839 Check getgroups() return and improve debugging.
1840 Fixed developed for diagnosis in bug 927 (which turned out to be
1843 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1844 Patch from Mark Zealey.
1846 PP/12 Bugzilla 1056: Improved spamd server selection.
1847 Patch from Mark Zealey.
1849 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1850 Based on patch from Heiko Schlittermann.
1852 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1853 Patch from Uwe Doering, sign-off by Michael Haardt.
1855 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1856 variable declaration deep within a block. Bug and patch from
1859 PP/15 lookups-Makefile IRIX compatibility coercion.
1861 PP/16 Make DISABLE_DKIM build knob functional.
1863 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1864 Patch by Simon Arlott
1866 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1867 variable argument macros. Our copy now differs from upstream.
1873 TF/01 Failure to get a lock on a hints database can have serious
1874 consequences so log it to the panic log.
1876 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1877 controlled using the smtp_confirmation log selector.
1879 TF/03 Include the error message when we fail to unlink a spool file.
1881 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1882 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1883 for maintaining out-of-tree patches for some time.
1885 PP/01 Bugzilla 139: Documentation and portability issues.
1886 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1887 Handle per-OS dynamic-module compilation flags.
1889 PP/02 Let /dev/null have normal permissions.
1890 The 4.73 fixes were a little too stringent and complained about the
1891 permissions on /dev/null. Exempt it from some checks.
1892 Reported by Andreas M. Kirchwitz.
1894 PP/03 Report version information for many libraries, including
1895 Exim version information for dynamically loaded libraries. Created
1896 version.h, now support a version extension string for distributors
1897 who patch heavily. Dynamic module ABI change.
1899 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1900 privilege escalation vulnerability whereby the Exim run-time user
1901 can cause root to append content of the attacker's choosing to
1904 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1907 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1908 If dropping privileges for untrusted macros, we disabled normal logging
1909 on the basis that it would fail; for the Exim run-time user, this is not
1910 the case, and it resulted in successful deliveries going unlogged.
1911 Fixed. Reported by Andreas Metzler.
1917 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1918 only prepend for the Resent-* case. Fixes regression introduced in
1919 Exim 4.70 by NM/22 for Bugzilla 607.
1921 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1922 increasing numbers of administrators be bitten by this.
1924 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1926 PP/03 Bugzilla 994: added openssl_options main configuration option.
1928 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1930 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1932 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1934 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1935 without a peer certificate, leading to a segfault because of an
1936 assumption that peers always have certificates. Be a little more
1937 paranoid. Problem reported by Martin Tscholak.
1939 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1940 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1941 NB: ClamAV planning to remove STREAM in "middle of 2010".
1942 CL also introduces -bmalware, various -d+acl logging additions and
1943 more caution in buffer sizes.
1945 PP/09 Implemented reverse_ip expansion operator.
1947 PP/10 Bugzilla 937: provide a "debug" ACL control.
1949 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1951 PP/12 Bugzilla 973: Implement --version.
1953 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1955 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1957 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1959 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1960 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1961 ignore trailing whitespace.
1963 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1965 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1968 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1969 Notification from Dr Andrew Aitchison.
1971 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1972 ExtendedDetectionInfo response format.
1973 Notification from John Horne.
1975 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1978 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1979 XSL and documented dependency on system catalogs, with examples of how
1982 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1985 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1986 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1987 configuration file which is writeable by the Exim user or group.
1989 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1990 of configuration files to cover files specified with the -C option if
1991 they are going to be used with root privileges, not just the default
1994 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1995 option (effectively making it always true).
1997 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1998 files to be used while preserving root privileges.
2000 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2001 that rogue child processes cannot use them.
2003 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2004 run-time user, instead of root.
2006 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2007 Exim run-time user without dropping privileges.
2009 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2010 result string, instead of calling string_vformat() twice with the same
2013 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2014 for other users. Others should always drop root privileges if they use
2015 -C on the command line, even for a whitelisted configure file.
2017 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2019 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2025 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2026 $data_path, and $header_path variables; fixed documentation bugs and
2029 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2030 exipick to access non-standard spools, including the "frozen" queue
2033 NM/01 Bugzilla 965: Support mysql stored procedures.
2034 Patch from Alain Williams
2036 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2038 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2039 Patch from Andreas Metzler
2041 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2042 Patch from Kirill Miazine
2044 NM/05 Bugzilla 671: Added umask to procmail example.
2046 JJ/03 installed exipick 20100323.0, fixing doc bug
2048 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2049 directory. Notification and patch from Dan Rosenberg.
2051 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2053 TK/02 Improve log output when DKIM signing operation fails.
2055 MH/01 Treat the transport option dkim_domain as a colon separated
2056 list, not as a single string, and sign the message with each element,
2057 omitting multiple occurences of the same signer.
2059 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2060 Bugzilla 985, 986. Patch by Simon Arlott
2062 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2063 Patch by Simon Arlott
2065 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2066 MBX locking. Notification from Dan Rosenberg.
2072 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2074 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2076 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2078 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2080 NM/04 Bugzilla 915: Flag broken perl installation during build.
2086 TK/01 Added patch by Johannes Berg that expands the main option
2087 "spamd_address" if it starts with a dollar sign.
2089 TK/02 Write list of recipients to X-Envelope-Sender header when building
2090 the mbox-format spool file for content scanning (suggested by Jakob
2093 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2094 (http://www.dcc-servers.net/) support via dccifd. Activated by
2095 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2097 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2098 by Mark Daniel Reidel <mr@df.eu>.
2100 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2101 When building exim an external PCRE library is now needed -
2102 PCRE is a system library on the majority of modern systems.
2103 See entry on PCRE_LIBS in EDITME file.
2105 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2106 conversation. Added nologin parameter to request.
2107 Patch contributed by Kirill Miazine.
2109 TF/01 Do not log submission mode rewrites if they do not change the address.
2111 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2113 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2114 log files in place. Contributed by Roberto Lima.
2116 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2118 TF/03 Bugzilla 615: When checking the local_parts router precondition
2119 after a local_part_suffix or local_part_prefix option, Exim now
2120 does not use the address's named list lookup cache, since this
2121 contains cached lookups for the whole local part.
2123 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2124 Robert Millan. Documentation is in experimental-spec.txt.
2126 TF/04 Bugzilla 668: Fix parallel build (make -j).
2128 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2130 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2131 Patch provided by Jan Srzednicki.
2133 TF/05 Leading white space used to be stripped from $spam_report which
2134 wrecked the formatting. Now it is preserved.
2136 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2137 that they are available at delivery time.
2139 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2141 TF/08 TLS error reporting now respects the incoming_interface and
2142 incoming_port log selectors.
2144 TF/09 Produce a more useful error message if an SMTP transport's hosts
2145 setting expands to an empty string.
2147 NM/06 Bugzilla 744: EXPN did not work under TLS.
2148 Patch provided by Phil Pennock.
2150 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2151 Patch provided by Richard Godbee.
2153 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2154 acl_smtp_notquit, added index entry.
2156 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2157 Patch provided by Eugene Bujak.
2159 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2160 accept(). Patch provided by Maxim Dounin.
2162 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2163 Patch provided by Phil Pennock.
2165 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2167 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2168 Patch provided by Brad "anomie" Jorsch.
2170 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2171 Patch provided by Dean Brooks.
2173 TK/05 Add native DKIM support (does not depend on external libraries).
2175 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2176 Patch provided by Graeme Fowler.
2178 NM/16 Bugzilla 851: Documentation example syntax fix.
2180 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2182 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2185 NM/19 Bugzilla 745: TLS version reporting.
2186 Patch provided by Phil Pennock.
2188 NM/20 Bugzilla 167: bool: condition support.
2189 Patch provided by Phil Pennock.
2191 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2192 clients. Patch provided by Phil Pennock.
2194 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2195 Patch provided by Brad "anomie" Jorsch.
2197 NM/23 Bugzilla 687: Fix misparses in eximstats.
2198 Patch provided by Heiko Schlittermann.
2200 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2201 Patch provided by Heiko Schlittermann.
2203 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2204 plus update to original patch.
2206 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2208 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2209 Patch provided by David Brownlee.
2211 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2213 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2215 NM/30 Bugzilla 888: TLS documentation bugfixes.
2217 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2219 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2220 Unlike the original bugzilla I have changed all shell scripts in src tree.
2222 NM/33 Bugzilla 898: Transport filter timeout fix.
2223 Patch by Todd Rinaldo.
2225 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2226 Patch by Serge Demonchaux.
2228 NM/35 Bugzilla 39: Base64 decode bug fixes.
2229 Patch by Jakob Hirsch.
2231 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2233 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2235 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2237 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2243 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2244 ALT-N's libdkim that I have put here:
2245 http://duncanthrax.net/exim-experimental/
2247 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2248 were called 'true' and it seems that C99 defines that as a reserved
2249 keyword to be used with 'bool' variable types. That means you could
2250 not include C99-style headers which use bools without triggering
2251 build errors in sieve.c.
2253 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2254 as mailq or other aliases. Changed the --help handling significantly
2255 to do whats expected. exim_usage() emits usage/help information.
2257 SC/01 Added the -bylocaldomain option to eximstats.
2259 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2261 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2263 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2269 PH/01 Another patch from the Sieve maintainer.
2271 PH/02 When an IPv6 address is converted to a string for single-key lookup
2272 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2273 dots are used instead of colons so that keys in lsearch files need not
2274 contain colons. This was done some time before quoting was made available
2275 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2276 (notated using the quote facility) so as to distinguish them from IPv4
2277 keys. This meant that lookups for IP addresses in host lists did not work
2278 for iplsearch lookups.
2280 This has been fixed by arranging for IPv6 addresses to be expressed with
2281 colons if the lookup type is iplsearch. This is not incompatible, because
2282 previously such lookups could never work.
2284 The situation is now rather anomalous, since one *can* have colons in
2285 ordinary lsearch keys. However, making the change in all cases is
2286 incompatible and would probably break a number of configurations.
2288 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2291 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2292 conversion specification without a maximum field width, thereby enabling
2293 a rogue spamd server to cause a buffer overflow. While nobody in their
2294 right mind would setup Exim to query an untrusted spamd server, an
2295 attacker that gains access to a server running spamd could potentially
2296 exploit this vulnerability to run arbitrary code as the Exim user.
2298 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2299 $primary_hostname instead of what libspf2 thinks the hosts name is.
2301 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2302 a directory entry by the name of the lookup key. Previously, if a
2303 symlink pointed to a non-existing file or a file in a directory that
2304 Exim lacked permissions to read, a lookup for a key matching that
2305 symlink would fail. Now it is enough that a matching directory entry
2306 exists, symlink or not. (Bugzilla 503.)
2308 PH/03 The body_linecount and body_zerocount variables are now exported in the
2311 PH/04 Added the $dnslist_matched variable.
2313 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2314 This means they are set thereafter only if the connection becomes
2317 PH/06 Added the client_condition to authenticators so that some can be skipped
2318 by clients under certain conditions.
2320 PH/07 The error message for a badly-placed control=no_multiline_responses left
2321 "_responses" off the end of the name.
2323 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2325 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2326 (without spaces) instead of just copying the configuration text.
2328 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2330 PH/11 Added $max_received_linelength.
2332 PH/12 Added +ignore_defer and +include_defer to host lists.
2334 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2335 way in which PCRE > 7.0 is built.
2337 PH/14 Implemented queue_only_load_latch.
2339 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2340 MAIL command. The effect was to mangle the value on 64-bit systems.
2342 PH/16 Another patch from the Sieve maintainer.
2344 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2346 PH/18 If a system quota error occurred while trying to create the file for
2347 a maildir delivery, the message "Mailbox is full" was not appended to the
2348 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2349 only to a quota excession during the actual writing of the file.
2351 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2352 characters?) which causes problems in log lines. The DN values are now
2353 passed through string_printing() before being added to log lines.
2355 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2356 and InterBase are left for another time.)
2358 PH/21 Added message_body_newlines option.
2360 PH/22 Guard against possible overflow in moan_check_errorcopy().
2362 PH/23 POSIX allows open() to be a macro; guard against that.
2364 PH/24 If the recipient of an error message contained an @ in the local part
2365 (suitably quoted, of course), incorrect values were put in $domain and
2366 $local_part during the evaluation of errors_copy.
2372 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2373 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2376 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2377 issue a MAIL command.
2379 PH/02 In an ACL statement such as
2381 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2383 if a client was not listed at all, or was listed with a value other than
2384 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2385 the condition was not true (as it should be), so access was not denied.
2386 The bug was that the ! inversion was incorrectly passed on to the second
2387 item. This has been fixed.
2389 PH/03 Added additional dnslists conditions == and =& which are different from
2390 = and & when the dns lookup returns more than one IP address.
2392 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2393 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2395 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2396 FSYNC, which compiles an option called disable_fsync that allows for
2397 bypassing fsync(). The documentation is heavily laced with warnings.
2399 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2401 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2402 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2403 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2404 including adding "make clean"; (3) Added -fPIC when compiling the test
2405 dynamically loaded module, to get rid of a warning.
2407 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2408 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2409 = 0s. The bug is otherwise harmless.
2411 PH/07 There was a bug in the dovecot authenticator such that the value of
2412 $auth1 could be overwritten, and so not correctly preserved, after a
2413 successful authentication. This usually meant that the value preserved by
2414 the server_setid option was incorrect.
2416 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2418 PH/09 Installed PCRE release 7.0.
2420 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2421 run for batched SMTP input. It is now run at the start of every message
2422 in the batch. While fixing this I discovered that the process information
2423 (output by running exiwhat) was not always getting set for -bs and -bS
2424 input. This is fixed, and it now also says "batched" for BSMTP.
2426 PH/11 Added control=no_pipelining.
2428 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2429 patch, slightly modified), and move the expansion of helo_data till after
2430 the connection is made in the smtp transport (so it can use these
2433 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2435 PH/14 Added log_selector = +pid.
2437 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2439 PH/16 Add ${if forany and ${if forall.
2441 PH/17 Added dsn_from option to vary the From: line in DSNs.
2443 PH/18 Flush SMTP output before performing a callout, unless control =
2444 no_callout_flush is set.
2446 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2447 was true (the default) a successful delivery failed to delete the retry
2448 item, thus causing premature timeout of the address. The bug is now
2451 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2453 PH/21 Long custom messages for fakedefer and fakereject are now split up
2454 into multiline responses in the same way that messages for "deny" and
2455 other ACL rejections are.
2457 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2458 with slight modification.
2460 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2461 draft, changing the syntax and factoring some duplicate code".
2463 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2464 for deliveries of the second and subsequent messages over the same SMTP
2467 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2468 ${reduce, with only minor "tidies".
2470 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2472 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2473 expansion side effects.
2475 PH/27 When a message times out after an over-quota error from an Exim-imposed
2476 quota, the bounce message says "mailbox is full". This message was not
2477 being given when it was a system quota that was exceeded. It now should
2480 MH/03 Made $recipients available in local_scan(). local_scan() already has
2481 better access to the recipient list through recipients_list[], but
2482 $recipients can be useful in postmaster-provided expansion strings.
2484 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2485 in the case of a MAIL command with additional options following the
2486 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2487 were accidentally chopped off.
2489 PH/29 SMTP synchronization checks are implemented when a command is read -
2490 there is a check that no more input is waiting when there shouldn't be
2491 any. However, for some commands, a delay in an ACL can mean that it is
2492 some time before the response is written. In this time, more input might
2493 arrive, invalidly. So now there are extra checks after an ACL has run for
2494 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2495 pipelining has not been advertised.
2497 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2499 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2500 returned in the SMTP response when smtp_return_error_details was false.
2501 This has been fixed.
2503 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2504 instead of the C I/O that was originally supplied, because problems were
2505 reported on Solaris.
2507 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2508 Exim which did not show up earlier: it was assuming that a call to
2509 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2510 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2511 was a macro that became an assignment, so it seemed to work. This has
2512 changed to a proper function call with a void return, hence the compile
2513 error. Exim's code has been fixed.
2515 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2518 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2520 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2522 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2523 because it turns out to be too big or there is a timeout) there is no
2524 "Completed" line in the log. When some messages of this type were
2525 selected by exigrep, they were listed as "not completed". Others were
2526 picked up by some special patterns. I have improved the selection
2527 criteria to be more general.
2529 PH/38 The host_find_failed option in the manualroute router can now be set
2530 to "ignore", to completely ignore a host whose IP address cannot be
2531 found. If all hosts are ignored, the behaviour is controlled by the new
2532 host_all_ignored option.
2534 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2535 homing or because of multiple MX records with /mx) generated more than
2536 one IP address, and the following item turned out to be the local host,
2537 all the secondary addresses of the first item were incorrectly removed
2538 from the list, along with the local host and any following hosts (which
2539 is what is supposed to happen).
2541 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2542 whoever called Exim into the -H file. In the case of the daemon it was
2543 behaving confusingly. When first started, it used values for whoever
2544 started the daemon, but after a SIGHUP it used the Exim user (because it
2545 calls itself on a restart). I have changed the code so that it now always
2548 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2549 message are rejected with the same error (e.g. no authentication or bad
2550 sender address), and a DATA command is nevertheless sent (as can happen
2551 with PIPELINING or a stupid MUA), the error message that was given to the
2552 RCPT commands is included in the rejection of the DATA command. This is
2553 intended to be helpful for MUAs that show only the final error to their
2556 PH/42 Another patch from the Sieve maintainer.
2558 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2559 Eximstats - Fixed some broken HTML links and added missing column headers
2561 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2562 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2564 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2566 PH/43 Yet another patch from the Sieve maintainer.
2568 PH/44 I found a way to check for a TCP/IP connection going away before sending
2569 the response to the final '.' that terminates a message, but only in the
2570 case where the client has not sent further data following the '.'
2571 (unfortunately, this is allowed). However, in many cases there won't be
2572 any further data because there won't be any more messages to send. A call
2573 to select() can be used: if it shows that the input is "ready", there is
2574 either input waiting, or the socket has been closed. An attempt to read
2575 the next input character can distinguish the two cases. Previously, Exim
2576 would have sent an OK response which the client would never have see.
2577 This could lead to message repetition. This fix should cure that, at
2578 least in a lot of common cases.
2580 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2581 advertised in response to EHLO.
2587 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2588 fixed by 4.65/MH/01 (is this a record?) are fixed:
2590 (i) An empty string was always treated as zero by the numeric comparison
2591 operators. This behaviour has been restored.
2593 (ii) It is documented that the numeric comparison operators always treat
2594 their arguments as decimal numbers. This was broken in that numbers
2595 starting with 0 were being interpreted as octal.
2597 While fixing these problems I realized that there was another issue that
2598 hadn't been noticed. Values of message_size_limit (both the global option
2599 and the transport option) were treated as octal if they started with 0.
2600 The documentation was vague. These values are now always treated as
2601 decimal, and I will make that clear in the documentation.
2607 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2608 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2611 MH/01 Don't check that the operands of numeric comparison operators are
2612 integers when their expansion is in "skipping" mode (fixes bug
2613 introduced by 4.64-PH/07).
2615 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2616 child addresses, Exim now panics and dies. Previously, because the count
2617 is held in a short int, deliveries were likely to be lost. As such a
2618 large number of recipients for a single message is ridiculous
2619 (performance will be very, very poor), I have chosen to impose a limit
2620 rather than extend the field.
2626 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2627 leftover -K file (the existence of which was triggered by #402).
2628 While we were at it, introduced process PID as part of the -K
2629 filename. This should rule out race conditions when creating
2632 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2633 processing considerably. Previous code took too long for large mails,
2634 triggering a timeout which in turn triggers #401.
2636 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2637 in the DK code in transports.c. sendfile() is not really portable,
2638 hence the _LINUX specificness.
2640 TF/01 In the add_headers option to the mail command in an Exim filter,
2641 there was a bug that Exim would claim a syntax error in any
2642 header after the first one which had an odd number of characters
2645 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2646 callout verification, Exim cached a "reject" for the entire domain. This
2647 is correct for most verifications, but it is not correct for a recipient
2648 verification with use_sender or use_postmaster set, because in that case
2649 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2650 case of MAIL FROM:<> rejection from other early rejections (e.g.
2651 rejection of HELO). When verifying a recipient using a non-null MAIL
2652 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2653 Whatever the result of the callout, the value of the domain cache is
2654 left unchanged (for any other kind of callout, getting as far as trying
2655 RCPT means that the domain itself is ok).
2657 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2660 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2661 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2662 session, the daemon ignores ECONNECT errors and logs others; it now
2663 ignores EPIPE as well.
2665 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2666 (quoted-printable decoding).
2668 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2669 later the small subsequent patch to fix an introduced bug.
2671 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2673 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2675 PH/08 An error is now given if message_size_limit is specified negative.
2677 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2678 to be given (somewhat) arbitrary names.
