1 Change log file for Exim from version 4.21
2 ------------------------------------------
3 This document describes *changes* to previous versions, that might
4 affect Exim's operation, with an unchanged configuration file. For new
5 options, and new features, see the NewStuff file next to this ChangeLog.
8 Since Exim version 4.90
9 -----------------------
11 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
12 When redis_servers is set to a list of > 1 element, and the Redis servers
13 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
14 case of MOVED into a DEFER case instead, thus moving the query onto the
15 next server in the list. For a cluster of N elements, all N servers must
16 be defined in redis_servers.
18 JH/01 Replace the store_release() internal interface with store_newblock(),
19 which internalises the check required to safely use the old one, plus
20 the allocate and data copy operations duplicated in both (!) of the
23 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
24 modifier. This matches the restriction on the commandline.
26 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
27 Previously only the last row was returned.
29 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
30 we assumed that tags in the header were well-formed, and parsed the
31 element content after inspecting only the first char of the tag.
32 Assumptions at that stage could crash the receive process on malformed
35 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
36 While running the DKIM ACL we operate on the Permanent memory pool so that
37 variables created with "set" persist to the DATA ACL. Also (at any time)
38 DNS lookups that fail create cache records using the Permanent pool. But
39 expansions release any allocations made on the current pool - so a dnsdb
40 lookup expansion done in the DKIM ACL releases the memory used for the
41 DNS negative-cache, and bad things result. Solution is to switch to the
42 Main pool for expansions.
43 While we're in that code, add checks on the DNS cache during store_reset,
44 active in the testsuite.
45 Problem spotted, and debugging aided, by Wolfgang Breyha.
47 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
48 When none of the hosts presented to a transport match an already-open
49 connection, close it and proceed with the list. Previously we would
50 queue the message. Spotted by Lena with Yahoo, probably involving
53 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
54 Previously a spurious "250 OK id=" response was appended to the proper
57 JH/08 The "support for" informational output now, which built with Content
58 Scanning support, has a line for the malware scanner interfaces compiled
59 in. Interface can be individually included or not at build time.
61 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
62 by the template makefile "src/EDITME". The "STREAM" support for an older
63 ClamAV interface method is removed.
65 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
66 rows affected is given instead).
68 JH/11 The runtime Berkeley DB library version is now additionally output by
69 "exim -d -bV". Previously only the compile-time version was shown.
71 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
72 SMTP connection. Previously, when one had more receipients than the
73 first, an abortive onward connection was made. Move to full support for
74 multiple onward connections in sequence, handling cutthrough connection
75 for all multi-message initiating connections.
77 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
78 routers. Previously, a multi-recipient message would fail to match the
79 onward-connection opened for the first recipient, and cause its closure.
81 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
82 a timeout on read on a GnuTLS initiating connection, resulting in the
83 initiating connection being dropped. This mattered most when the callout
84 was marked defer_ok. Fix to keep the two timeout-detection methods
87 JH/15 Relax results from ACL control request to enable cutthrough, in
88 unsupported situations, from error to silently (except under debug)
89 ignoring. This covers use with PRDR, frozen messages, queue-only and
92 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
94 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
95 metadata, resulting in a crash in free().
97 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
98 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
99 Broken also in d185889f4, with init system revamp.
101 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
102 Previously we abruptly closed the connection after reading a malware-
103 found indication; now we go on to read the "scan ok" response line,
106 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
107 ACL. Previously, a crash would result.
109 JH/19 Speed up macro lookups during configuration file read, by skipping non-
110 macro text after a replacement (previously it was only once per line) and
111 by skipping builtin macros when searching for an uppercase lead character.
113 JH/20 DANE support moved from Experimental to mainline. The Makefile control
114 for the build is renamed.
116 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
117 was allocated for every new TLS startup, meaning one per message. Fix
118 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
120 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
121 reported the original. Fix to report (as far as possible) the ACL
122 result replacing the original.
124 JH/23 Fix memory leak during multi-message connections using STARTTLS under
125 OpenSSL. Certificate information is loaded for every new TLS startup,
126 and the resources needed to be freed.
128 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
130 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
133 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
134 DATA response info to the (existing) per-recipient response info for
135 the "C=" log element. It can have useful tracking info from the
136 destination system. Patch from Simon Arlott.
138 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
139 length value. Previously this would segfault.
141 HS/02 Support Avast multiline protoocol, this allows passing flags to
142 newer versions of the scanner.
144 JH/28 Ensure that variables possibly set during message acceptance are marked
145 dead before release of memory in the daemon loop. This stops complaints
146 about them when the debug_store option is enabled. Discovered specifically
147 for sender_rate_period, but applies to a whole set of variables.
148 Do the same for the queue-runner loop, for variables set from spool
155 JH/01 Rework error string handling in TLS interface so that the caller in
156 more cases is responsible for logging. This permits library-sourced
157 string to be attached to addresses during delivery, and collapses
158 pairs of long lines into single ones.
160 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
161 during configuration. Wildcards are allowed and expanded.
163 JH/02 Rework error string handling in DKIM to pass more info back to callers.
164 This permits better logging.
166 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
167 do not close it down and have the child transport start it up again on
168 the passed-on TCP connection. Instead, proxy the child (and any
169 subsequent ones) for TLS via a unix-domain socket channel. Logging is
170 affected: the continued delivery log lines do not have any DNSSEC, TLS
171 Certificate or OCSP information. TLS cipher information is still logged