2680 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2683 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2684 miscellaneous code fixes
2686 PH/10 Added the log_reject_target ACL modifier to specify where to log
2689 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2690 hostname. This is wrong, because it relates to the incoming message (and
2691 probably the interface on which it is arriving) and not to the outgoing
2692 callout (which could be using a different interface). This has been
2693 changed to use the value of the helo_data option from the smtp transport
2694 instead - this is what is used when a message is actually being sent. If
2695 there is no remote transport (possible with a router that sets up host
2696 addresses), $smtp_active_hostname is used.
2698 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2699 tweaks were necessary in order to get it to work (see also 21 below):
2700 (a) The code assumed that strncpy() returns a negative number on buffer
2701 overflow, which isn't the case. Replaced with Exim's string_format()
2703 (b) There were several signed/unsigned issues. I just did the minimum
2704 hacking in of casts. There is scope for a larger refactoring.
2705 (c) The code used strcasecmp() which is not a standard C function.
2706 Replaced with Exim's strcmpic() function.
2707 (d) The code set only $1; it now sets $auth1 as well.
2708 (e) A simple test gave the error "authentication client didn't specify
2709 service in request". It would seem that Dovecot has changed its
2710 interface. Fortunately there's a specification; I followed it and
2711 changed what the client sends and it appears to be working now.
2713 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2716 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2717 address A is aliased to B and C, where B exists and C does not. Without
2718 -v the output is "A verified" because verification stops after a
2719 successful redirection if more than one address is generated. However,
2720 with -v the child addresses are also verified. Exim was outputting "A
2721 failed to verify" and then showing the successful verification for C,
2722 with its parentage. It now outputs "B failed to verify", showing B's
2723 parentage before showing the successful verification of C.
2725 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2726 look up a TXT record in a specific list after matching in a combined
2729 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2730 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2731 they consult the DNS. I had assumed they would set it the way they
2732 wanted; and indeed my experiments on Linux seem to show that in some
2733 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2734 To be on the safe side, however, I have now made the interface to
2735 host_find_byname() similar to host_find_bydns(), with an argument
2736 containing the DNS resolver options. The host_find_byname() function now
2737 sets these options at its start, just as host_find_bydns() does. The smtp
2738 transport options dns_qualify_single and dns_search_parents are passed to
2739 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2740 of host_find_byname() use the default settings of RES_DEFNAMES
2741 (qualify_single) but not RES_DNSRCH (search_parents).
2743 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2744 spool_read_header() do less string testing, by means of a preliminary
2745 switch on the second character of optional "-foo" lines. (This is
2746 overdue, caused by the large number of possibilities that now exist.
2747 Originally there were few.) While I was there, I also converted the
2748 str(n)cmp tests so they don't re-test the leading "-" and the first
2749 character, in the hope this might squeeze out yet more improvement.
2751 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2752 flag allowing group syntax was set by the header_syntax check but not
2753 turned off, possible causing trouble later; (2) The flag was not being
2754 set at all for the header_verify test, causing "group"-style headers to
2755 be rejected. I have now set it in this case, and also caused header_
2756 verify to ignore an empty address taken from a group. While doing this, I
2757 came across some other cases where the code for allowing group syntax
2758 while scanning a header line wasn't quite right (mostly, not resetting
2759 the flag correctly in the right place). These bugs could have caused
2760 trouble for malformed header lines. I hope it is now all correct.
2762 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2763 with the "reply" argument non-NULL. The code, however (which originally
2764 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2765 but it didn't always do it. This confused somebody who was copying the
2766 code for some other use. I have removed all the tests.
2768 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2769 feature that was used to support insecure browsers during the U.S. crypto
2770 embargo. It requires special client support, and Exim is probably the
2771 only MTA that supported it -- and would never use it because real RSA is
2772 always available. This code has been removed, because it had the bad
2773 effect of slowing Exim down by computing (never used) parameters for the
2774 RSA_EXPORT functionality.
2776 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2777 authenticator to fail if there's a tab character in the incoming data
2778 (there should never be unless someone is messing about, as it's supposed
2779 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2780 if the connection is using TLS or if the remote IP is the same as the
2781 local IP, and the "valid-client-cert option" if a client certificate has
2784 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2785 authenticators. This can be used for authorization after authentication
2786 succeeds. (In the case of plaintext, it servers for both authentication
2789 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2790 if any retry times were supplied.
2792 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2793 connection, where there is no client IP address to check. In this
2794 situation, the verify now always succeeds.
2796 PH/25 Applied John Jetmore's -Mset patch.
2798 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2800 PH/27 In a string expansion for a processed (not raw) header when multiple
2801 headers of the same name were present, leading whitespace was being
2802 removed from all of them, but trailing whitespace was being removed only
2803 from the last one. Now trailing whitespace is removed from each header
2804 before concatenation. Completely empty headers in a concatenation (as
2805 before) are ignored.
2807 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2808 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2810 PH/29 [Removed. This was a change that I later backed out, and forgot to
2811 correct the ChangeLog entry (that I had efficiently created) before
2812 committing the later change.]
2814 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2815 address errors (4xx response to RCPT) over the same connection as other
2816 messages routed to the same hosts. Such deliveries are always "forced",
2817 so retry times are not inspected. This resulted in far too many retries
2818 for the affected addresses. The effect occurred only when there were more
2819 hosts than the hosts_max_try setting in the smtp transport when it had
2820 the 4xx errors. Those hosts that it had tried were not added to the list
2821 of hosts for which the message was waiting, so if all were tried, there
2822 was no problem. Two fixes have been applied:
2824 (i) If there are any address or message errors in an SMTP delivery, none
2825 of the hosts (tried or untried) are now added to the list of hosts
2826 for which the message is waiting, so the message should not be a
2827 candidate for sending over the same connection that was used for a
2828 successful delivery of some other message. This seems entirely
2829 reasonable: after all the message is NOT "waiting for some host".
2830 This is so "obvious" that I'm not sure why it wasn't done
2831 previously. Hope I haven't missed anything, but it can't do any
2832 harm, as the worst effect is to miss an optimization.
2834 (ii) If, despite (i), such a delivery is accidentally attempted, the
2835 routing retry time is respected, so at least it doesn't keep
2836 hammering the server.
2838 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2839 in ${readsocket because some servers need this prod.
2841 PH/32 Added some extra debug output when updating a wait-xxx database.
2843 PH/33 The hint "could be header name not terminated by colon", which has been
2844 given for certain expansion errors for a long time, was not being given
2845 for the ${if def:h_colon_omitted{... case.
2847 PH/34 The spec says: "With one important exception, whenever a domain list is
2848 being scanned, $domain contains the subject domain." There was at least
2849 one case where this was not true.
2851 PH/35 The error "getsockname() failed: connection reset by peer" was being
2852 written to the panic log as well as the main log, but it isn't really
2853 panic-worthy as it just means the connection died rather early on. I have
2854 removed the panic log writing for the ECONNRESET error when getsockname()
2857 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2858 runs only) independently of the message's sender address. This meant
2859 that, if the 4xx error was in fact related to the sender, a different
2860 message to the same recipient with a different sender could confuse
2861 things. In particular, this can happen when sending to a greylisting
2862 server, but other circumstances could also provoke similar problems.
2863 I have changed the default so that the retry time for these errors is now
2864 based a combination of the sender and recipient addresses. This change
2865 can be overridden by setting address_retry_include_sender=false in the
2868 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2869 remote server are returned as part of bounce messages. This was not
2870 happening for LMTP over a pipe (the lmtp transport), but now it is the
2871 same for both kinds of LMTP.
2873 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2874 in header lines that were in fact CNAMEs. This is no longer the case.
2876 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2877 and queue runs started by the daemon processed all messages. This has
2878 been fixed so that -R and -S can now usefully be given with -q<time>.
2880 PH/40 Import PCRE release 6.7 (fixes some bugs).
2882 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2884 PH/42 Give an error if -q is specified more than once.
2886 PH/43 Renamed the variables $interface_address and $interface_port as
2887 $received_ip_address and $received_port, to make it clear that these
2888 values apply to message reception, and not to the outgoing interface when
2889 a message is delivered. (The old names remain recognized, of course.)
2891 PH/44 There was no timeout on the connect() call when using a Unix domain
2892 socket in the ${readsocket expansion. There now is.
2894 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2895 be meaningful with "accept".
2897 SC/01 Eximstats V1.43
2898 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2900 SC/02 Eximstats V1.44
2901 Use a glob alias rather than an array ref in the generated
2902 parser. This improves both readability and performance.
2904 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2905 Collect SpamAssassin and rejection statistics.
2906 Don't display local sender or destination tables unless
2907 there is data to show.
2908 Added average volumes into the top table text output.
2910 SC/04 Eximstats V1.46
2911 Collect data on the number of addresses (recipients)
2912 as well as the number of messages.
2914 SC/05 Eximstats V1.47
2915 Added 'Message too big' to the list of mail rejection
2916 reasons (thanks to Marco Gaiarin).
2918 SC/06 Eximstats V1.48
2919 Mainlog lines which have GMT offsets and are too short to
2920 have a flag are now skipped.
2922 SC/07 Eximstats V1.49 (Alain Williams)
2923 Added the -emptyok flag.
2925 SC/08 Eximstats V1.50
2926 Fixes for obtaining the IP address from reject messages.
2928 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2929 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2930 whitespace changes from 4.64-PH/27
2932 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2935 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2936 are found, allow negative numbers in numeric criteria)
2938 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2940 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2941 to match changes made in 4.64-PH/43
2943 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2945 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2946 "domain" is actually an IP address, and force a failure. This locks out
2947 those revolvers/nameservers that support "A-for-A" lookups, in
2948 contravention of the specifications.
2950 PH/48 When a host name was looked up from an IP address, and the subsequent
2951 forward lookup of the name timed out, the host name was left in
2952 $sender_host_name, contrary to the specification.
2954 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2955 restricted to single-key lookups, Exim was not diagnosing an error if
2956 * or *@ was used with a query-style lookup.
2958 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2960 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2961 long ago, but noone interested enough thought of it. Let's just say that
2962 the "1.1" means that there are some new functions that weren't there at
2963 some point in the past.
2965 PH/51 Error processing for expansion failure of helo_data from an smtp
2966 transport during callout processing was broken.
2968 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2969 tested/used via the -bh/-bhc/-bs options.
2971 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2972 bug, fixed in subsequent PCRE releases).
2974 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2975 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2977 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2983 SC/01 Use a glob alias rather than an array ref in eximstats generated
2984 parser. This improves both readability and performance.
2986 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2987 Don't display local sender or destination tables in eximstats unless
2988 there is data to show.
2989 Added average volumes into the eximstats top table text output.
2991 SC/03 Collect data on the number of addresses (recipients) as well
2992 as the number of messages in eximstats.
2994 TF/01 Correct an error in the documentation for the redirect router. Exim
2995 does (usually) call initgroups() when daemonizing.
2997 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2998 with consistent privilege compared to when running as a daemon.
3000 TF/03 Note in the spec that $authenticated_id is not set for local
3001 submissions from trusted users.
3003 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3004 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3006 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3007 by adding some example configuration directives to the default
3008 configuration file. A little bit of work is required to uncomment the
3009 directives and define how usernames and passwords are checked, but
3010 there is now a framework to start from.
3012 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3013 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3014 without this. I don't know how relevant this is to other LDAP libraries.
3016 PH/02 Add the verb name to the "unknown ACL verb" error.
3018 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3020 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3022 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3023 directory not expanded when it should be if an expanded home directory
3024 was set for the address (which is overridden by the transport).
3026 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3029 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3030 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3031 because it is too late at that time, and has no effect.
3033 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3034 security issue with \' (bugzilla #107). I could not use the
3035 PQescapeStringConn() function, because it needs a PGconn value as one of
3038 PH/08 When testing addresses using -bt, indicate those final addresses that
3039 are duplicates that would not cause an additional delivery. At least one
3040 person was confused, thinking that -bt output corresponded to deliveries.
3041 (Suppressing duplicates isn't a good idea as you lose the information
3042 about possibly different redirections that led to the duplicates.)
3044 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3045 systems where poll() doesn't work, in particular OS X.
3047 PH/10 Added more information to debugging output for retry time not reached.
3049 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3050 operations in malware.c.
3052 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3055 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3056 syslog_duplication set false, log lines that would normally be written
3057 both the the main log and to the reject log were not written to syslog at
3060 PH/14 In the default configuration, change the use of "message" in ACL warn
3061 statements to "add_header".
3063 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3064 not followed by a command (e.g. "seen endif").
3066 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3067 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3070 PH/17 Added extra conditions to the default value of delay_warning_condition
3074 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3075 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3076 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3079 The Auto-Submitted: and various List- headers are standardised, whereas I
3080 don't think Precedence: ever was.
3082 PH/18 Refactored debugging code in route_finduser() to show more information,
3083 in particular, the error code if getpwnam() issues one.
3085 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3086 This is apparently needed in addition to the PH/07 change above to avoid
3087 any possible encoding problems.
3089 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3090 but not after initializing Perl.
3092 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3093 output them only if debugging. By default they are written stderr,
3094 apparently, which is not desirable.
3096 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3099 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3102 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3104 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3105 authenticated or an ident call has been made. Suppress the default
3106 values for $authenticated_id and $authenticated_sender (but permit -oMai
3107 and -oMas) when testing with -bh.
3109 PH/24 Re-jigged the order of the tests in the default configuration so that the
3110 tests for valid domains and recipients precede the DNS black list and CSA
3111 tests, on the grounds that those ones are more expensive.
3113 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3114 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3115 This bug exists in every version of Exim that I still have, right back to
3118 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3119 However, an attempt to turn on case-sensitivity in a regex key by
3120 including (?-i) didn't work because the subject string was already
3121 lowercased, and the effects were non-intuitive. It turns out that a
3122 one-line patch can be used to allow (?-i) to work as expected.
3128 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3129 other effects) broke the use of negated acl sub-conditions.
3131 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3134 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3135 "Deny" causes Exim to reject the incoming connection with a 554 error.
3136 Unfortunately, if there is a major crisis, such as a disk failure,
3137 tcp-wrappers gives "deny", whereas what one would like would be some
3138 kind of temporary error. A kludge has been added to help with this.
3139 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3140 554 error is used if errno is still zero or contains ENOENT (which occurs
3141 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3144 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3146 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3147 errors. Otherwise a message that provokes a temporary error (when other
3148 messages do not) can cause a whole host to time out.
3150 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3151 addresses were routed directly to files or pipes from a redirect router.
3152 File deliveries just didn't batch; pipe deliveries might have suffered
3155 PH/06 A failure to get a lock for a hints database would erroneously always say
3156 "Failed to get write lock", even when it was really a read lock.
3158 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3159 of 0600. This has been changed to use the value of the lockfile_mode
3160 option (which defaults to 0600).
3162 PH/08 Applied small patch from the Sieve maintainer.
3164 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3165 folder from quota calculations, a direct delivery into this folder messed
3166 up the contents of the maildirsize file. This was because the regex was
3167 used only to exclude .Trash (or whatever) when the size of the mailbox
3168 was calculated. There was no check that a delivery was happening into an
3169 excluded directory. This bug has been fixed by ignoring all quota
3170 processing for deliveries into excluded directories.
3172 PH/10 Added the maildirfolder_create_regex option to appendfile.
3178 PH/01 The code for finding all the local interface addresses on a FreeBSD
3179 system running IPv6 was broken. This may well have applied to all BSD
3180 systems, as well as to others that have similar system calls. The broken
3181 code found IPv4 interfaces correctly, but gave incorrect values for the
3182 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3183 that it would not match correctly against @[] and not recognize the IPv6
3186 PH/02 The ipliteral router was not recognizing addresses of the form user@
3187 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3189 PH/03 Added disable_ipv6.
3191 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3192 decoded form, because it is most often used to construct To: headers
3193 lines in autoreplies, and the decoded form may well be syntactically
3194 invalid. However, $reply_address has leading white space removed, and all
3195 newlines turned into spaces so that the autoreply transport does not
3198 PH/05 If group was specified without a user on a router, and no group or user
3199 was specified on a transport, the group from the router was ignored.
3201 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3202 for visible compile-time settings that can be used to change these
3203 numbers, for those that want even more. Backwards compatibility with old
3204 spool files has been maintained. However, going back to a previous Exim
3205 release will lost any variables that are in spool files.
3207 PH/07 Two small changes when running in the test harness: increase delay when
3208 passing a TCP/IP connection to a new process, in case the original
3209 process has to generate a bounce, and remove special handling of
3210 127.0.0.2 (sic), which is no longer necessary.
3212 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3213 be the same on different OS.
3215 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3218 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3219 whether --show-vars was specified or not
3221 JJ/02 exipick: Added support for new ACL variable spool format introduced
3224 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3225 syntactically invalid From: or Reply-to: line, and a filter used this to
3226 generate an autoreply, and therefore failed to obtain an address for the
3227 autoreply, Exim could try to deliver to a non-existent relative file
3228 name, causing unrelated and misleading errors. What now happens is that
3229 it logs this as a hard delivery error, but does not attempt to create a
3232 PH/11 The exinext utility has a -C option for testing purposes, but although
3233 the given file was scanned by exinext itself; it wasn't being passed on
3234 when Exim was called.
3236 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3237 an end-of-file indication when reading a command response.
3239 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3240 compiled. In many other places in Exim, IPv6 addresses are always
3241 recognized, so I have changed this. It also means that IPv4 domain
3242 literals of the form [IPV4:n.n.n.n] are now always recognized.
3244 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3245 used if the router is not running as root, for example, when verifying at
3246 ACL time, or when using -bh. The debugging output from this situation was
3247 non-existent - all you got was a failure to exec. I have made two
3250 (a) Failures to set uid/gid, the current directory, or a process leader
3251 in a subprocess such as that created by queryprogram now generate
3252 suitable debugging output when -d is set.
3254 (b) The queryprogram router detects when it is not running as root,
3255 outputs suitable debugging information if -d is set, and then runs
3256 the subprocess without attempting to change uid/gid.
3258 PH/15 Minor change to Makefile for building test_host (undocumented testing
3261 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3262 additional section of a DNS packet that returns MX or SRV records.
3263 Instead, it always explicitly searches for A/AAAA records. This avoids
3264 major problems that occur when a DNS server includes only records of one
3265 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3266 fixed another bug: if SRV records were looked up and the corresponding
3267 address records were *not* found in the additional section, the port
3268 values from the SRV records were lost.
3270 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3271 using the correct key (the original address) when searching the retry
3272 rules in order to find which one to use for generating the retry hint.
3274 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3275 adding the default one. Similarly, if it contains a Reply-To: header, the
3276 errors_reply_to option, if set, is not used.
3278 PH/19 When calculating a retry time, Exim used to measure the "time since
3279 failure" by looking at the "first failed" field in the retry record. Now
3280 it does not use this if it is later than than the arrival time of the
3281 message. Instead it uses the arrival time. This makes for better
3282 behaviour in cases where some deliveries succeed, thus re-setting the
3283 "first failed" field. An example is a quota failure for a huge message
3284 when small messages continue to be delivered. Without this change, the
3285 "time since failure" will always be short, possible causing more frequent
3286 delivery attempts for the huge message than are intended.
3287 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3289 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3290 $1, $2, $3) because the numerical variables can be reset during some
3291 expansion items (e.g. "match"), thereby losing the authentication data.
3293 PH/21 Make -bV show the size of off_t variables so that the test suite can
3294 decide whether to run tests for quotas > 2G.
3296 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3297 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3298 filecount value is greater than 2G or if a quota value is greater than 2G
3299 on a system where the size of off_t is not greater than 4, a panic error
3302 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3303 never match. The debug and -bh output now contains an explicit error
3304 message indicating a malformed IPv4 address or mask.
3306 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3307 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3308 PH/23 above applies.
3310 PH/25 Do not write to syslog when running in the test harness. The only
3311 occasion when this arises is a failure to open the main or panic logs
3312 (for which there is an explicit test).
3314 PH/26 Added the /no_tell option to "control=freeze".
3316 PH/27 If a host name lookup failed very early in a connection, for example, if
3317 the IP address matched host_lookup and the reverse lookup yielded a name
3318 that did not have a forward lookup, an error message of the form "no IP
3319 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3320 could be logged. Now it outputs the IP address instead of "NULL".
3322 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3323 allows the sender and the authenticated sender to be set when
3324 submitting a message from within Exim. Since child_open_exim() is
3325 documented for local_scan(), the new function should be too.
3327 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3328 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3329 results in an empty string is now treated as unset.
3331 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3333 PH/31 Added sender_verify_fail logging option.
3335 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3336 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3337 tidied the source and removed it altogether.
3339 PH/33 When a queue run was abandoned because the load average was too high, a
3340 log line was always written; now it is written only if the queue_run log
3341 selector is set. In addition, the log line for abandonment now contains
3342 information about the queue run such as the pid. This is always present
3343 in "start" and "stop" lines but was omitted from the "abandon" line.
3345 PH/34 Omit spaces between a header name and the colon in the error message that
3346 is given when verify = headers_syntax fails (if there are lots of them,
3347 the message gets confusing).
3349 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3350 names, as there are now some PTR records that contain slashes. This check
3351 is only to protect against broken name servers that fall over on strange
3352 characters, so the fact that it applies to all lookups doesn't matter.
3354 PH/36 Now that the new test suite is complete, we can remove some of the
3355 special code in Exim that was needed for the old test suite. For example,
3356 sorting DNS records because real resolvers return them in an arbitrary
3357 order. The new test suite's fake resolver always returns records in the
3360 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3361 bounces) except when queue_only is set, to avoid logging races between
3362 the different processes.
3364 PH/38 Panic-die if .include specifies a non-absolute path.
3366 PH/39 A tweak to the "H" retry rule from its user.
3368 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3369 a label. They prevented compilation on older perls.
3371 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3372 a warning to be raised on newish perls.
3374 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3375 on queue. Changes to match documented behaviour of showing count of
3376 messages matching specified criteria.
3378 PH/40 Changed the default ident timeout from 30s to 5s.
3380 PH/41 Added support for the use of login_cap features, on those BSD systems
3381 that have them, for controlling the resources used by pipe deliveries.
3383 PH/42 The content-scanning code uses fopen() to create files in which to put
3384 message data. Previously it was not paying any attention to the mode of
3385 the files. Exim runs with umask(0) because the rest of the code creates
3386 files with open(), and sets the required mode explicitly. Thus, these
3387 files were ending up world-writeable. This was not a big issue, because,
3388 being within the spool directory, they were not world-accessible. I have
3389 created a function called modefopen, which takes an additional mode
3390 argument. It sets umask(777), creates the file, chmods it to the required
3391 mode, then resets the umask. All the relevant calls to fopen() in the
3392 content scanning code have been changed to use this function.
3394 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3395 to 24 hours. This avoids potential overflow problems when processing G
3396 and H retry rules. I suspect nobody ever tinkers with this value.
3398 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3400 PH/45 When the plaintext authenticator is running as a client, the server's
3401 challenges are checked to ensure they are valid base64 strings. By
3402 default, the authentication attempt is cancelled if an invalid string is
3403 received. Setting client_ignore_invalid_base64 true ignores these errors.
3404 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3405 they are received. Thus, the responses can be made to depend on the
3406 challenges. If an invalid string is ignored, an empty string is placed in
3409 PH/46 Messages that are created by the autoreply transport now contains a
3410 References: header, in accordance with RFCs 2822 and 3834.
3412 PH/47 Added authenticated_sender_force to the smtp transport.
3414 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3416 PH/49 Installed latest patch from the Sieve maintainer.
3418 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3419 was also set, the appendfile transport was unnecessarily scanning a
3420 directory of message files (e.g. for maildir delivery) to find the count
3421 of files (along with the size), even though it did not need this
3422 information. It now does the scan only if it needs to find either the
3423 size of the count of files.
3425 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3427 PH/52 Two bugs concerned with error handling when the smtp transport is
3430 (i) Exim was not creating retry information for temporary errors given
3431 for individual recipients after the DATA command when the smtp transport
3432 was used in LMTP mode. This meant that they could be retried too
3433 frequently, and not timed out correctly.
3435 (ii) Exim was setting the flag that allows error details to be returned
3436 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3437 recipients that were returned after the DATA command.
3439 PH/53 This is related to PH/52, but is more general: for any failing address,
3440 when detailed error information was permitted to be returned to the
3441 sender, but the error was temporary, then after the final timeout, only
3442 "retry timeout exceeded" was returned. Now it returns the full error as
3443 well as "retry timeout exceeded".
3445 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3446 do this, and (what is worse) MTAs that accept it.
3448 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3449 will now be deprecated.
3451 PH/56 New os.c-cygwin from the Cygwin maintainer.
3453 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3454 formats (previously only available in exim formats via -bpr, -bpru,
3455 and -bpra. Now also available in native and exiqgrep formats)
3457 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3458 with very large, slow to parse queues
3460 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3462 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3464 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3465 responses to authentication challenges, though it was showing the
3466 challenges; (ii) I've removed the CR characters from the debug output for
3469 PH/58 Allow for the insertion of a newline as well as a space when a string
3470 is turned into more than one encoded-word during RFC 2047 encoding. The
3471 Sieve code now uses this.
3473 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3474 data_4xx, lost_connection, tls_required.
3476 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3477 message was being sent as an SMTP response.
3479 PH/61 Add -l and -k options to exicyclog.
3481 PH/62 When verifying, if an address was redirected to one new address, so that
3482 verification continued, and the new address failed or deferred after
3483 having set something in $address_data, the value of $address_data was not
3484 passed back to the ACL. This was different to the case when no
3485 redirection occurred. The value is now passed back in both cases.
3487 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3488 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3489 use login_cap.h, so on its own it isn't the distinguishing feature. The
3490 new name refers directly to the setclassresources() function.
3492 PH/65 Added configuration files for NetBSD3.
3494 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3496 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3497 is preferred over IPv4.
3499 PH/68 The bounce_return_message and bounce_return_body options were not being
3500 honoured for bounces generated during the reception of non-SMTP messages.
3501 In particular, this applied to messages rejected by the ACL. This bug has
3502 been fixed. However, if bounce_return_message is true and bounce_return_
3503 body is false, the headers that are returned for a non-SMTP message
3504 include only those that have been read before the error was detected.
3505 (In the case of an ACL rejection, they have all been read.)
3507 PH/69 The HTML version of the specification is now built in a directory called
3508 spec_html instead of spec.html, because the latter looks like a path with
3509 a MIME-type, and this confuses some software.
3511 PH/70 Catch two compiler warnings in sieve.c.
3513 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3514 function verify_get_ident() calls ip_connect() to connect a socket, but
3515 if the "connect()" function timed out, ip_connect() used to close the
3516 socket. However, verify_get_ident() also closes the socket later, and in
3517 between Exim writes to the log, which may get opened at this point. When
3518 the socket was closed in ip_connect(), the log could get the same file
3519 descriptor number as the socket. This naturally causes chaos. The fix is
3520 not to close the socket in ip_connect(); the socket should be closed by
3521 the function that creates it. There was only one place in the code where
3522 this was missing, in the iplookup router, which I don't think anybody now
3523 uses, but I've fixed it anyway.
3525 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3526 well as to direct DNS lookups. Otherwise the handling of names in host
3527 lists is inconsistent and therefore confusing.
3533 PH/01 Two changes to the default runtime configuration:
3535 (1) Move the checks for relay_from_hosts and authenticated clients from
3536 after to before the (commented out) DNS black list checks.
3538 (2) Add control=submission to the relay_from_hosts and authenticated
3539 clients checks, on the grounds that messages accepted by these
3540 statements are most likely to be submissions.
3542 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3544 (1) Generate an error if the third argument for the ${prvs expansion is
3547 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3550 (3) Reset the variables that are obtained from the first argument of
3551 ${prvscheck and used in the second argument before leaving the code,
3552 because their memory is reclaimed, so using them afterwards may do
3555 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3556 one (it's much easier than Tom thought :-).
3558 (5) Because of (4), we can now allow for the use of $prvscheck_result
3559 inside the third argument.
3561 PH/03 For some reason, the default setting of PATH when running a command from
3562 a pipe transport was just "/usr/bin". I have changed it to
3565 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3566 anything to be listed in the output from -bV.
3568 PH/05 When a filter generated an autoreply, the entire To: header line was
3569 quoted in the delivery log line, like this:
3571 => >A.N.Other <ano@some.domain> <original@ddress> ...
3573 This has been changed so that it extracts the operative address. There
3574 may be more than one such address. If so, they are comma-separated, like
3577 => >ano@some.domain,ona@other.domain <original@ddress> ...
3579 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3580 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3581 not being looked up in the rDNS to get a host name, Exim was showing the
3582 IP address twice in Received: lines, even though the IP addresses were
3583 identical. For example:
3585 Received: from [1.2.3.4] (helo=[1.2.3.4])
3587 However, if the real host name was known, it was omitting the HELO data
3588 if it matched the actual IP address. This has been tidied up so that it
3589 doesn't show the same IP address twice.
3591 PH/07 When both +timestamp and +memory debugging was on, the value given by
3592 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3593 called by the debug printing, thereby overwriting the timestamp buffer.
3594 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3596 PH/08 When the original message was included in an autoreply transport, it
3597 always said "this is a copy of the message, including all the headers",
3598 even if body_only or headers_only was set. It now gives an appropriate
3601 PH/09 Applied a patch from the Sieve maintainer which:
3603 o fixes some comments
3604 o adds the (disabled) notify extension core
3605 o adds some debug output for the result of if/elsif tests
3606 o points to the current vacation draft in the documentation
3607 and documents the missing references header update
3611 o fixes a bug in processing the envelope test (when testing
3612 multiple envelope elements, the last element determined the
3615 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3616 Electronic Mail") by including:
3618 Auto-submitted: auto-generated
3620 in the messages that it generates (bounce messages and others, such as
3621 warnings). In the case of bounce messages for non-SMTP messages, there was
3622 also a typo: it was using "Auto_submitted" (underscore instead of
3623 hyphen). Since every message generated by Exim is necessarily in response
3624 to another message, thes have all been changed to:
3626 Auto-Submitted: auto-replied
3628 in accordance with these statements in the RFC:
3630 The auto-replied keyword:
3632 - SHOULD be used on messages sent in direct response to another
3633 message by an automatic process,
3635 - MUST NOT be used on manually-generated messages,
3637 - MAY be used on Delivery Status Notifications (DSNs) and Message
3638 Disposition Notifications (MDNs),
3640 - MUST NOT be used on messages generated by automatic or periodic
3641 processes, except for messages which are automatic responses to
3644 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3645 to the default Received: header definition.
3647 PH/12 Added log selector acl_warn_skipped (default on).
3649 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3650 variables because (a) they are in the wrong storage pool and (b) even if
3651 they were copied, it wouldn't work properly because of the caching.
3653 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3654 checking when decoding. Apparently there are clients that generate
3655 overlong encoded strings. Why am I not surprised?
3657 PH/15 If the first argument of "${if match_address" was not empty, but did not
3658 contain an "@" character, Exim crashed. Now it writes a panic log message
3659 and treats the condition as false.
3661 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3663 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3664 extension "envelope-auth". The code is finished and in agreement with
3665 other implementations, but there is no documentation so far and in fact,
3666 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3667 not changing the active code.
3669 Print executed "if" and "elsif" statements when debugging is used. This
3670 helps a great deal to understand what a filter does.
3672 Document more things not specified clearly in RFC3028. I had all this
3673 sorted out, when out of a sudden new issues came to my mind. Oops."
3675 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3678 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3679 canonical form (as documented). However, after a host name lookup from
3680 the IP address, check_host() was doing a simple string comparison with
3681 addresses acquired from the DNS when checking that the found name did
3682 have the original IP as one of its addresses. Since any found IPv6
3683 addresses are likely to be in abbreviated form, the comparison could
3684 fail. Luckily, there already exists a function for doing the comparison
3685 by converting both addresses to binary, so now that is used instead of
3686 the text comparison.
3688 PH/20 There was another similar case to PH/19, when a complete host name was
3689 given in a host list; looking up its IP address could give an abbreviated
3690 form, whereas the current host's name might or might not be abbreviated.
3691 The same fix has been applied.
3697 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3698 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3701 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3702 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3704 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3706 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3707 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3708 Also, for "Auto-Submitted", treat anything other than "no" as
3709 non-personal, in accordance with RFC 3834. (Previously it treated
3710 anything starting "auto-" as non-personal.)
3712 TF/01 The control=submission/name=... option had a problem with syntax
3713 errors if the name included a slash character. The /name= option
3714 now slurps the rest of the string, so it can include any characters
3715 but it must come last in the list of options (after /sender_retain
3718 PH/05 Some modifications to the interface to the fake nameserver for the new
3726 TK/01 Added the "success_on_redirect" address verification option. See
3727 NewStuff for rationale and an example.
3729 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3731 PH/02 Patch to exigrep to allow it to work on syslog lines.
3733 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3734 fread() to skip over the body file's header line, because in Cygwin the
3735 header line is locked and is inaccessible.
3737 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3738 co-exist for some time) to make it clear that it is the Exim ID that is
3739 referenced, not the Message-ID: header line.
3741 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3742 string_format() function, because snprintf() does not exist on all
3745 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3746 ${stat: expansion item.
3748 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3749 protocol synchronization error", to keep the pedants happy.
3751 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3752 well as for IRIX systems, when gcc is being used. See the host.c source
3755 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3757 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3760 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3761 transport and to the smtp transport in LMTP mode.
3763 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3765 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3766 run a filter in a subprocess. This could lead to confusion in subsequent
3767 lookups in the parent process. There should also be a search_tidyup() at
3768 the end of the subprocess.
3770 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3771 only if the host matched helo_try_verify_hosts, which caused the
3772 verification to occur when the EHLO/HELO command was issued. The ACL just
3773 tested the remembered result. Now, if a previous verification attempt has
3774 not happened, "verify = helo" does it there and then.
3776 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3778 TK/03 Fix log output including CR from clamd.
3780 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3781 exist provoked a memory error which could cause a segfault.
3783 PH/15 Installed PCRE 6.2
3785 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3787 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3788 of the problem. Specifically, suggested +O2 rather than +O1 for the
3791 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3793 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3794 record, the port was not correctly logged when the outgoing_port log
3795 selector was set (it logged the transort's default port).
3797 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3798 fallback_hosts, and "hosts" in the smtp transport.
3800 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3801 host errors such as "Connection refused".
3803 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3804 authentication with radiusclient 0.4.9:
3806 - Error returned from rc_read_config was caught wrongly
3807 - Username/password not passed on to radius server due to wrong length.
3809 The presumption is that some radiusclient API changes for 4.51/PH/17
3810 were not taken care of correctly. The code is still untested by me (my
3811 Linux distribution still has 0.3.2 of radiusclient), but it was
3812 contributed by a Radius user.
3814 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3815 expanding the "port" option of the smtp transport.
3817 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3818 while reading a MIME header. Thanks to Tom Hughes for a patch.
3820 PH/24 Include config.h inside local_scan.h so that configuration settings are
3823 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3824 that in an ACL for RCPT (for example), you can examine exactly what was
3827 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3828 commands, but it was not correctly comparing the address with the actual
3829 client host address. Thus, it would show the EHLO address in Received:
3830 header lines when this was not necessary.
3832 PH/27 Added the % operator to ${eval:}.
3834 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3835 it should be ignoring failures (because with -C, for example, it has lost
3836 privilege). It wasn't ignoring creation failures other than "already
3839 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3842 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3843 because an input line was too long, either on its own, or by virtue of
3844 too many continuations, the temporary file was not being removed, and the
3845 return code was incorrect.
3847 PH/31 Missing "BOOL" in function definition in filtertest.c.
3849 PH/32 Applied Sieve patches from the maintainer.
3851 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3853 PH/33 Added "verify = not_blind".
3855 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3856 Local/Makefile (with some defaults set). These are used in built scripts
3857 such as exicyclog, but they have never been used in the exim_install
3858 script (though there are many overriding facilities there). I have
3859 arranged that the exim_install script now takes note of these two
3862 PH/35 Installed configuration files for Dragonfly.
3864 PH/36 When a locally submitted message by a trusted user did not contain a
3865 From: header, and the sender address was obtained from -f or from an SMTP
3866 MAIL command, and the trusted user did not use -F to supply a sender
3867 name, $originator_name was incorrectly used when constructing a From:
3868 header. Furthermore, $originator_name was used for submission mode
3869 messages from external hosts without From: headers in a similar way,
3870 which is clearly wrong.
3872 PH/37 Added control=suppress_local_fixups.
3874 PH/38 When log_selector = +received_sender was set, and the addition of the
3875 sender made the log line's construction buffer exactly full, or one byte
3876 less than full, an overflow happened when the terminating "\n" was
3879 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3880 when the result of a list match is failure because a DNS lookup failed.
3882 PH/40 RM_COMMAND is now used in the building process.
3884 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3885 the "build-* directories that it finds.
3887 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3888 address in a domain literal was a prefix of an interface address.
3890 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3891 when verifying a sender address, unless rewrite_headers is false.
3893 PH/44 Wrote a long comment about why errors_to addresses are verified as
3894 recipients, not senders.
3896 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3897 the ratelimit ACL was added.
3899 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3901 PH/46 Added extra information about PostgreSQL errors to the error string.
3903 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3904 avoiding the need to install special zones in a real server. This is
3905 backwards compatible; if it can't find the fake resolver, it drops back.
3906 Thus, both old and new test suites can be run.
3908 TF/02 Added util/ratelimit.pl
3910 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3911 clock is set back in time.
3913 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3914 Candler <B.Candler@pobox.com>.
3916 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3917 OK for addresses that are the result of redirections.
3919 PH/48 A number of further additions for the benefit of the new test suite,
3920 including a fake gethostbyname() that interfaces to the fake DNS resolver
3923 TF/06 The fix for widen_domains has also been applied to qualify_single and
3924 search_parents which are the other dnslookup options that can cause
3927 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3930 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3932 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3933 service" if it encountered an MX record with an empty target hostname.
3934 The message is now "an MX or SRV record indicated no SMTP service".
3936 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3937 if the DNS of the sending site is misconfigured. This is quite a
3938 common situation. This change restores the behaviour of treating a
3939 helo verification defer as a failure.
3941 PH/51 If self=fail was set on a router, the bounce message did not include the
3942 actual error message.
3948 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3950 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3951 command itself ended in error, the underlying message about the transport
3952 filter timeout was being overwritten with the pipe command error. Now the
3953 underlying error message should be appended to the second error message.
3955 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3957 PH/02 Reduce the amount of output that "make" produces by default. Full output
3958 can still be requested.
3960 PH/03 The warning log line about a condition test deferring for a "warn" verb
3961 was being output only once per connection, rather than after each
3962 occurrence (because it was using the same function as for successful
3963 "warn" verbs). This seems wrong, so I have changed it.
3965 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3966 it should not have, which might have caused a crash in the right
3967 circumstances, but probably never did.
3969 PH/04 Installed a modified version of Tony Finch's patch to make submission
3970 mode fix the return path as well as the Sender: header line, and to
3971 add a /name= option so that you can make the user's friendly name appear
3974 TF/03 Added the control = fakedefer ACL modifier.
3976 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3977 Mark Lowes for thorough testing.
3979 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3981 TK/03 Merged latest SRS patch from Miles Wilton.
3983 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3984 with the definition in sysexits.h (which is #included earlier).
3985 Fortunately, Exim does not actually use EX_OK. The code used to try to
3986 preserve the sysexits.h value, by assuming that macro definitions were
3987 scanned for macro replacements. I have been disabused of this notion,
3988 so now the code just undefines EX_OK before #including unistd.h.
3990 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3991 in the smtp transport. When a block could not be written in a single
3992 write() function, the timeout was being re-applied to each part-write.
3993 This seems wrong - if the receiver was accepting one byte at a time it
3994 would take for ever. The timeout is now adjusted when this happens. It
3995 doesn't have to be particularly precise.
3997 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3998 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4000 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4001 without a domain if the check to <postmaster@domain> fails.
4003 SC/01 Eximstats: added -xls and the ability to specify output files
4004 (patch written by Frank Heydlauf).
4006 SC/02 Eximstats: use FileHandles for outputting results.
4008 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4010 SC/04 Eximstats: fixed display of large numbers with -nvr option
4012 SC/05 Eximstats: fixed merging of reports with empty tables.
4014 SC/06 Eximstats: added the -include_original_destination flag
4016 SC/07 Eximstats: removed tabs and trailing whitespace.
4018 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4020 TK/06 MBOX spool code: Add real "From " MBOX separator line
4021 so the .eml file is really in mbox format (even though
4022 most programs do not really care). Patch from Alex Miller.
4024 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4025 The latter is generated from $received_to and is only set if the
4026 message has one envelope recipient. SA can use these headers,
4027 obviously out-of-the-box. Patch from Alex Miller.
4029 PH/08 The ${def test on a variable was returning false if the variable's
4030 value was "0", contrary to what the specification has always said!
4031 The result should be true unless the variable is empty.
4033 PH/09 The syntax error of a character other than { following "${if
4034 def:variable_name" (after optional whitespace) was not being diagnosed.
4035 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4036 accidental colon was present, for example, could give incorrect results.
4038 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4039 result is used (not including appendfile, where other changes are about
4042 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4043 This involved changing a lot of size variables from int to off_t. It
4044 should work with maildirs and everything.
4046 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4047 spamd dying while we are connected to it.
4049 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4052 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4053 function for BDB 4.3.
4055 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4057 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4058 the sources for PCRE have been re-arranged and more files are now
4061 PH/15 The code I had for printing potentially long long variables in PH/11
4062 above was not the best (it lost precision). The length of off_t variables
4063 is now inspected at build time, and an appropriate printing format (%ld
4064 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4065 to be "long long int" or "long int". This is needed for the internal
4066 formatting function string_vformat().
4068 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4069 the configuration file to be ":syslog", then the script "guesses" where
4070 the logs files are, rather than using the compiled in default. In our
4071 case the guess is not the same as the compiled default, so the script
4072 suddenly stopped working when I started to use syslog. The patch checks
4073 to see if log_file_path is "". If so, it attempts to read it from exim
4074 with no configuration file to get the compiled in version, before it
4075 falls back to the previous guessing code."
4077 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4078 implementing BATV in an Exim configuration. See NewStuff for the gory
4081 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4082 Makefile that are specific to HP-UX.
4084 PH/18 If the "use_postmaster" option was set for a recipient callout together
4085 with the "random" option, the postmaster address was used as the MAIL
4086 FROM address for the random test, but not for the subsequent recipient
4087 test. It is now used for both.
4089 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4090 patch removes a few documentation additions to RFC 3028, because the
4091 latest draft now contains them. It adds the new en;ascii-case comparator
4092 and a new error check for 8bit text in MIME parts. Comparator and
4093 require names are now matched exactly. I enabled the subaddress
4094 extension, but it is not well tested yet (read: it works for me)."
4096 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4097 rework some of the code of TK/09 above to avoid the hardwired use of
4098 "%lld" and "long long". Replaced the call to snprintf() with a call to
4101 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4102 records point to non-existent hosts", "retry timeout exceeded", and
4103 "retry time not reached for any host after a long failure period".
4105 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4106 experimental DomainKeys support:
4108 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4109 (2) On an error such as an illegally used "control", the wrong name for
4110 the control was given.
4112 These problems did NOT occur unless DomainKeys support was compiled.
4114 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4116 PH/24 Added ${if match_ip condition.
4118 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4119 databases so that it will be absolutely obvious if a crash occurs in the
4120 DB library. This is a regular occurrence (often caused by mis-matched
4123 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4124 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4125 picked up on a user's system that detects such things. There doesn't seem
4126 to be a gcc warning option for this - only an attribute that has to be
4127 put on the function's prototype. It seems that in Fedora Core 4 they have
4128 set this on a number of new functions. No doubt there will be more in due
4131 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4132 specify a transport. However, if an address that was verified by such a
4133 router was the subject of a callout, Exim crashed because it tried to
4134 read the rcpt_include_affixes from the non-existent transport. Now it
4135 just assumes that the setting of that option is false. This bug was
4136 introduced by 4.51/PH/31.
4138 PH/28 Changed -d+all to exclude +memory, because that information is very
4139 rarely of interest, but it makes the output a lot bigger. People tend to
4140 do -d+all out of habit.
4142 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4143 code in os-type was giving problems when libc.so lives in lib64, like on
4146 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4147 aren't the modern standard, and it seems that some systems' include files
4148 don't always have them. Exim was already checking for some of the newer
4149 ones like T_AAAA, and defining it itself. I've added checks for all the
4150 record types that Exim uses.
4152 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4153 not automatically generating a new one, as it is supposed to. This
4154 prevented TLS from working. If the file did exist, but contained invalid
4155 data, a new version was generated, as expected. It was only the case of a
4156 non-existent file that was broken.
4158 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4159 with a change in libdomainkeys > 0.64.
4161 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4162 from DNS. If the selector record carries the flag, it now has
4163 precedence over the domain-wide flag.
4165 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4167 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4168 the use of a port name that isn't defined in /etc/services) occurred, the
4169 message was deferred as in a normal delivery, and thus remained on the
4170 spool, instead of being failed because of the mua_wrapper setting. This
4171 is now fixed, and I tidied up some of the mua_wrapper messages at the
4174 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4175 the messages in a hash of arrays rather than using individual hashes.
4176 This is a bit cleaner and results in dramatic memory savings, albeit
4177 at a slight CPU cost.
4179 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4180 as requested by Marc Sherman.
4182 SC/10 Eximstats: added histograms for user specified patterns as requested
4185 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4187 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4188 fopen() in the content-scanning modules that did not already have it.
4194 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4195 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4197 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4199 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4201 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4202 to test Sieve filters that use "vacation".
4204 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4205 that changes the way the GnuTLS parameters are stored in the cache file.
4206 The new format can be generated externally. For backward compatibility,
4207 if the data in the cache doesn't make sense, Exim assumes it has read an
4208 old-format file, and it generates new data and writes a new file. This
4209 means that you can't go back to an older release without removing the
4212 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4213 work if there are any delivery delays because "one_time" forces the
4214 parent to be marked "delivered", so its unseen clone is never tried
4215 again. For this reason, Exim now forbids the simultaneous setting of
4218 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4219 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4220 entry if you want to know the details. The fix, however, neglected to
4221 consider the case when local delivery batching is involved. The test for
4222 "previously delivered" was not happening when checking to see if an
4223 address could be batched with a previous (undelivered) one; under
4224 certain circumstances this could lead to multiple deliveries to the same
4227 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4228 in its include files, and this causes problems building Exim.
4230 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4231 header_syntax) but Exim was just ignoring anything given after a slash.
4232 In particular, this caused confusion with an attempt to use "verify =
4233 reverse_host_lookup/defer_ok". An error is now given when options are
4234 supplied for verify items that do not have them. (Maybe reverse_host_
4235 lookup should have a defer_ok option, but that's a different point.)
4237 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4238 defined by RFC 821) to 2048, because there were problems with some AUTH
4239 commands, and RFC 1869 says the size should be increased for extended
4240 SMTP commands that take arguments.
4242 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4245 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4246 "unknown" error; now it says that the functionality isn't in the binary.
4248 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4249 an address' error message when a string expansion fails (syntax or
4250 whatever). Otherwise the password may appear in the log. Following change
4251 PH/42 below, there is no longer a chance of it appearing in a bounce
4254 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4256 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4257 of its addresses was ever tried. (Bugzilla bug #2.)
4259 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4260 the result incorrectly in the debug output. (It correctly added a newline
4261 to what was transported.)
4263 TF/01 Added $received_time.
4265 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4266 commented out examples of how to interface to a virus scanner and to
4267 SpamAssassin. Also added commented examples of av_scanner and
4268 spamd_address settings.
4270 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4271 and controls are allowed in which ACLs. There were a couple of minor
4272 errors. Some of the entries in the conditions table (which is a table of
4273 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4274 negation of where the condition IS allowed.
4276 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4278 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4279 header file does not have a version number, so I've had to invent a new
4280 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4281 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4282 radiusclient), but it was contributed by a Radius user.
4284 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4285 files or directories, for OpenSSL.
4287 PH/19 When an Exim process that is running as root has to create an Exim log
4288 file, it does so in a subprocess that runs as exim:exim so as to get the
4289 ownership right at creation (otherwise, other Exim processes might see
4290 the file with the wrong ownership). There was no test for failure of this
4291 fork() call, which would lead to the process getting stuck as it waited
4292 for a non-existent subprocess. Forks do occasionally fail when resources
4293 run out. I reviewed all the other calls to fork(); they all seem to check
4296 PH/20 When checking for unexpected SMTP input at connect time (before writing
4297 the banner), Exim was not dealing correctly with a non-positive return
4298 from the read() function. If the client had disconnected by this time,
4299 the result was a log entry for a synchronization error with an empty
4300 string after "input=" when read() returned zero. If read() returned -1
4301 (an event I could not check), uninitialized data bytes were printed.
4302 There were reports of junk text (parts of files, etc) appearing after
4305 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4307 PH/22 Added support for macro redefinition, and (re)definition in between
4308 driver and ACL definitions.
4310 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4311 forgetting to use the resulting value; it was using the unexpanded value.
4313 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4314 hadn't been configured. The fix is from Juergen Kreileder, who
4315 understands it better than I do:
4317 "Here's what I see happening with three configured cyrus_sasl
4318 authenticators configured (plain, login, cram-md5):
4320 On startup auth_cyrus_sasl_init() gets called for each of these.
4321 This means three calls to sasl_listmech() without a specified mech_list.
4322 => SASL tests which mechs of all available mechs actually work
4323 => three warnings about OTP not working
4324 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4326 With the patch, sasl_listmech() also gets called three times. But now
4327 SASL's mech_list option is set to the server_mech specified in the the
4328 authenticator. Or in other words, the answer from sasl_listmech()
4329 gets limited to just the mech you're testing for (which is different
4331 => the return list contains just 'plain' or 'login', 'cram-md5' or
4332 nothing depending on the value of ob->server_mech.
4334 I've just tested the patch: Authentication still works fine,
4335 unavailable mechs specified in the exim configuration are still
4336 caught, and the auth.log warnings about OTP are gone."
4338 PH/25 When debugging is enabled, the contents of the command line are added
4339 to the debugging output, even when log_selector=+arguments is not
4342 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4343 answer is "GNU", and only if the return is "GNU/something" is the answer
4346 PH/27 $acl_verify_message is now set immediately after the failure of a
4347 verification in an ACL, and so is available in subsequent modifiers. In
4348 particular, the message can be preserved by coding like this:
4350 warn !verify = sender
4351 set acl_m0 = $acl_verify_message
4353 Previously, $acl_verify_message was set only while expanding "message"
4354 and "log_message" when a very denied access.
4356 PH/28 Modified OS/os.c-Linux with
4358 -#ifndef OS_LOAD_AVERAGE
4359 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4361 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4362 nomenclature these days.)
4364 PH/29 Installed patch from the Sieve maintainer that adds the options
4365 sieve_useraddress and sieve_subaddress to the redirect router.
4367 PH/30 In these circumstances:
4368 . Two addresses routed to the same list of hosts;
4369 . First host does not offer TLS;
4370 . First host accepts first address;
4371 . First host gives temporary error to second address;
4372 . Second host offers TLS and a TLS session is established;
4373 . Second host accepts second address.
4374 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4375 and peerdn, if requested) that were in fact used only for the second
4378 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4379 attention to any local part prefix or suffix that was matched by the
4380 router that accepted the address. It now behaves in the same way as it
4381 does for delivery: the affixes are removed from the local part unless
4382 rcpt_include_affixes is set on the transport.
4384 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4385 timeout during the DATA phase of an incoming message.
4387 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4388 applied a patch sanctioned by the Sieve maintainer.
4390 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4391 the uid or gid is negative. A case of a negative gid caused this to be
4392 noticed. The fix allows for either to be negative.
4394 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4395 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4398 PH/36 The change PH/12 above was broken. Fixed it.
4400 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4401 the grounds that routing the same address twice would always produce the
4402 same answer. This might have been true once, but it is certainly no
4403 longer true now. Routing a child address may depend on the previous
4404 routing that produced that child. Some complicated redirection strategies
4405 went wrong when messages had multiple recipients, and made Exim's
4406 behaviour dependent on the order in which the addresses were given.
4408 I have moved the duplicate checking until after the routing is complete.
4409 Exim scans the addresses that are assigned to local and remote
4410 transports, and removes any duplicates. This means that more work will be
4411 done, as duplicates will always all be routed, but duplicates are
4412 presumably rare, so I don't expect this is of any significance.
4414 For deliveries to pipes, files, and autoreplies, the duplicate checking
4415 still happens during the routing process, since they are not going to be
4418 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4419 It corrects a timeout issue with spamd. This is Ian's comment: "The
4420 background is that sometimes spamd either never reads data from a
4421 connection it has accepted, or it never writes response data. The exiscan
4422 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4423 blindly assumes that writes won't block so it may never time out."
4425 PH/39 Allow G after quota size as well as K and M.
4427 PH/40 The value set for $authenticated_id in an authenticator may not contain
4428 binary zeroes or newlines because the value is written to log lines and
4429 to spool files. There was no check on this. Now the value is run through
4430 the string_printing() function so that such characters are converted to
4431 printable escape sequences.
4433 PH/41 $message_linecount is a new variable that contains the total number of
4434 lines in the message. Compare $body_linecount, which is the count for the
4437 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4438 bounce and delay warning messages, except in certain special cases, which
4441 (a) An SMTP error message from a remote host;
4442 (b) A message specified in a :fail: redirection;
4443 (c) A message specified in a "fail" command in a system filter;
4444 (d) A message specified in a FAIL return from the queryprogram router;
4445 (e) A message specified by the cannot_route_message router option.
4447 In these cases only, Exim does include the error details in bounce and
4448 warning messages. There are also a few cases where bland messages such
4449 as "unrouteable address" or "local delivery error" are given.
4451 PH/43 $value is now also set for the "else" part of a ${run expansion.
4453 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4454 being worked on, but at least Exim now implements the latest version to
4457 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4458 process to complete was treated as a delivery failure, a timeout while
4459 writing the message to the pipe was logged, but erroneously treated as a
4460 successful delivery. Such timeouts include transport filter timeouts. For
4461 consistency with the overall process timeout, these timeouts are now
4462 treated as errors, giving rise to delivery failures by default. However,
4463 there is now a new Boolean option for the pipe transport called
4464 timeout_defer, which, if set TRUE, converts the failures into defers for
4465 both kinds of timeout. A transport filter timeout is now identified in
4468 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4469 systems where "make" and "gmake" are different, calling "gmake" at top
4470 level broke things. I've arranged for the value of $(MAKE) to be passed
4471 from the Makefile to this script so that it can call the same version of
4475 A note about Exim versions 4.44 and 4.50
4476 ----------------------------------------
4478 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4479 changes of various kinds. As a consequence, a big documentation update was
4480 needed. This delayed the release for rather longer than seemed good, especially
4481 in the light of a couple of (minor) security issues. Therefore, the changes
4482 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4483 release. So 4.44 and 4.50 are in effect two different branches that both start
4486 I have left the 4.50 change log unchanged; it contains all the changes since
4487 4.43. The change log for 4.44 is below; many of its items are identical to
4488 those for 4.50. This seems to be the most sensible way to preserve the
4489 historical information.
4495 1. Minor wording change to the doc/README.SIEVE file.
4497 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4498 computation of the current number of files was incorrect.
4500 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4501 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4504 4. Give more explanation in the error message when the command for a transport
4505 filter fails to execute.
4507 5. There are several places where Exim runs a non-Exim command in a
4508 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4509 was being done only for the command run by the queryprogram router. It is
4510 now done for all such subprocesses. The other cases are: ${run, transport
4511 filters, and the commands run by the lmtp and pipe transports.
4513 6. Added CONFIGURE_GROUP build-time option.
4515 7. Some older OS have a limit of 256 on the maximum number of file
4516 descriptors. Exim was using setrlimit() to set 1000 as a large value
4517 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4518 systems. I've change it so that if it can't get 1000, it tries for 256.
4520 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4521 was an oversight, and furthermore, ever since the addition of extra
4522 controls (e.g. 4.43/32), the checks on when to allow different forms of
4523 "control" were broken. There should now be diagnostics for all cases when a
4524 control that does not make sense is encountered.
4526 9. Added the /retain_sender option to "control=submission".
4528 10. $recipients is now available in the predata ACL (oversight).
4530 11. Tidy the search cache before the fork to do a delivery from a message
4531 received from the command line. Otherwise the child will trigger a lookup
4532 failure and thereby defer the delivery if it tries to use (for example) a
4533 cached ldap connection that the parent has called unbind on.
4535 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4536 of $address_data from the recipient verification was clobbered by the
4537 sender verification.
4539 13. The value of address_data from a sender verification is now available in
4540 $sender_address_data in subsequent conditions in the ACL statement.
4542 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4544 15. Added a new option "connect=<time>" to callout options, to set a different
4547 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4548 was its contents. (It was OK if the option was not defined at all.)
4550 17. A "Completed" log line is now written for messages that are removed from
4551 the spool by the -Mrm option.
4553 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4554 information about exactly what failed.
4556 19. Added -dd to debug only the daemon process.
4558 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4559 handles timeouts, both on the server side and network timeouts. Renamed the
4560 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4562 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4563 It is now set to "smtps".
4565 22. $host_address is now set to the target address during the checking of
4566 ignore_target_hosts.
4568 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4569 being passed; this would have caused $sender_host_name to have been used if
4570 matching the list had actually called for a host name (not very likely,
4571 since this list is usually IP addresses). A host name is now passed as
4574 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4575 code that specifies a non-restarting handler (typically sigaction() in
4576 modern systems) in an attempt to fix a rare and obscure crash bug.
4578 25. Narrowed the window for a race in the daemon that could cause it to ignore
4579 SIGCHLD signals. This is not a major problem, because they are used only to
4580 wake it up if nothing else does.
4582 26. A malformed maildirsize file could cause Exim to calculate negative values
4583 for the mailbox size or file count. Odd effects could occur as a result.
4584 The maildirsize information is now recalculated if the size or filecount
4587 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4588 support for a long time. Removed HAVE_SYS_VFS_H.
4590 28. Installed the latest version of exipick from John Jetmore.
4592 29. In an address list, if the pattern was not a regular expression, an empty
4593 subject address (from a bounce message) matched only if the pattern was an
4594 empty string. Non-empty patterns were not even tested. This was the wrong
4595 because it is perfectly reasonable to use an empty address as part of a
4596 database query. An empty address is now tested by patterns that are
4597 lookups. However, all the other forms of pattern expect the subject to
4598 contain a local part and a domain, and therefore, for them, an empty
4599 address still always fails if the pattern is not itself empty.
4601 30. Exim went into a mad DNS loop when attempting to do a callout where the
4602 host was specified on an smtp transport, and looking it up yielded more
4603 than one IP address.
4605 31. Re-factored the code for checking spool and log partition space into a
4606 function that finds that data and another that does the check. The former
4607 is then used to implement four new variables: $spool_space, $log_space,
4608 $spool_inodes, and $log_inodes.
4610 32. The RFC2047 encoding function was originally intended for short strings
4611 such as real names; it was not keeping to the 75-character limit for
4612 encoded words that the RFC imposes. It now respects the limit, and
4613 generates multiple encoded words if necessary. To be on the safe side, I
4614 have increased the buffer size for the ${rfc2047: expansion operator from
4617 33. It is now permitted to omit both strings after an "if" condition; if the
4618 condition is true, the result is "true". As before, when the second string
4619 is omitted, a false condition yields an empty string. This makes it less
4620 cumbersome to write custom ACL and router conditions.
4622 34. Failure to deliver a bounce message always caused it to be frozen, even if
4623 there was an errors_to setting on the router. The errors_to setting is now
4626 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4627 canonical form (fully expanded) before being placed in
4628 $sender_host_address.
4630 36. The table in the code that translates DNS record types into text (T_A to
4631 "A" for instance) was missing entries for NS and CNAME. It is just possible
4632 that this could have caused confusion if both these types were looked up
4633 for the same domain, because the text type is used as part of Exim's
4634 per-process caching. But the chance of anyone hitting this buglet seems
4637 37. The dnsdb lookup has been extended in a number of ways.
4639 (1) There is a new type, "zns", which walks up the domain tree until it
4640 finds some nameserver records. It should be used with care.
4642 (2) There is a new type, "mxh", which is like "mx" except that it returns
4643 just the host names, not the priorities.
4645 (3) It is now possible to give a list of domains (or IP addresses) to be
4646 looked up. The behaviour when one of the lookups defers can be
4647 controlled by a keyword.
4649 (4) It is now possible to specify the separator character for use when
4650 multiple records are returned.
4652 38. The dnslists ACL condition has been extended: it is now possible to supply
4653 a list of IP addresses and/or domains to be looked up in a particular DNS
4656 39. Added log_selector=+queue_time_overall.
4658 40. When running the queue in the test harness, wait just a tad after forking a
4659 delivery process, to get repeatability of debugging output.
4661 41. Include certificate and key file names in error message when GnuTLS fails
4662 to set them up, because the GnuTLS error message doesn't include the name
4663 of the failing file when there is a problem reading it.
4665 42. Allow both -bf and -bF in the same test run.
4667 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4669 44. The "Exiscan patch" is now merged into the mainline Exim source.
4671 45. Sometimes the final signoff response after QUIT could fail to get
4672 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4673 < 0 before doing a fflush(). This bug looks as though it goes back to the
4674 introduction of TLS in release 3.20, but "sometimes" must have been rare
4675 because the tests only now provoked it.
4677 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4678 (this can affect the format of dates).
4680 47. exim_tidydb, when checking for the continued existence of a message for
4681 which it has found a message-specific retry record, was not finding
4682 messages that were in split spool directories. Consequently, it was
4683 deleting retry records that should have stayed in existence.
4685 48. Steve fixed some bugs in eximstats.
4687 49. The SPA authentication driver was not abandoning authentication and moving
4688 on to the next authenticator when an expansion was forced to fail,
4689 contradicting the general specification for all authenticators. Instead it
4690 was generating a temporary error. It now behaves as specified.
4692 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4693 (the order specifies the preference for clients). The order is now AES256,
4694 AES128, 3DES, ARCFOUR128.
4696 51. Small patch to Sieve code - explicitly set From: when generating an
4699 52. Exim crashed if a remote delivery caused a very long error message to be
4700 recorded - for instance if somebody sent an entire SpamAssassin report back
4701 as a large number of 550 error lines. This bug was coincidentally fixed by
4702 increasing the size of one of Exim's internal buffers (big_buffer) that
4703 happened as part of the Exiscan merge. However, to be on the safe side, I
4704 have made the code more robust (and fixed the comments that describe what
4707 53. Now that there can be additional text after "Completed" in log lines (if
4708 the queue_time_overall log selector is set), a one-byte patch to exigrep
4709 was needed to allow it to recognize "Completed" as not the last thing in
4712 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4713 patch that reportedly fixes this has been added. I am not expert enough to
4714 create a test for it. This is what the patch creator wrote:
4716 "I found a little strange behaviour of ldap code when working with
4717 Windows 2003 AD Domain, where users was placed in more than one
4718 Organization Units. When I tried to give exim partial DN, the exit code
4719 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4720 But simultaneously result of request was absolutely normal ldap result,
4721 so I produce this patch..."
4723 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4724 REFERENCE, so I have modified the code to exclude the patch when that macro
4727 55. Some experimental protocols are using DNS PTR records for new purposes. The
4728 keys for these records are domain names, not reversed IP addresses. The
4729 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4730 leaves it alone. Component reversal etc. now happens only for IP addresses.
4733 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4735 57. Double the size of the debug message buffer (to 2048) so that more of very
4736 long debug lines gets shown.
4738 58. The exicyclog utility now does better if the number of log files to keep
4739 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4741 59. Two changes related to the smtp_active_hostname option:
4743 (1) $smtp_active_hostname is now available as a variable.
4744 (2) The default for smtp_banner uses $smtp_active_hostname instead
4745 of $primary_hostname.
4747 60. The host_aton() function is supposed to be passed a string that is known
4748 to be a valid IP address. However, in the case of IPv6 addresses, it was
4749 not checking this. This is a hostage to fortune. Exim now panics and dies
4750 if the condition is not met. A case was found where this could be provoked
4751 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4752 components; fortuitously, this particular loophole had already been fixed
4753 by change 4.50/55 above.
4755 If there are any other similar loopholes, the new check in host_aton()
4756 itself should stop them being exploited. The report I received stated that
4757 data on the command line could provoke the exploit when Exim was running as
4758 exim, but did not say which command line option was involved. All I could
4759 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4760 running as the user.
4763 61. There was a buffer overflow vulnerability in the SPA authentication code
4764 (which came originally from the Samba project). I have added a test to the
4765 spa_base64_to_bits() function which I hope fixes it.
4768 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4769 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4771 63. The daemon start-up calls getloadavg() while still root for those OS that
4772 need the first call to be done as root, but it missed one case: when
4773 deliver_queue_load_max is set with deliver_drop_privilege. This is
4774 necessary for the benefit of the queue runner, because there is no re-exec
4775 when deliver_drop_privilege is set.
4777 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4778 This has been fixed.
4780 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4781 was in use, was not putting the data itself into the right store pool;
4782 consequently, it could be overwritten for a subsequent message in the same
4783 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4786 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4788 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4789 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4790 the function were treating the return as a boolean value, which happened to
4791 work because 0=false and not-0=true, but is not correct code.
4793 68. The host_aton() function was not handling scoped IPv6 addresses (those
4794 with, for example, "%eth0" on the end) correctly.
4796 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4797 negated items (that is, ~something) in unsigned ints. Some compilers
4798 apparently mutter when there is no cast.
4800 70. If an address verification called from an ACL failed, and did not produce a
4801 user-specific message (i.e. there was only a "system" message), nothing was
4802 put in $acl_verify_message. In this situation, it now puts the system
4805 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4806 session; change 4.31/43 added the unwanted input to the log line - except
4807 that it did not do this in the start of session case. It now does.
4809 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4810 This is wrong and can cause the other end to generate a synchronization
4811 error if it is another Exim or anything else that does the synchronization
4812 check. A QUIT command is no longer sent after a timeout.
4814 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4815 during host lookups.
4817 74. The defer_ok option of callout verification was not working if it was used
4818 when verifying addresses in header lines, that is, for this case:
4820 verify = header_sender/callout=defer_ok
4822 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4823 those file descriptors could be used for SMTP connections. If anything
4824 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4825 could be sent to the SMTP client, causing chaos. The daemon now opens
4826 stdin, stdout, and stderr to /dev/null when it puts itself into the
4829 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4830 log by default. The user can override this with suitable Perl magic.
4832 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4833 the log message when discard triggers, was not working for the DATA ACL or
4834 for the non-SMTP ACL.
4836 78. Error message wording change in sieve.c.
4838 79. If smtp_accept_max_per_host was set, the number of connections could be
4839 restricted to fewer than expected, because the daemon was trying to set up
4840 a new connection before checking whether the processes handling previous
4841 connections had finished. The check for completed processes is now done
4842 earlier. On busy systems, this bug wouldn't be noticed because something
4843 else would have woken the daemon, and it would have reaped the completed
4846 80. If a message was submitted locally by a user whose login name contained one
4847 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4848 It caused a spool format error. I have fixed the spool reading code. A
4849 related problem was that the "from" clause in the Received: line became
4850 illegal because of the space(s). It is now covered by ${quote_local_part.
4852 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4855 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4857 83. Patch from Sieve maintainer to fix unterminated string problem in
4858 "vacation" handling.
4860 84. Some minor changes to the Linux configuration files to help with other
4861 OS variants using glibc.
4863 85. One more patch for Sieve to update vacation handling to latest spec.
4866 ----------------------------------------------------
4867 See the note above about the 4.44 and 4.50 releases.
4868 ----------------------------------------------------
4874 1. Change 4.43/35 introduced a bug that caused file counts to be
4875 incorrectly computed when quota_filecount was set in an appendfile
4878 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4879 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4882 3. Give more explanation in the error message when the command for a transport
4883 filter fails to execute.
4885 4. There are several places where Exim runs a non-Exim command in a
4886 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4887 was being done only for the command run by the queryprogram router. It is
4888 now done for all such subprocesses. The other cases are: ${run, transport
4889 filters, and the commands run by the lmtp and pipe transports.
4891 5. Some older OS have a limit of 256 on the maximum number of file
4892 descriptors. Exim was using setrlimit() to set 1000 as a large value
4893 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4894 systems. I've change it so that if it can't get 1000, it tries for 256.
4896 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4897 was an oversight, and furthermore, ever since the addition of extra
4898 controls (e.g. 4.43/32), the checks on when to allow different forms of
4899 "control" were broken. There should now be diagnostics for all cases when a
4900 control that does not make sense is encountered.
4902 7. $recipients is now available in the predata ACL (oversight).
4904 8. Tidy the search cache before the fork to do a delivery from a message
4905 received from the command line. Otherwise the child will trigger a lookup
4906 failure and thereby defer the delivery if it tries to use (for example) a
4907 cached ldap connection that the parent has called unbind on.
4909 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4910 of $address_data from the recipient verification was clobbered by the
4911 sender verification.
4913 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4914 was its contents. (It was OK if the option was not defined at all.)
4916 11. A "Completed" log line is now written for messages that are removed from
4917 the spool by the -Mrm option.
4919 12. $host_address is now set to the target address during the checking of
4920 ignore_target_hosts.
4922 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4923 being passed; this would have caused $sender_host_name to have been used if
4924 matching the list had actually called for a host name (not very likely,
4925 since this list is usually IP addresses). A host name is now passed as
4928 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4929 code that specifies a non-restarting handler (typically sigaction() in
4930 modern systems) in an attempt to fix a rare and obscure crash bug.
4932 15. Narrowed the window for a race in the daemon that could cause it to ignore
4933 SIGCHLD signals. This is not a major problem, because they are used only to
4934 wake it up if nothing else does.
4936 16. A malformed maildirsize file could cause Exim to calculate negative values
4937 for the mailbox size or file count. Odd effects could occur as a result.
4938 The maildirsize information is now recalculated if the size or filecount
4941 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4942 support for a long time. Removed HAVE_SYS_VFS_H.
4944 18. Updated exipick to current release from John Jetmore.
4946 19. Allow an empty sender to be matched against a lookup in an address list.
4947 Previously the only cases considered were a regular expression, or an
4950 20. Exim went into a mad DNS lookup loop when doing a callout where the
4951 host was specified on the transport, if the DNS lookup yielded more than
4954 21. The RFC2047 encoding function was originally intended for short strings
4955 such as real names; it was not keeping to the 75-character limit for
4956 encoded words that the RFC imposes. It now respects the limit, and
4957 generates multiple encoded words if necessary. To be on the safe side, I
4958 have increased the buffer size for the ${rfc2047: expansion operator from
4961 22. Failure to deliver a bounce message always caused it to be frozen, even if
4962 there was an errors_to setting on the router. The errors_to setting is now
4965 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4966 canonical form (fully expanded) before being placed in
4967 $sender_host_address.
4969 24. Updated eximstats to version 1.33
4971 25. Include certificate and key file names in error message when GnuTLS fails
4972 to set them up, because the GnuTLS error message doesn't include the name
4973 of the failing file when there is a problem reading it.
4975 26. Expand error message when OpenSSL has problems setting up cert/key files.
4978 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4979 (this can affect the format of dates).
4981 28. exim_tidydb, when checking for the continued existence of a message for
4982 which it has found a message-specific retry record, was not finding
4983 messages that were in split spool directories. Consequently, it was
4984 deleting retry records that should have stayed in existence.
4986 29. eximstats updated to version 1.35
4987 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4988 1.35 - bugfix such that pie charts by volume are generated correctly
4990 30. The SPA authentication driver was not abandoning authentication and moving
4991 on to the next authenticator when an expansion was forced to fail,
4992 contradicting the general specification for all authenticators. Instead it
4993 was generating a temporary error. It now behaves as specified.
4995 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4996 (the order specifies the preference for clients). The order is now AES256,
4997 AES128, 3DES, ARCFOUR128.
4999 31. Small patch to Sieve code - explicitly set From: when generating an
5002 32. Exim crashed if a remote delivery caused a very long error message to be
5003 recorded - for instance if somebody sent an entire SpamAssassin report back
5004 as a large number of 550 error lines. This bug was coincidentally fixed by
5005 increasing the size of one of Exim's internal buffers (big_buffer) that
5006 happened as part of the Exiscan merge. However, to be on the safe side, I
5007 have made the code more robust (and fixed the comments that describe what
5010 33. Some experimental protocols are using DNS PTR records for new purposes. The
5011 keys for these records are domain names, not reversed IP addresses. The
5012 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5013 leaves it alone. Component reversal etc. now happens only for IP addresses.
5016 34. The host_aton() function is supposed to be passed a string that is known
5017 to be a valid IP address. However, in the case of IPv6 addresses, it was
5018 not checking this. This is a hostage to fortune. Exim now panics and dies
5019 if the condition is not met. A case was found where this could be provoked
5020 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5021 components; fortuitously, this particular loophole had already been fixed
5022 by change 4.50/55 or 4.44/33 above.
5024 If there are any other similar loopholes, the new check in host_aton()
5025 itself should stop them being exploited. The report I received stated that
5026 data on the command line could provoke the exploit when Exim was running as
5027 exim, but did not say which command line option was involved. All I could
5028 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5029 running as the user.
5032 35. There was a buffer overflow vulnerability in the SPA authentication code
5033 (which came originally from the Samba project). I have added a test to the
5034 spa_base64_to_bits() function which I hope fixes it.
5037 36. The daemon start-up calls getloadavg() while still root for those OS that
5038 need the first call to be done as root, but it missed one case: when
5039 deliver_queue_load_max is set with deliver_drop_privilege. This is
5040 necessary for the benefit of the queue runner, because there is no re-exec
5041 when deliver_drop_privilege is set.
5043 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5044 was in use, was not putting the data itself into the right store pool;
5045 consequently, it could be overwritten for a subsequent message in the same
5046 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5049 38. Sometimes the final signoff response after QUIT could fail to get
5050 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5051 < 0 before doing a fflush(). This bug looks as though it goes back to the
5052 introduction of TLS in release 3.20, but "sometimes" must have been rare
5053 because the tests only now provoked it.
5059 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5060 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5061 Now it returns an integer. A number of calls were still expecting a T/F
5062 return. Fortuitously, in all cases, the tests worked in OK situations,
5063 which is the norm. However, things would have gone wrong on any write
5064 failures on the smtp file descriptor. This function is used when sending
5065 messages over SMTP and also when doing verify callouts.
5067 2. When Exim is called to do synchronous delivery of a locally submitted
5068 message (the -odf or -odi options), it no longer closes stderr before doing
5071 3. Implemented the mua_wrapper option.
5073 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5075 5. Implemented the functions header_remove(), header_testname(),
5076 header_add_at_position(), and receive_remove_recipient(), and exported them
5079 6. If an ACL "warn" statement specified the addition of headers, Exim already
5080 inserted X-ACL-Warn: at the start if there was no header name. However, it
5081 was not making this test for the second and subsequent header lines if
5082 there were newlines in the string. This meant that an invalid header could
5083 be inserted if Exim was badly configured.
5085 7. Allow an ACL "warn" statement to add header lines at the start or after all
5086 the Received: headers, as well as at the end.
5088 8. Added the rcpt_4xx retry error code.
5090 9. Added postmaster_mailfrom=xxx to callout verification option.
5092 10. Added mailfrom=xxxx to the callout verification option, for verify=
5095 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5096 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5098 12. Inserted some casts to stop certain compilers warning when using pointer
5099 differences as field lengths or precisions in printf-type calls (mostly
5100 affecting debugging statements).
5102 13. Added optional readline() support for -be (dynamically loaded).
5104 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5105 same clock tick as a message's arrival, so that its received time was the
5106 same as the "first fail" time on the retry record, and that message
5107 remained on the queue past the ultimate address timeout, every queue runner
5108 would try a delivery (because it was past the ultimate address timeout) but
5109 after another failure, the ultimate address timeout, which should have then
5110 bounced the address, did not kick in. This was a "< instead of <=" error;
5111 in most cases the first failure would have been in the next clock tick
5112 after the received time, and all would be well.
5114 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5115 being recognized when the domain list was tested by the match_domain
5116 condition in an expansion string.
5118 16. Added the ${str2b64: operator.
5120 17. Exim was always calling setrlimit() to set a large limit for the number of
5121 processes, without checking whether the existing limit was already
5122 adequate. (It did check for the limit on file descriptors.) Furthermore,
5123 errors from getrlimit() and setrlimit() were being ignored. Now they are
5124 logged to the main and panic logs, but Exim does carry on, to try to do its
5125 job under whatever limits there are.
5127 18. Imported PCRE 5.0.
5129 19. Trivial typo in log message " temporarily refused connection" (the leading
5132 20. If the log selector return_path_on_delivery was set and an address was
5133 redirected to /dev/null, the delivery process crashed because it assumed
5134 that a return path would always be set for a "successful" delivery. In this
5135 case, the whole delivery is bypassed as an optimization, and therefore no
5138 21. Internal re-arrangement: the function for sending a challenge and reading
5139 a response while authentication was assuming a zero-terminated challenge
5140 string. It's now changed to take a pointer and a length, to allow for
5141 binary data in such strings.
5143 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5145 23. Exim was not respecting finduser_retries when seeking the login of the
5146 uid under which it was called; it was always trying 10 times. (The default
5147 setting of finduser_retries is zero.) Also, it was sleeping after the final
5148 failure, which is pointless.
5150 24. Implemented tls_on_connect_ports.
5152 25. Implemented acl_smtp_predata.
5154 26. If the domain in control=submission is set empty, Exim assumes that the
5155 authenticated id is a complete email address when it generates From: or
5156 Sender: header lines.
5158 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5159 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5160 chown and chgrp in /bin and hostname in /usr/bin.
5162 28. Exim was keeping the "process log" file open after each use, just as it
5163 does for the main log. This opens the possibility of it remaining open for
5164 long periods when the USR1 signal hits a daemon. Occasional processlog
5165 errors were reported, that could have been caused by this. Anyway, it seems
5166 much more sensible not to leave this file open at all, so that is what now
5169 29. The long-running daemon process does not normally write to the log once it
5170 has entered its main loop, and it closes the log before doing so. This is
5171 so that log files can straightforwardly be renamed and moved. However,
5172 there are a couple of unusual error situations where the daemon does write
5173 log entries, and I had neglected to close the log afterwards.
5175 30. The text of an SMTP error response that was received during a remote
5176 delivery was being truncated at 512 bytes. This is too short for some of
5177 the long messages that one sometimes sees. I've increased the limit to
5180 31. It is now possible to make retry rules that apply only when a message has a
5181 specific sender, in particular, an empty sender.
5183 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5184 it possible to be selective about when SMTP synchronization is enforced.
5186 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5188 32. Implemented hosts_connection_nolog.
5190 33. Added an ACL for QUIT.
5192 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5195 35. Added mailbox_size and mailbox_filecount to appendfile.
5197 36. Added control = no_multiline_responses to ACLs.
5199 37. There was a bug in the logic of the code that waits for the clock to tick
5200 in the case where the clock went backwards by a substantial amount such
5201 that the microsecond fraction of "now" was more than the microsecond
5202 fraction of "then" (but the whole seconds number was less).
5204 38. Added support for the libradius Radius client library this is found on
5205 FreeBSD (previously only the radiusclient library was supported).
5211 1. When certain lookups returned multiple values in the form name=value, the
5212 quoting of the values was not always being done properly. Specifically:
5213 (a) If the value started with a double quote, but contained no whitespace,
5215 (b) If the value contained whitespace other than a space character (i.e.
5216 tabs or newlines or carriage returns) it was not quoted.
5217 This fix has been applied to the mysql and pgsql lookups by writing a
5218 separate quoting function and calling it from the lookup code. The fix
5219 should probably also be applied to nisplus, ibase and oracle lookups, but
5220 since I cannot test any of those, I have not disturbed their existing code.
5222 2. A hit in the callout cache for a specific address caused a log line with no
5223 reason for rejecting RCPT. Now it says "Previous (cached) callout
5224 verification failure".
5226 3. There was an off-by-one bug in the queryprogram router. An over-long
5227 return line was truncated at 256 instead of 255 characters, thereby
5228 overflowing its buffer with the terminating zero. As well as fixing this, I
5229 have increased the buffer size to 1024 (and made a note to document this).
5231 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5232 when Exim is waiting for an SMTP response from a remote server, Exim
5233 restarts its select() call on the socket, thereby resetting its timeout.
5234 This is not a problem when such interrupts are rare. Somebody set up a cron
5235 job to run exiwhat every 2 minutes, which is less than the normal select()
5236 timeout (5 or 10 minutes). This meant that the select() timeout never
5237 kicked in because it was always reset. I have fixed this by comparing the
5238 time when an interrupt arrives with the time at the start of the first call
5239 to select(). If more time than the timeout has elapsed, the interrupt is
5240 treated as a timeout.
5242 5. Some internal re-factoring in preparation for the addition of Sieve
5243 extensions (by MH). In particular, the "personal" test is moved to a
5244 separate function, and given an option for scanning Cc: and Bcc: (which is
5245 not set for Exim filters).
5247 6. When Exim created an email address using the login of the caller as the
5248 local part (e.g. when creating a From: or Sender: header line), it was not
5249 quoting the local part when it contained special characters such as @.
5251 7. Installed new OpenBSD configuration files.
5253 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5254 try to make them clearer.
5256 9. Callout options, other than the timeout value, were being ignored when
5257 verifying sender addresses in header lines. For example, when using
5259 verify = header_sender/callout=no_cache
5261 the cache was (incorrectly) being used.
5263 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5264 only the Cygwin environment.
5266 11. When return_path_on_delivery was set as a log selector, if different remote
5267 addresses in the same message used different return paths and parallel
5268 remote delivery occurred, the wrong values would sometimes be logged.
5269 (Whenever a remote delivery process finished, the return path value from
5270 the most recently started remote delivery process was logged.)
5272 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5273 lines when AUTH and/or TLS are in use. This is the "received protocol"
5274 field. Exim used to use "asmtp" for authenticated SMTP, without any
5275 indication (in the protocol name) for TLS use. Now it follows the RFC and
5276 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5277 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5278 names appear in log lines as well as in Received: header lines.
5280 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5281 extensions, and comparison tests, and to fix some bugs.
5283 14. Changes to the "personal" filter test:
5285 (1) The test was buggy in that it was just doing the equivalent of
5286 "contains" tests on header lines. For example, if a user's address was
5287 anne@some.where, the "personal" test would incorrectly be true for
5289 To: susanne@some.where
5291 This test is now done by extracting each address from the header in turn,
5292 and checking the entire address. Other tests that are part of "personal"
5293 are now done using regular expressions (for example, to check local parts
5294 of addresses in From: header lines).
5296 (2) The list of non-personal local parts in From: addresses has been
5297 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5298 taken from the Sieve specification recommendations.
5300 (3) If the message contains any header line starting with "List-" it is
5301 treated as non-personal.
5303 (4) The test for "circular" in the Subject: header line has been removed
5304 because it now seems ill-conceived.
5306 15. Minor typos in src/EDITME comments corrected.
5308 16. Installed latest exipick from John Jetmore.
5310 17. If headers_add on a router specified a text string that was too long for
5311 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5312 of string_sprintf() is now avoided.
5314 18. $message_body_size was not set (it was always zero) when running the DATA
5315 ACL and the local_scan() function.
5317 19. For the "mail" command in an Exim filter, no default was being set for
5318 the once_repeat time, causing a random time value to be used if "once" was
5319 specified. (If the value happened to be <= 0, no repeat happened.) The
5320 default is now 0s, meaning "never repeat". The "vacation" command was OK
5321 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5322 (I found it when inspecting the code).
5324 20. There is now an overall timeout for performing a callout verification. It
5325 defaults to 4 times the callout timeout, which applies to individual SMTP
5326 commands during the callout. The overall timeout applies when there is more
5327 than one host that can be tried. The timeout is checked before trying the
5328 next host. This prevents very long delays if there are a large number of
5329 hosts and all are timing out (e.g. when the network connections are timing
5330 out). The value of the overall timeout can be changed by specifying an
5331 additional sub-option for "callout", called "maxwait". For example:
5333 verify = sender/callout=5s,maxwait=20s
5335 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5336 to the end before writing, but this should make it even safer).
5338 22. Exim was forgetting that it had advertised PIPELINING for the second and
5339 subsequent messages on an SMTP connection. It was also not resetting its
5340 memory on STARTTLS and an internal HELO.
5342 23. When Exim logs an SMTP synchronization error within a session, it now
5343 records whether PIPELINING has been advertised or not.
5345 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5346 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5347 rather than long int.
5349 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5351 26. Added the never_mail option to autoreply.
5357 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5358 crash if the getsockname() call failed; this can happen if a connection is
5359 closed very soon after it is established. The problem was simply in the
5360 order in which certain operations were done, causing Exim to try to write
5361 to the SMTP stream before it had set up the file descriptor. The bug has
5362 been fixed by making things happen in the correct order.
5368 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5369 before the connection was closed, thus losing the rejection response.
5371 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5372 some early Solaris releases, but causes trouble in current releases where
5373 socklen_t is defined.
5375 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5378 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5381 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5382 that releasing the top of it at the end releases what was used for sub-
5383 expansions (unless the block got too big). However, discard this block if
5384 the first thing is a variable or header, so that we can use its block when
5385 it is dynamic (useful for very large $message_headers, for example).
5387 6. Lookups now cache *every* query, not just the most recent. A new, separate
5388 store pool is used for this. It can be recovered when all lookup caches are
5389 flushed. Lookups now release memory at the end of their result strings.
5390 This has involved some general refactoring of the lookup sources.
5392 7. Some code has been added to the store_xxx() functions to reduce the amount
5393 of flapping under certain conditions.
5395 8. log_incoming_interface used to affect only the <= reception log lines. Now
5396 it causes the local interface and port to be added to several more SMTP log
5397 lines, for example "SMTP connection from", and rejection lines.
5399 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5401 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5403 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5404 could be overwritten at the end of the current message (or the start of a
5405 new message if it was set in a HELO ACL). The value is now preserved for
5406 the duration of the SMTP connection.
5408 12. If a transport had a headers_rewrite setting, and a matching header line
5409 contained an unqualified address, that address was qualified, even if it
5410 did not match any rewriting rules. The underlying bug was that the values
5411 of the flags that permit the existence of unqualified sender and recipient
5412 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5413 non-local messages, and by -bnq for local messages) were not being
5414 preserved with the message after it was received.
5416 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5417 "next input=" as part of the text comprising the host identity instead of
5418 the correct text. The code was using the same buffer for two different
5419 strings. However, depending on which order the printing function evaluated
5420 its arguments, the bug did not always show up. Under Linux, for example, my
5421 test suite worked just fine.
5423 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5424 doesn't work with some older versions of Perl. It has been changed to "my",
5425 which in any case is probably the better facility to use.
5427 15. A really picky compiler found some instances of statements for creating
5428 error messages that either had too many or two few arguments for the format
5431 16. The size of the buffer for calls to the DNS resolver has been increased
5432 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5433 for addresses that have a lot of PTR records. This alleviates a problem; it
5434 does not fully solve it.
5436 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5437 buffer now truncates the list and logs the incident, which is the same
5438 action as happens when Exim is looking up a host name and its aliases.
5439 Previously in this situation something unpredictable would happen;
5440 sometimes it was "internal error: store_reset failed".
5442 18. If a server dropped the connection unexpectedly when an Exim client was
5443 using GnuTLS and trying to read a response, the client delivery process
5444 crashed while trying to generate an error log message.
5446 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5447 string, for example:
5449 warn message = H1: something\nH2: something
5451 the text was added as a single header line from Exim's point of view
5452 though it ended up OK in the delivered message. However, searching for the
5453 second and subsequent header lines using $h_h2: did not work. This has been
5454 fixed. Similarly, if a system filter added multiple headers in this way,
5455 the routers could not see them.
5457 20. Expanded the error message when iplsearch is called with an invalid key to
5458 suggest using net-iplsearch in a host list.
5460 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5461 ACLs are no longer actually imposed (and a message to that effect is
5464 22. If a "gecos" field in a passwd entry contained escaped characters, in
5465 particular, if it contained a \" sequence, Exim got it wrong when building
5466 a From: or a Sender: header from that name. A second bug also caused
5467 incorrect handling when an unquoted " was present following a character
5468 that needed quoting.
5470 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5471 was not being matched caselessly.
5473 24. Arranged for all hyphens in the exim.8 source to be escaped with
5476 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5477 properly. Recipient callout cache records were still being keyed to include
5478 the sender, even when use_sender was set false. This led to far more
5479 callouts that were necessary. The sender is no longer included in the key
5480 when use_sender is false.
5482 26. Added "control = submission" modifier to ACLs.
5484 27. Added the ${base62d: operator to decode base 62 numbers.
5486 28. dnsdb lookups can now access SRV records.
5488 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5489 the configuration file.
5491 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5492 (-v) mode. This makes the output for a verbose queue run more intelligible.
5494 31. Added a use_postmaster feature to recipient callouts.
5496 32. Added the $body_zerocount variable, containing the number of binary zero
5497 bytes in the message body.
5499 33. The time of last modification of the "new" subdirectory is now used as the
5500 "mailbox time last read" when there is a quota error for a maildir
5503 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5505 35. Added +ignore_unknown as a special item in host lists.
5507 36. Code for decoding IPv6 addresses in host lists is now included, even if
5508 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5509 address was recognized as an IP address, but was then not correctly decoded
5510 into binary, causing unexpected and incorrect effects when compared with
5517 1. Very minor rewording of debugging text in manualroute to say "list of
5518 hosts" instead of "hostlist".
5520 2. If verify=header_syntax was set, and a header line with an unqualified
5521 address (no domain) and a large number of spaces between the end of the
5522 name and the colon was received, the reception process suffered a buffer
5523 overflow, and (when I tested it) crashed. This was caused by some obsolete
5524 code that should have been removed. The fix is to remove it!
5526 3. When running in the test harness, delay a bit after writing a bounce
5527 message to get a bit more predictability in the log output.
5529 4. Added a call to search_tidyup() just before forking a reception process. In
5530 theory, someone could use a lookup in the expansion of smtp_accept_max_
5531 per_host which, without the tidyup, could leave open a database connection.
5533 5. Added the variables $recipient_data and $sender_data which get set from a
5534 lookup success in an ACL "recipients" or "senders" condition, or a router
5535 "senders" option, similar to $domain_data and $local_part_data.
5537 6. Moved the writing of debug_print from before to after the "senders" test
5540 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5541 problems for message scanning, either using a data ACL, or using
5542 local_scan() because the Received: header was not generated till after they
5543 were called (in order to set the time as the time of reception completion).
5544 I have revised the way this works. The header is now generated after the
5545 body is received, but before the ACL or local_scan() are called. After they
5546 are run, the timestamp in the header is updated.
5552 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5553 before starting a queue runner without re-exec. This happened only when
5554 deliver_drop_privilege was set or when the Exim user was set to root. The
5555 effect of the bug was that timeouts during subsequent deliveries caused
5556 crashes instead of being properly handled. The handler is now left at its
5557 default (and expected) setting.
5559 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5560 message, again when deliver_drop_privilege is set or Exim is run as root.
5561 The bug described in (1) was not present in this case, but the tidying up
5562 of the other signals was missing. I have made the two cases consistent.
5564 3. The ignore_target_hosts setting on a manualroute router was being ignored
5565 for hosts that were looked up using the /MX notation.
5567 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5570 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5571 operated on the sender address. After changing the $sender_address to <>
5572 for the sender address verify, Exim was re-instated it as the original
5573 (before rewriting) address, but remembering that it had rewritten it, so it
5574 wasn't rewriting it again. This bug also had the effect of breaking the
5575 sender address verification caching when the sender address was rewritten.
5577 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5578 This has been changed so that if the ip literal address matches
5579 ignore_target_hosts, the router declines.
5581 7. Added expansion conditions match_domain, match_address, and match_local_
5582 part (NOT match_host).
5584 8. The placeholder for the Received: header didn't have a length field set.
5586 9. Added code to Exim itself and to exim_lock to test for a specific race
5587 condition that could lead to file corruption when using MBX delivery. The
5588 issue is with the lockfile that is created in /tmp. If this file is removed
5589 after a process has opened it but before that process has acquired a lock,
5590 there is the potential for a second process to recreate the file and also
5591 acquire a lock. This could lead to two Exim processes writing to the file
5592 at the same time. The added code performs the same test as UW imapd; it
5593 checks after acquiring the lock that its file descriptor still refers to
5594 the same named file.
5596 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5597 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5600 11. Added the smtp_active_hostname option. If used, this will typically be made
5601 to depend on the incoming interface address. Because $interface_address is
5602 not set up until the daemon has forked a reception process, error responses
5603 that can happen earlier (such as "too many connections") no longer contain
5606 12. If an expansion in a condition on a "warn" statement fails because a lookup
5607 defers, the "warn" statement is abandoned, and the next ACL statement is
5608 processed. Previously this caused the whole ACL to be aborted.
5610 13. Added the iplsearch lookup type.
5612 14. Added ident_timeout as a log selector.
5614 15. Added tls_certificate_verified as a log selector.
5616 16. Added a global option tls_require_ciphers (compare the smtp transport
5617 option of the same name). This controls incoming TLS connections.
5619 17. I finally figured out how to make tls_require_ciphers do a similar thing
5620 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5621 before starting the TLS session.
5623 18. Tabs are now shown as \t in -bP output.
5625 19. If the log selector return_path_on_delivery was set, Exim crashed when
5626 bouncing a message because it had too many Received: header lines.
5628 20. If two routers both had headers_remove settings, and the first one included
5629 a superfluous trailing colon, the final name in the first list and the
5630 first name in the second list were incorrectly joined into one item (with a
5631 colon in the middle).
5637 1. Added -C and -D options to the exinext utility, mainly to make it easier
5638 to include in the automated testing, but these could be helpful when
5639 multiple configurations are in use.
5641 2. The exinext utility was not formatting the output nicely when there was
5642 an alternate port involved in the retry record key, nor when there was a
5643 message id as well (for retries that were specific to a specific message
5644 and a specific host). It was also confused by IPv6 addresses, because of
5645 the additional colons they contain. I have fixed the IPv4 problem, and
5646 patched it up to do a reasonable job for IPv6.
5648 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5649 delivery, the log line now contains "pipelined" if PIPELINING was used.
5651 4. An SMTP transport process used to panic and die if the bind() call to set
5652 an explicit outgoing interface failed. This has been changed; it is now
5653 treated in the same way as a connect() failure.
5655 5. A reference to $sender_host_name in the part of a conditional expansion
5656 that was being skipped was still causing a DNS lookup. This no longer
5659 6. The def: expansion condition was not recognizing references to header lines
5660 that used bh_ and bheader_.
5662 7. Added the _cache feature to named lists.
5664 8. The code for checking quota_filecount in the appendfile transport was
5665 allowing one more file than it should have been.
5667 9. For compatibility with Sendmail, the command line option
5675 and sets the incoming protocol and host name (for trusted callers). The
5676 host name and its colon can be omitted when only the protocol is to be set.
5677 Note the Exim already has two private options, -pd and -ps, that refer to
5678 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5679 "s", but I don't think that's a major issue.
5681 10. A number of refactoring changes to the code, none of which should affect
5684 (a) The number of logging options was getting close to filling up the
5685 32-bit word that was used as a bit map. I have split them into two classes:
5686 those that are passed in the argument to log_write(), and those that are
5687 only ever tested independently outside of that function. These are now in
5688 separate 32-bit words, so there is plenty of room for expansion again.
5689 There is no change in the user interface or the logging behaviour.
5691 (b) When building, for example, log lines, the code previously used a
5692 macro that called string_cat() twice, in order to add two strings. This is
5693 not really sufficiently general. Furthermore, there was one instance where
5694 it was actually wrong because one of the argument was used twice, and in
5695 one call a function was used. (As it happened, calling the function twice
5696 did not affect the overall behaviour.) The macro has been replaced by a
5697 function that can join an arbitrary number of extra strings onto a growing
5700 (c) The code for expansion conditions now uses a table and a binary chop
5701 instead of a serial search (which was left over from when there were very
5702 few conditions). Also, it now recognizes conditions like "pam" even when
5703 the relevant support is not compiled in: a suitably worded error message is
5704 given if an attempt is made to use such a condition.
5706 11. Added ${time_interval:xxxxx}.
5708 12. A bug was causing one of the ddress fields not to be passed back correctly
5709 from remote delivery subprocesses. The field in question was not being
5710 subsequently used, so this caused to problems in practice.
5712 13. Added new log selectors queue_time and deliver_time.
5714 14. Might have fixed a bug in maildirsizefile handling that threw up
5715 "unexpected character" debug warnings, and recalculated the data
5716 unnecessarily. In any case, I expanded the warning message to give more
5719 15. Added the message "Restricted characters in address" to the statements in
5720 the default ACL that block characters like @ and % in local parts.
5722 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5723 Three changes have been made:
5725 (a) There was a serious bug; a negative response to MAIL caused the whole
5726 recipient domain to be cached as invalid, thereby blocking all messages
5727 to all local parts at the same domain, from all senders. This bug has
5728 been fixed. The domain is no longer cached after a negative response to
5729 MAIL if the sender used is not empty.
5731 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5734 (c) A new callout option, "use_sender" has been added for people who want
5735 the modified behaviour.
5741 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5744 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5745 indeed breaks things for older releases.
5747 3. Added additional logging to the case where there is a problem reading data
5748 from a filter that is running in a subprocess using a pipe, in order to
5749 try to track down a specific problem.
5751 4. Testing facility fudge: when running in the test harness and attempting
5752 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5753 getting "No route to host". Convert this to a timeout.
5755 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5758 6. Some OS don't have socklen_t but use size_t instead. This affects the
5759 fifth argument of getsockopt() amongst other things. This is now
5760 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5761 can be set for individual OS. I have set it for SunOS5, OSF1, and
5762 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5763 some earlier ones do not.
5765 7. Change 4.30/15 was not doing the test caselessly.
5767 8. The standard form for an IPv6 address literal was being rejected by address
5768 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5769 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5770 this, as well as the form without the "IPv6" on the front (but only when
5771 address literals are enabled, of course).
5773 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5775 10. Exim crashed if a message with an empty sender address specified by -f
5776 encountered a router with an errors_to setting. This could be provoked only
5777 by a command such as
5781 where an empty string was supplied; "<>" did not hit this bug.
5783 11. Installed PCRE release 4.5.
5785 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5786 remained set. It is now erased.
5788 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5789 times from message ids (which are base 36 rather than the normal 62).
5791 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5792 were being counted as actual protocol errors, and logged if the log
5793 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5794 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5795 and DATA following a set of rejected RCPTs do not count as protocol errors.
5796 In other words, Exim assumes they were pipelined, though this may not
5797 actually be the case. Of course, in all cases the client gets an
5798 appropriate error code.
5800 15. If a lookup fails in an ACL condition, a message about the failure may
5801 be available; it is used if testing the ACL cannot continue, because most
5802 such messages specify what the cause of the deferral is. However, some
5803 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5804 that caused an old message to be retained and used if a later statement
5805 caused a defer, replacing the real cause of the deferral.
5807 16. If an IP address had so many PTR records that the DNS lookup buffer
5808 was not large enough to hold them, Exim could crash while trying to process
5809 the truncated data. It now detects and logs this case.
5811 17. Further to 4.21/58, another change has been made: if (and only if) the
5812 first line of a message (the first header line) ends with CRLF, a bare LF
5813 in a subsequent header line has a space inserted after it, so as not to
5814 terminate the header.
5816 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5817 unnecessarily, used atoi() instead of strtol(), and didn't check the
5818 termination when getting file sizes from file names by regex.
5820 19. Completely re-implemented the support for maildirsize files, in the light
5821 of a number of problems with the previous contributed implementation
5822 (4.30/29). In particular:
5824 . If the quota is zero, the maildirsize file is maintained, but no quota is
5827 . If the maildir directory does not exist, it is created before any attempt
5828 to write a maildirsize file.
5830 . The quota value in the file is just a cache; if the quota is changed in
5831 the transport, the new value overrides.
5833 . A regular expression is available for excluding directories from the
5836 20. The autoreply transport checks the characters in options that define the
5837 message's headers; it allows continued headers, but it was checking with
5838 isspace() after an embedded newline instead of explicitly looking for a
5841 21. If all the "regular" hosts to which an address was routed had passed their
5842 expiry times, and had not reached their retry times, the address was
5843 bounced, even if fallback hosts were defined. Now Exim should go on to try
5846 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5847 equivalent code in the SMTP transport. Some hosts send humungous responses
5848 to HELO/EHLO, more than 1024 it seems.
5850 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5851 gives compiler warnings in some environments. I've now done it "properly",
5854 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5855 (because of problems with the built-in one) was declared to return uschar *
5856 instead of char *, causing compiler failure.
5858 25. Fixed a file descriptor leak when processing alias/forward files.
5860 26. Fixed a minor format string issue in dbfn.c.
5862 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5864 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5865 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5868 29. When a sender address is verified, it is cached, to save repeating the test
5869 when there is more than one recipient in a message. However, when the
5870 verification involves a callout, it is possible for different callout
5871 options to be set for different recipients. It is too complicated to keep
5872 track of this in the cache, so now Exim always runs a verification when a
5873 callout is required, relying on the callout cache for the optimization.
5874 The overhead is duplication of the address routing, but this should not be
5877 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5878 to be verified with callout=postmaster, and the main callout worked but the
5879 postmaster check failed, the verification correctly failed. However, if a
5880 subsequent RCPT command asked for sender verification *without* the
5881 postmaster check, incorrect caching caused this verification also to fail,
5884 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5885 it was not caching the DNS options (qualify_single, search_parents) that
5886 were used when the lookup failed. A subsequent lookup with different
5887 options therefore always gave the same answer, though there were cases
5888 where it should not have. (Example: a "domains = !$mx_any" option on a
5889 dnslookup router: the "domains" option is always processed without any
5890 widening, but the router might have qualify_single set.) Now Exim uses the
5891 cached value only when the same options are set.
5893 32. Added John Jetmore's "exipick" utility to the distribution.
5895 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5896 than a timeout (e.g. a certificate is required, and is not provided), an
5897 Exim server now closes the connection immediately. Previously it waited for
5898 the client to close - but if the client is SSL, it seems that they each
5899 wait for each other, leading to a delay before one of them times out.
5901 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5902 maintained 0.8.x compatibility because I don't think many are using it, and
5903 it is clearly obsolete.
5905 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5908 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5909 client certificate was expired. A simple patch fixes this, though I don't
5910 understand the full logic of why the verify callback is called multiple
5913 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5914 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5915 which causes problems with some clients (such as the Certicom SSL Plus
5916 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5917 disables the coutermeasure allowing Eudora to connect."
5919 38. Exim was not checking that a write() to a log file succeeded. This could
5920 lead to Bad Things if a log got too big, in particular if it hit a file
5921 size limit. Exim now panics and dies if it cannot write to a log file, just
5922 as it does if it cannot open a log file.
5924 39. Modified OS/Makefile-Linux so that it now contains
5926 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5928 The two -D definitions ensure that Exim is compiled with large file
5929 support, which makes it possible to handle log files that are bigger than
5932 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5933 instance) a domain was checked against a named list that involved a lookup,
5934 causing $domain_data to be set, then another domain was checked against the
5935 same list, then the first domain was re-checked, the value of $domain_data
5936 after the final check could be wrong. In particular, if the second check
5937 failed, it could be set empty. This bug probably also applied to
5940 41. The strip_trailing_dot option was not being applied to the address given
5941 with the -f command-line option.
5943 42. The code for reading a message's header from the spool was incrementing
5944 $received_count, but never initializing it. This meant that the value was
5945 incorrect (doubled) while delivering a message in the same process in which
5946 it was received. In the most common configuration of Exim, this never
5947 happens - a fresh exec is done - but it can happen when
5948 deliver_drop_privilege is set.
5950 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5951 it now includes up to 150 characters of the unexpected data in the log
5954 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5955 and building data strings. The size of both of these buffers was 10 000
5956 bytes - far larger than anybody would *ever* want, thought I. Needless to
5957 say, somebody hit the limit. I have increased the maximum line length to
5958 20 000 and the maximum data length of concatenated lines to 100 000. I have
5959 also fixed two bugs, because there was no checking on these buffers. Tsk,
5960 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5961 buffer is too small.
5963 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5964 lsearch lookups. Now it does.
5966 46. When parsing a route_list item in a manualroute router, a fixed-length
5967 buffer was used for the list of hosts. I made this 1024 bytes long,
5968 thinking that nobody would ever have a list of hosts that long. Wrong.
5969 Somebody had a whole pile of complicated expansion conditions, and the
5970 string was silently truncated, leading to an expansion error. It turns out
5971 that it is easier to change to an unlimited length (owing to other changes
5972 that have happened since this code was originally written) than to build
5973 structure for giving a limitation error. The length of the item that
5974 expands into the list of hosts is now unlimited.
5976 47. The lsearch lookup could not handle data where the length of text line was
5977 more than 4095 characters. Such lines were truncated, leading to shortened
5978 data being returned. It should now handle lines of any length.
5980 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5981 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5984 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5985 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5986 no longer happens in comment lines. A list of replacements is now placed
5987 at the head of all of the source files, except those whose only change is
5988 to replace PERL_COMMAND in the very first #! line.
5990 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5991 messages on the queue, with a bottom-up merge sort, using code contributed
5992 by Michael Haardt. This should make operations like -bp somewhat faster on
5993 large queues. It won't affect queue runners, except when queue_run_in_order
5996 51. Installed eximstats 1.31 in the distribution.
5998 52. Added support for SRV lookups to the dnslookup router.
6000 53. If an ACL referred to $message_body or $message_body_end, the value was not
6001 reset for any messages that followed in the same SMTP session.
6003 54. The store-handling optimization for building very long strings was not
6004 differentiating between the different store pools. I don't think this
6005 actually made any difference in practice, but I've tidied it.
6007 55. While running the routers to verify a sender address, $sender_address
6008 was still set to the sender address. This is wrong, because when routing to
6009 send a bounce to the sender, it would be empty. Therefore, I have changed
6010 it so that, while verifying a sender address, $sender_address is set to <>.
6011 (There is no change to what happens when verifying a recipient address.)
6013 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6014 target A or AAAA records (if not already returned) without resetting the
6015 qualify_single or search_parents options of the DNS resolver. These are
6016 inappropriate in this case because the targets of MX and SRV records must
6017 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6018 target that, when qualified, matched something in the local domain. These
6019 two options are now turned off when doing these lookups.
6021 57. It seems that at least some releases of Reiserfs (which does not have the
6022 concept of a fixed number of inodes) returns zero and not -1 for the
6023 number of available inodes. This interacted badly with check_spool_inodes,
6024 which assumed that -1 was the "no such thing" setting. What I have done is
6025 to check that the total number of inodes is greater than zero before doing
6026 the test of how many are available.
6028 58. When a "warn" ACL statement has a log_message modifier, the message is
6029 remembered, and not repeated. This is to avoid a lot of repetition when a
6030 message has many recipients that cause the same warning to be written.
6031 However, Exim was preserving the list of already written lines for an
6032 entire SMTP session, which doesn't seem right. The memory is now reset if a
6033 new message is started.
6035 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6036 debugging output unless log_rewrite was also set.
6038 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6039 of (int)(handle) when we know that handle contains (void *)(-1).
6041 61. The Exim daemon panic-logs an error return when it closes the incoming
6042 connection. However "connection reset by peer" seems to be common, and
6043 isn't really an error worthy of noting specially, so that particular error
6046 62. When Exim is trying to find all the local interfaces, it used to panic and
6047 die if the ioctl to get the interface flags failed. However, it seems that
6048 on at least one OS (Solaris 9) it is possible to have an interface that is
6049 included in the list of interfaces, but for which you get a failure error
6050 for this call. This happens when the interface is not "plumbed" into a
6051 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6052 failure of the "get flags" call assumes that the interface is down.
6054 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6055 makes life easier for people who are doing arithmetic on fields extracted
6056 from dates, where you often get leading zeros that should not be
6057 interpreted as octal.
6059 64. Added qualify_domain to the redirect router, to override the global
6062 65. If a pathologically long header line contained very many addresses (the
6063 report of this problem mentioned 10 000) and each of them was rewritten,
6064 Exim could use up a very large amount of memory. (It kept on making new
6065 copies of the header line as it rewrote, and never released the old ones.)
6066 At the expense of a bit more processing, the header rewriting function has
6067 been changed so that it no longer eats memory in this way.
6069 66. The generation of the Received: header has been moved from the time that a
6070 message starts to be received, to the time that it finishes. The timestamp
6071 in the Received: header should now be very close to that of the <= log
6072 line. There are two side-effects of this change:
6074 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6075 logged header lines no longer include the local Received: line, because
6076 it has not yet been created. The same applies to a copy of the message
6077 that is returned to a non-SMTP sender when a message is rejected.
6079 (b) When a filter file is tested using -bf, no additional Received: header
6080 is added to the test message. After some thought, I decided that this
6083 This change does not affect the value of $received_for. It is still set
6084 after address rewriting, but before local_scan() is called.
6086 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6088 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6089 gave an unhelpful panic error message, and a defer error. I have managed to
6090 change this behaviour so that it now rejects any supplied certificate,
6091 which seems right, as the list of acceptable certificates is empty.
6093 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6094 gave an unhelpful defer error. I have not managed to make this reject any
6095 supplied certificates, but the error message it gives is "no certificate
6096 supplied", which is not helpful.
6098 70. exigrep's output now also includes lines that are not associated with any
6099 message, but which match the given pattern. Implemented by a patch from
6100 Martin Sluka, which also tidied up the Perl a bit.
6102 71. Recipient callout verification, like sender verification, was using <> in
6103 the MAIL FROM command. This isn't really the right thing, since the actual
6104 sender may affect whether the remote host accepts the recipient or not. I
6105 have changed it to use the actual sender in the callout; this means that
6106 the cache record is now keyed on a recipient/sender pair, not just the
6107 recipient address. There doesn't seem to be a real danger of callout loops,
6108 since a callout by the remote host to check the sender would use <>.
6109 [SEE ABOVE: changed after hitting problems.]
6111 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6112 temporary errors. However, in the case of such a code being given after
6113 the end of a data transmission (i.e. after ".") Exim was failing to write
6114 a retry record for the message. (Yes, there was some broken host that was
6115 actually sending 8xx at this point.)
6117 73. An unknown lookup type in a host list could cause Exim to panic-die when
6118 the list was checked. (An example that provoked this was putting <; in the
6119 middle of a list instead of at the start.) If this happened during a DATA
6120 ACL check, a -D file could be left lying around. This kind of configuration
6121 error no longer causes Exim to die; instead it causes a defer error. The
6122 incident is still logged to the main and panic logs.
6124 74. Buglet left over from Exim 3 conversion. The message "too many messages
6125 in one connection" was written to the rejectlog but not the mainlog, except
6126 when address rewriting (yes!) was being logged.
6128 75. Added write_rejectlog option.
6130 76. When a system filter was run not as root (that is, when system_filter_user
6131 was set), the values of the $n variables were not being returned to the
6132 main process; thus, they were not subsequently available in the $sn
6135 77. Added +return_path_on_delivery log selector.
6137 78. A connection timeout was being treated differently from recipients deferred
6138 when testing hosts_max_try with a message that was older than the host's
6139 retry timeout. (The host should not be counted, thus allowing all hosts to
6140 be tried at least once before bouncing.) This may have been the cause of an
6141 occasionally reported bug whereby a message would remain on the queue
6142 longer than the retry timeout, but would be bounced if a delivery was
6143 forced. I say "may" because I never totally pinned down the problem;
6144 setting up timeout/retry tests is difficult. See also the next item.
6146 79. The ultimate address timeout was not being applied to errors that involved
6147 a combination of host plus message (for example, a timeout on a MAIL
6148 command). When an address resolved to a number of possible hosts, and they
6149 were not all tried for each delivery (e.g. because of hosts_max_try), a
6150 message could remain on the queue longer than the retry timeout.
6152 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6155 81. Fixed an obscure SMTP outgoing bug which required at least the following
6156 conditions: (a) there was another message waiting for the same server;
6157 (b) the server returned 5xx to all RCPT commands in the first message so
6158 that the message was not completed; (c) the server dropped the connection
6159 or gave a negative response to the RSET that Exim sends to abort the
6160 transaction. The observed case was a dropped connection after DATA that had
6161 been sent in pipelining mode. That is, the server had advertised PIPELINING
6162 but was not implementing it correctly. The effect of the bug was incorrect
6163 behaviour, such as trying another host, and this could lead to a crash.
6169 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6170 and daemon.c were passed as pointers to ints; they should have been
6171 pointers to socklen_t variables (which are typically unsigned ints).
6173 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6176 3. Fixed a really odd bug that affected only the testing scheme; patching a
6177 certain fixed string in the binary changed the value of another string that
6178 happened to be identical to the end of the original first string.
6180 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6181 name", it returns that address as the IP address. On some operating
6182 systems (e.g. Solaris), it also passes back the IP address string as the
6183 "host name". However, on others (e.g. Linux), it passes back an empty
6184 string. Exim wasn't checking for this, and was changing the host name to an
6185 empty string, assuming it had been canonicalized.
6187 5. Although rare, it is permitted to have more than one PTR record for a given
6188 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6189 all the names associated with an address, because they do in Solaris.
6190 However, it seems that they do not in Linux for data that comes from the
6191 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6192 I found this out when I moved to a new Linux workstation and tried to run
6193 the Exim test suite.
6195 To get round this problem I have changed the code so that it now does its
6196 own call to the DNS to look up PTR records when searching for a host name.
6197 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6198 addresses that are only in /etc/hosts are still found.
6200 This behaviour is, however, controlled by an option called host_lookup_
6201 order, which defaults to "bydns:byaddr". If people want to use the other
6202 order, or indeed, just use one or the other means of lookup, they can
6203 specify it in this variable.
6205 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6206 some operating systems, this comes back from gethostbyaddr() as an empty
6207 string, and this is what Exim used to test for. However, it seems that in
6208 other systems, "." is yielded. Exim now tests for this case too.
6210 7. The values of check_spool_space and check_log_space are now held internally
6211 as a number of kilobytes instead of an absolute number of bytes. If a
6212 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6213 kilobyte. This means that much larger values can be stored.
6215 8. Exim monitor: an attempt to get the action menu when not actually pointing
6216 at a message produces an empty menu entitled "No message selected". This
6217 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6218 no entries in it ("Shell widget menu has zero width and/or height"). So I
6219 have added a single, blank menu entry in this case.
6221 9. Added ${quote_local_part.
6223 10. MIME decoding is now applied to the contents of Subject: header lines when
6226 11. Now that a reference to $sender_host_address automatically causes a reverse
6227 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6228 host lookup before query-style lookups in lists that might use this
6229 variable. This has therefore been abolished, and the "net-" prefix is no
6230 longer necessary for query-style lookups.
6232 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6233 have been a typo for LFLAGS, so it has been changed.
6235 13. The install script calls Exim with "-C /dev/null" in order to find the
6236 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6237 to be output. However, since Exim outputs its version number before the
6238 error, it didn't break the script. It just looked ugly. I fixed this by
6239 always allowing "-C /dev/null" if the caller is root.
6241 14. Ignore overlarge ACL variable number when reading spool file - insurance
6242 against a later release with more variables having written the file.
6244 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6245 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6246 this, as well as the form without the "IPv6" on the front.
6248 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6249 OS/Makefile-Darwin file.
6251 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6252 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6254 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6255 variable. However, some OS use ints for inodes. I've added cast to long int
6256 to get rid of the compiler warning.
6258 19. I had forgotten to lock out "/../" in configuration file names when
6259 ALT_CONFIG_PREFIX was set.
6261 20. Routers used for verification do not need to specify transports. However,
6262 if such a router generated a host list, and callout was configured, Exim
6263 crashed, because it could not find a port number from the (non-existent)
6264 transport. It now assumes port 25 in this circumstance.
6266 21. Added the -t option to exigrep.
6268 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6269 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6270 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6271 them is set, LOOKUP_LSEARCH is forced.
6273 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6275 24. Added sender and host information to the "rejected by local_scan()" log
6276 line; previously there was no indication of these.
6278 25. Added .include_if_exists.
6280 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6281 Linux. It turns out that not all file systems support this. Apparently some
6282 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6283 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6284 not supported on the file descriptor, is now ignored when Exim is trying to
6285 sync a directory. This applies only to Linux.
6287 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6289 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6290 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6293 29. Added support for maildirsize files from supplied patch (modified a bit).
6295 30. The use of :fail: followed by an empty string could lead Exim to respond to
6296 sender verification failures with (e.g.):
6298 550 Verification failed for <xxx>
6299 550 Sender verify failed
6301 where the first response line was missing the '-' that indicates it is not
6302 the final line of the response.
6304 31. The loop for finding the name of the user that called Exim had a hardwired
6305 limit of 10; it now uses the value of finduser_retries, which is used for
6306 all other user lookups.
6308 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6311 33. Exim was neglecting to zero errno before one call of strtol() when
6312 expanding a string and expecting an integer value. On some systems this
6313 resulted in spurious "integer overflow" errors. Also, it was casting the
6314 result into an int without checking.
6316 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6317 did not work. The code looks as if it has *never* worked, though it appears
6318 to have been documented since at least release 1.62. I have made it work.
6320 35. The "timeout_DNS" error in retry rules, also documented since at least
6321 1.62, also never worked. As it isn't clear exactly what this means, and
6322 clearly it isn't a major issue, I have abolished the feature by treating it
6323 as "timeout", and writing a warning to the main and panic logs.
6325 36. The display of retry rules for -brt wasn't always showing the error code
6328 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6329 timeout_connect_A, timeout_connect_MX.
6331 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6332 to the empty sender.
6334 39. The daemon was not analysing the content of -oX till after it had closed
6335 stderr and disconnected from the controlling terminal. This meant that any
6336 syntax errors were only noted on the panic log, and the return code from
6337 the command was 0. By re-arranging the code a little, I've made the
6338 decoding happen first, so such errors now appear on stderr, and the return
6339 code is 1. However, the actual setting up of the sockets still happens in
6340 the disconnected process, so errors there are still only recorded on the
6343 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6344 connections (as happens on some IP stacks) was logged at start up time as
6345 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6346 it from "IPv6 and IPv4", which means that two separate sockets are being
6349 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6350 says whether AF_INET or AF_INET6 was passed as an argument.
6352 42. Exiwhat output was messed up when time zones were included in log
6355 43. Exiwhat now gives more information about the daemon's listening ports,
6356 and whether -tls-on-connect was used.
6358 44. The "port" option of the smtp transport is now expanded.
6360 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6361 silently ignored. Now an error message is written to the main and panic
6364 46. There's a new ACL modifier called "logwrite" which writes to a log file
6365 as soon as it is encountered.
6367 47. Added $local_user_uid and $local_user_gid at routing time.
6369 48. Exim crashed when trying to verify a sender address that was being
6372 49. Exim was recognizing only a space character after ".include". It now also
6373 recognizes a tab character.
6375 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6376 extracting the relevant information from the specification. The man page no
6377 longer contains scrambled data for the -d option, and I've added a section
6378 at the front about calling Exim under different names.
6380 51. Added "extra_headers" argument to the "mail" command in filter files.
6382 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6385 53. Installed eximstats 1.29.
6387 54. Added transport_filter_timeout as a generic transport option.
6389 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6390 Cc: header lines. This was required by RFC 822, but it not required by RFC
6393 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6394 incoming messages that did not have them. Now it does so only if the
6395 message originates locally, that is, if there is no associated remote host
6396 address. When Resent- header lines are present, this applies to the Resent-
6397 lines rather than the non-Resent- lines.
6399 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6400 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6402 58. Messages for configuration errors now include the name of the main
6403 configuration file - useful now that there may be more than one file in a
6404 list (.included file names were always shown).
6406 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6407 for those rare installations that do not start the daemon as root or run it
6408 setuid root. I've cut out the call to initgroups() if the daemon is not
6411 60. The Exim user and group can now be bound into the binary as text strings
6412 that are looked up at the start of Exim's processing.
6414 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6416 62. Added $mailstore_basename variable.
6418 63. Installed patch to sieve.c from Michael Haardt.
6420 64. When Exim failed to open the panic log after failing to open the main log,
6421 the original message it was trying to log was written to stderr and debug
6422 output, but if they were not available (the usual case in production), it
6423 was lost. Now it is written to syslog before the two lines that record the
6424 failures to open the logs.
6426 65. Users' Exim filters run in subprocesses under the user's uid. It is
6427 possible for a "deliver" command or an alias in a "personal" command to
6428 provoke an address rewrite. If logging of address rewriting is configured,
6429 this fails because the process is not running as root or exim. There may be
6430 a better way of dealing with this, but for the moment (because 4.30 needs
6431 to be released), I have disabled address rewrite logging when running a
6432 filter in a non-root, non-exim process.
6438 1. The buildconfig auxiliary program wasn't quoting the value set for
6439 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6440 not defined. This bug was masked in 4.22 by the effect that was fixed in
6443 2. Some messages that were rejected after a message id was allocated were
6444 shown as "incomplete" by exigrep. It no longer does this for messages that
6445 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6447 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6448 have allow_domain_literals set, the ID did not get logged in the <= line.
6449 Domain literals are now always recognized in Message-ID: header lines.
6451 4. The first argument for a ${extract expansion item is the key name or field
6452 number. Leading and trailing spaces in this item were not being ignored,
6453 causing some misleading effects.
6455 5. When deliver_drop_privilege was set, single queue runner processes started
6456 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6457 same command in the process it spins off) were not dropping privilege.
6459 6. When the daemon running as "exim" started a queue runner, it always
6460 re-executed Exim in the spun-off process. This is a waste of effort when
6461 deliver_drop_privilege is set. The new process now just calls the
6462 queue-runner function directly.
6468 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6471 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6472 ignored. Though the use of -oP was forcing the writing of a pid file, it
6473 was always written to the default place.
6475 3. If the message "no IP address found for host xxxx" is generated during
6476 incoming verification, it is now followed by identification of the incoming
6477 connection (so you can more easily find what provoked it).
6479 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6481 5. Added some features to "harden" Exim a bit more against certain attacks:
6483 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6484 be put in Local/Makefile. This is like the never_users runtime option,
6485 but it cannot be overridden. The default setting is "root".
6487 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6488 prefix string with which any file named in a -C command line option
6491 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6492 is retained for -C and -D only if the caller of Exim is root. Without
6493 it, the exim user may also use -C and -D and retain privilege.
6495 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6496 command line option is disabled.
6498 6. Macro names set by the -D option must start with an upper case letter, just
6499 like macro names defined in the configuration file.
6501 7. Added "dereference=" facility to LDAP.
6503 8. Two instances of the typo "uknown" in the source files are fixed.
6505 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6506 the Configure-Makefile script screwed up while processing it.
6508 10. Incorporated PCRE 4.4.
6510 11. The SMTP synchronization check was not operating right at the start of an
6511 SMTP session. For example, it could not catch a HELO sent before the client
6512 waited for the greeting. There is now a check for outstanding input at the
6513 point when the greeting is written. Because of the duplex, asynchronous
6514 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6515 way, but not yet received, when the check is performed.
6517 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6518 on TCP/IP sockets, because this apparently causes some broken clients to
6521 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6522 unchanged) from the Cygwin maintainer.
6524 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6525 was supported instead of testing for mbx. Effectively a typo.
6527 15. The spa authenticator server code was not checking that the input it
6528 received was valid base64.
6530 16. The debug output line for the "set" modifier in ACLs was not showing the
6531 name of the variable that was being set.
6533 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6535 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6536 lookup on its own. Something else was needed to trigger the lookup. For
6537 example, a match in host_lookup or the need for a host name in a host list.
6538 Now, if $sender_host_name is referenced and the host name has not yet been
6539 looked up, a lookup is performed. If the lookup fails, the variable remains
6540 empty, and $host_lookup_failed is set to "1".
6542 19. Added "eqi" as a case-independent comparison operator.
6544 20. The saslauthd authentication condition could segfault if neither service
6545 nor realm was specified.
6547 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6548 error message that was logged was misleading, and incoming SMTP
6549 connections were dropped. The message is now more accurate, and temporary
6550 errors are given to SMTP connections.
6552 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6553 (except RSET) with a 421 error, until QUIT is received. However, it was
6554 failing to send a response to QUIT.
6556 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6557 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6558 ACL is now run after the helo_try_verify_hosts code.
6560 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6561 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6562 case-independent, so other case variants are also recognized). Apparently
6563 some systems use these upper case variants.
6565 25. If more than two messages were waiting for the same host, and a transport
6566 filter was specified for the transport, Exim sent two messages over the
6567 same TCP/IP connection, and then failed with "socket operation on non-
6568 socket" when it tried to send the third.
6570 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6572 27. The extern definition of crypt16() in expand.c was not being excluded when
6573 the OS had its own crypt16() function.
6575 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6576 as a preferred synonym for return_size_limit, both as an option and as an
6579 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6581 30. Changed the default configuration ACL to relax the local part checking rule
6582 for addresses that are not in any local domains. For these addresses,
6583 slashes and pipe symbols are allowed within local parts, but the sequence
6584 /../ is explicitly forbidden.
6586 31. SPA server authentication was not clearing the challenge buffer before
6589 32. log_message in a "warn" ACL statement was writing to the reject log as
6590 well as to the main log, which contradicts the documentation and doesn't
6591 seem right (because no rejection is happening). So I have stopped it.
6593 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6594 However, I am unable to do any testing of this.
6596 34. Fixed an infelicity in the appendfile transport. When checking directories
6597 for a mailbox, to see if any needed to be created, it was accidentally
6598 using path names with one or more superfluous leading slashes; tracing
6599 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6601 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6602 contents are added to the log line that is written for every discarded
6603 recipient. (Previously a log_message setting was ignored.)
6605 36. The ${quote: operator now quotes the string if it is empty.
6607 37. The install script runs exim in order to find its version number. If for
6608 some reason other than non-existence or emptiness, which it checks, it
6609 could not run './exim', it was installing it with an empty version number,
6610 i.e. as "exim-". This error state is now caught, and the installation is
6613 38. An argument was missing from the function that creates an error message
6614 when Exim fails to connect to the socket for saslauthd authentication.
6615 This could cause Exim to crash, or give a corrupted message.
6617 39. Added isip, isip4, and isip6 to ${if conditions.
6619 40. The ACL variables $acl_xx are now saved with the message, and can be
6620 accessed later in routers, transports, and filters.
6622 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6623 strings in the file are not string-expanded.
6625 42. If a MAIL command specified a SIZE value that was too large to fit into an
6626 int variable, the check against message_size_limit failed. Such values are
6627 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6628 day this will have to be increased, but I don't think I want to be around
6629 when emails are that large.
6636 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6637 iconv() is not standard in FreeBSD.
6639 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6640 IPv6 enabled. The observed symptom was a segmentation fault on return from
6641 the function os_common_find_running_interfaces() in src/os.c.
6643 3. In the check_special_case() function in daemon.c I had used "errno" as an
6644 argument name, which causes warnings on some systems. This was basically a
6645 typo, since it was named "eno" in the comments!
6647 4. The code that waits for the clock to tick (at a resolution of some fraction
6648 of a second) so as to ensure message-id uniqueness was always waiting for
6649 at least one whole tick, when it could have waited for less. [This is
6650 almost certainly not relevant at current processor speeds, where it is
6651 unlikely to ever wait at all. But we try to future-proof.]
6653 5. The function that sleeps for a time interval that includes fractions of a
6654 second contained a race. It did not block SIGALRM between setting the
6655 timer, and suspending (a couple of lines later). If the interval was short
6656 and the sigsuspend() was delayed until after it had expired, the suspension
6657 never ended. On busy systems this could lead to processes getting stuck for
6660 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6661 process, before it forks any delivery processes. The open lookup caching
6662 mechanism meant that the open file or database connection was passed into
6663 the delivery process. The problem was that delivery processes always tidy
6664 up cached lookup data. This could cause a problem for the next delivery
6665 process started by the queue runner, because the external queue runner
6666 process does not know about the closure. So the next delivery process
6667 still has data in the lookup cache. In the case of a file lookup, there was
6668 no problem because closing a file descriptor in a subprocess doesn't affect
6669 the parent. However, if the lookup was caching a connection to a database,
6670 the connection was closed, and the second delivery process was likely to
6671 see errors such as "PGSQL: query failed: server closed the connection
6672 unexpectedly". The problem has been fixed by closing all cached lookups
6673 in a queue runner before running a delivery process.
6675 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6676 seem to have the "const" qualifier which it has on other OS. I've
6679 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6680 *of the same type* (client or server) with the same public name that an
6681 error should be diagnosed.
6683 9. When Exim looked up a host name for an IP address, but failed to find the
6684 original IP address when looking up the host name (a safety check), it
6685 output the message "<ip address> does not match any IP for NULL", which was
6686 confusing, to say the least. The bug was that the host name should have
6687 appeared instead of "NULL".
6689 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6690 user that is built into the binary, and the -C or -D options is used, root
6691 privilege is dropped before the configuration file is read. In addition,
6692 logging is switched to stderr instead of the normal log files. If the
6693 configuration then re-defines the Exim user, the unprivileged environment
6694 is probably not what is expected, so Exim logs a panic warning message (but
6697 However, if deliver_drop_privilege is set, the unprivileged state may well
6698 be exactly what is intended, so the warning has been cut out in that case,
6699 and Exim is allowed to try to write to its normal log files.
6705 1. smtp_return_error_details was not giving details for temporary sender
6706 or receiver verification errors.
6708 2. Diagnose a configuration error if two authenticators have the same public
6711 3. Exim used not to create the message log file for a message until the first
6712 delivery attempt. This could be confusing when incoming messages were held
6713 for policy or load reasons. The message log file is now created at the time
6714 the message is received, and an initial "Received" line is written to it.
6716 4. The automatically generated man page for command line options had a minor
6717 bug that caused no ill effects; however, a more serious problem was that
6718 the procedure for building the man page automatically didn't always
6719 operate. Consequently, release 4.20 contains an out-of-date version. This
6720 shouldn't happen again.
6722 5. When building Exim with embedded Perl support, the script that builds the
6723 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6724 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6726 6. The freeze_tell option was not being used for messages that were frozen on
6727 arrival, either by an ACL or by local_scan().
6729 7. Added the smtp_incomplete_transaction log selector.
6731 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6732 was accepting AUTH without a new EHLO.
6734 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6735 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6738 10. Logging of TCP/IP connections (when configured) now happens in the main
6739 daemon process instead of the child process, so that the TCP/IP connection
6740 count is more accurate (but it can never be perfect).
6742 11. The use of "drop" in a nested ACL was not being handled correctly in the
6743 outer ACL. Now, if condition failure induced by the nested "drop" causes
6744 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6745 or "require"), the connection is dropped.
6747 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6748 that yield "discard" can now be used with an "accept" or a "discard" verb,
6749 but an error is generated for any others (because I can't see a useful way
6750 to define what should happen).
6752 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6753 are now processed in the same way as lines in the Exim configuration file.
6754 In particular, continuation lines are supported.
6756 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6758 15. Added -ti meaning -t -i.
6760 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6761 domains, and warn by logging if others are found.
6763 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6764 returned from GIFCONF when Exim is trying to find the list of interfaces on
6765 a host. The code in os.c has been modified to copy each ifreq to an aligned
6766 structure in all cases.
6768 Also, in some cases, the returned ifreq's were being copied to a 'struct
6769 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6770 means the last couple of bytes of an IPv6 address could be chopped if the
6771 ifreq contained only a normal sockaddr (14 bytes storage).
6773 18. Named domain lists were not supported in the hosts_treat_as_local option.
6774 An entry such as +xxxx was not recognized, and was treated as a literal
6777 19. Ensure that header lines added by a DATA ACL are included in the reject log
6778 if the ACL subsequently rejects the message.
6780 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6781 MD5 (which is deprecated).
6783 21. When testing a filter file using -bf, Exim was writing a message when it
6784 took the sender from a "From " line in the message, but it was not doing so
6785 when it took $return_path from a Return-Path: header line. It now does.
6787 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6788 with a valid header line field name (a series of printing characters
6789 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6791 23. Changed "disc" in the source to "disk" to conform to the documentation and
6792 the book and for uniformity.
6794 24. Ignore Sendmail's -Ooption=value command line item.
6796 25. When execve() failed while trying to run a command in a pipe transport,
6797 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6798 could be confused with a return value of 69 from the command itself. This
6799 has been changed to 127, the value the shell returns if it is asked to run
6800 a non-existent command. The wording for the related log line suggests a
6801 non-existent command as the problem.
6803 26. If received_header_text expands to an empty string, do not add a Received:
6804 header line to the message. (Well, it adds a token one on the spool, but
6805 marks it "old" so that it doesn't get used or transmitted.)
6807 27. Installed eximstats 1.28 (addition of -nt option).
6809 28. There was no check for failure on the call to getsockname() in the daemon
6810 code. This can fail if there is a shortage of resources on the system, with
6811 ENOMEM, for example. A temporary error is now given on failure.
6813 29. Contrary to the C standard, it seems that in some environments, the
6814 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6815 program. Exim now does this explicitly; it affects the formatting of
6816 timestamps using strftime().
6818 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6819 complaints. I've now initialized all the variables, to avoid this.
6821 32. Header lines added by a system filter were not being "seen" during
6822 transport-time rewrites.
6824 33. The info_callback() function passed to OpenSSL is set up with type void
6825 (*)(SSL *, int, int), as described somewhere. However, when calling the
6826 function (actually a macro) that sets it up, the type void(*)() is
6827 expected. I've put in a cast to prevent warnings from picky compilers.
6829 34. If a DNS black list lookup found a CNAME record, but there were no A
6830 records associated with the domain it pointed at, Exim crashed.
6832 35. If a DNS black list lookup returned more than one A record, Exim ignored
6833 all but the first. It now scans all returned addresses if a particular IP
6834 value is being sought. In this situation, the contents of the
6835 $dnslist_value variable are a list of all the addresses, separated by a
6838 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6839 to accept a host name and all its aliases if the forward lookup for any of
6840 them yielded the IP address of the incoming connection. Now it accepts only
6841 those names whose forward lookup yields the correct IP address. Any other
6842 names are discarded. This closes a loophole whereby a rogue DNS
6843 administrator could create reverse DNS records to break through a
6844 wildcarded host restriction in an ACL.
6846 37. If a user filter or a system filter that ran in a subprocess used any of
6847 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6848 the wrong values were passed to the pipe command ($thisaddress had the
6849 value of $0, $0 had the value of $1, etc). This bug was introduced by
6850 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6852 38. Improved the line breaking for long SMTP error messages from ACLs.
6853 Previously, if there was no break point between 40 and 75 characters, Exim
6854 left the rest of the message alone. Two changes have been made: (a) I've
6855 reduced the minimum length to 35 characters; (b) if it can't find a break
6856 point between 35 and 75 characters, it looks ahead and uses the first one
6857 that it finds. This may give the occasional overlong line, but at least the
6858 remaining text gets split now.
6860 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6861 file descriptors might be low, and that setting 1000 would always raise it.
6862 It turns out that in some environments, the limit is already over 1000 and
6863 that lowering it causes trouble. So now Exim takes care not to decrease it.
6865 40. When delivering a message, the value of $return_path is set to $sender_
6866 address at the start of routing (routers may change the value). By an
6867 oversight, this default was not being set up when an address was tested by
6868 -bt or -bv, which affected the outcome if any router or filter referred to
6871 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6872 log only when "message" or "log_message" are set. However, if one of the
6873 conditions was an address verification, or a call to a nested ACL, the
6874 messages generated by the underlying test were being passed through. This
6875 no longer happens. The underlying message is available in $acl_verify_
6876 message for both "message" and "log_message" expansions, so it can be
6877 passed through if needed.
6879 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6880 new expansion $bh_ to give the encoded byte string without charset
6881 translation. Translation happens only if iconv() is available; HAVE_ICONV
6882 indicates this at build time. HEADERS_CHARSET gives the charset to
6883 translate to; headers_charset can change it in the configuration, and
6884 "headers charset" can change it in an individual filter file.
6886 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6887 that creates RFC 2047 encoded "words" labels them as that charset instead
6888 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6889 expansion operator; (ii) when Exim creates a From: line for a local
6890 message; (iii) when a header line is rewritten to include a "phrase" part.
6892 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6893 buggy, causing it to skip the first lines of messages whose message ID
6894 ended in 'D'. This would not have bitten before Exim release 4.14, because
6895 message IDs were unlikely to end in 'D' before then. The effect was to have
6896 incorrect size information for certain domains.
6898 45. #include "config.h" was missing at the start of the crypt16.c module. This
6899 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6902 46. If there was a timeout during a "random" callout check, Exim treated it as
6903 a failure of the random address, and carried on sending RSET and the real
6904 address. If the delay was just some slowness somewhere, the response to the
6905 original RCPT would be taken as a response to RSET and so on, causing
6906 mayhem of various kinds.
6908 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6909 when I implemented it. It didn't allow for the fact that some option values
6910 may legitimately be negative (e.g. size_addition), and it didn't even do
6911 the right test for positive values.
6913 48. Domain names in DNS records are case-independent. Exim always looks them up
6914 in lower case. Some resolvers return domain names in exactly the case they
6915 appear in the zone file, that is, they may contain uppercase letters. Not
6916 all resolvers do this - some return always lower case. Exim was treating a
6917 change of case by a resolver as a change of domain, similar to a widening
6918 of a domain abbreviation. This triggered its re-routing code and so it was
6919 trying to route what was effectively the same domain again. This normally
6920 caused routing to fail (because the router wouldn't handle the domain
6921 twice). Now Exim checks for this case specially, and just changes the
6922 casing of the domain that it ultimately uses when it transmits the message
6925 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6928 50. If a filter generated a file delivery with a non-absolute name (possible if
6929 no home directory exists for the router), the forbid_file option was not
6932 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6933 the existing equality matching.
6935 52. Exim was using ints instead of ino_t variables in some places where it was
6936 dealing with inode numbers.
6938 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6939 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6940 called TMPDIR, and if it finds it is different, it changes its value.
6942 54. The smtp_printf() function is now made available to local_scan() so
6943 additional output lines can be written before returning. There is also an
6944 smtp_fflush() function to enable the detection of a dropped connection.
6945 The variables smtp_input and smtp_batched_input are exported to
6948 55. Changed the default runtime configuration: the message "Unknown user"
6949 has been removed from the ACL, and instead placed on the localuser router,
6950 using the cannot_route_message feature. This means that any verification
6951 failures that generate their own messages won't get overridden. Similarly,
6952 the "Unrouteable address" message that was in the ACL for unverifiable
6953 relay addresses has also been removed.
6955 56. Added hosts_avoid_esmtp to the smtp transport.
6957 57. The exicyclog script was not checking for the esoteric option
6958 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6959 will work only if exicyclog is run under the appropriate euid.
6961 58. Following a discussion on the list, the rules by which Exim recognises line
6962 endings on incoming messages have been changed. The -dropcr and drop_cr
6963 options are now no-ops, retained only for backwards compatibility. The
6964 following line terminators are recognized: LF CRLF CR. However, special
6965 processing applies to CR:
6967 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6968 nor a local message in the state where . is a terminator.
6970 (ii) If a bare CR is encountered in a header line, an extra space is added
6971 after the line terminator so as not to end the header. The reasoning
6972 behind this is that bare CRs in header lines are most likely either
6973 to be mistakes, or people trying to play silly games.
6975 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6976 was being incorrectly given as 18 bytes larger than it should have been.
6977 This is a VOB (very old bug).
6979 60. This may never have affected anything current, but just in case it has:
6980 When the local host is found other than at the start of a list of hosts,
6981 the local host, those with the same MX, and any that follow, are discarded.
6982 When the list in question was part of a longer list of hosts, the following
6983 hosts (not currently being processed) were also being discarded. This no
6984 longer happens. I'm not sure if this situation could ever has previously
6987 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6990 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6991 header. This is something that is recommended in a new Internet Draft, and
6992 is something that is documented as being done by Sendmail. There are two
6993 possible values. For messages generated by the autoreply transport, Exim
6996 Auto-Submitted: auto-replied
6998 whereas for all other generated messages (e.g. bounces) it adds
7000 Auto-Submitted: auto-generated
7002 63. The "personal" condition in filters now includes a test for the
7003 Auto-Submitted: header. If it contains the string "auto-" the message it
7004 not considered personal.
7006 64. Added rcpt_include_affixes as a generic transport option.
7008 65. Added queue_only_override (default true).
7010 66. Added the syslog_duplication option.
7012 67. If what should have been the first header line of a message consisted of
7013 a space followed by a colon, Exim was mis-interpreting it as a header line.
7014 It isn't of course - it is syntactically invalid and should therefore be
7015 treated as the start of the message body. The misbehaviour could have
7016 caused a number of strange effects, including loss of data in subsequent
7017 header lines, and spool format errors.
7019 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7020 client host had authenticated. This control can now be exercised by an ACL
7021 for more flexibility.
7023 69. By default, callouts do not happen when testing with -bh. There is now a
7024 variant, -bhc, which does actually run the callout code, including
7025 consulting and updating the callout cache.
7027 70. Added support for saslauthd authentication, courtesy of Alexander
7030 71. If statvfs() failed on the spool or log directories while checking their
7031 size for availability, Exim confusingly gave the error "space shortage".
7032 Furthermore, in debugging mode it crashed with a floating point exception.
7033 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7034 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7035 serious problem, Exim now writes to the main and panic logs when this
7036 happens, with details of the failure. It then refuses to accept the
7037 incoming message, giving the message "spool directory problem" or "log
7038 directory problem" with a 421 code for SMTP messages.
7040 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7041 0, 1, and 2 exist, because some OS complain for execs without them (see
7042 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7043 descriptors. However, the code omitted to check that the open succeeded,
7044 causing mysterious errors if for some reason the permissions on /dev/null
7045 got screwed. Now Exim writes a message to the main and panic logs, and
7046 bombs out if it can't open /dev/null.
7048 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7049 interact so that it is all more flexible. It is supposed to remain
7050 backwards compatible. Also added extra_local_interfaces.
7052 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7053 to bomb out with an assertion failure - to the client this appears as a
7054 connection drop. This problem occurs in the part of the code that was taken
7055 from the Samba project. Fortunately, the assertion is in a very simple
7056 function, so I have fixed this by reproducing the function inline in the
7057 one place where it is called, and arranging for authentication to fail
7058 instead of killing the process with assert().
7060 75. The SPA client code was not working when the server requested OEM rather
7061 than Unicode encoding.
7063 76. Added code to make require_files with a specific uid setting more usable in
7064 the case where statting the file as root fails - usually a non-root-mounted
7065 NFS file system. When this happens and the failure is EACCES, Exim now
7066 forks a subprocess and does the per-uid checking as the relevant uid.
7068 77. Added process_log_path.
7070 78. If log_file_path was not explicitly set, a setting of check_log_space or
7071 check_log_inodes was ignored.
7073 79. If a space check for the spool or log partitions fails, the incident is now
7074 logged. Of course, in the latter case the data may get lost...
7076 80. Added the %p formatting code to string_format() so that it can be used to
7077 print addresses in debug_print(). Adjusted all the address printing in the
7078 debugging in store.c to use %p rather than %d.
7080 81. There was a concern that a line of code in smtp_in.c could overflow a
7081 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7082 initially expressed, the concern was not well-founded, because trailing
7083 spaces are removed early. However, if the trailing spaces were followed by
7084 a NULL, they did not get removed, so the overflow was possible. Two fixes
7087 (a) I re-wrote the offending code in a cleaner fashion.
7088 (b) If an incoming SMTP command contains a NULL character, it is rejected
7091 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7092 runs initgroups(), so that if the Exim user is in any additional groups,
7093 they will be used during message reception.
7099 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob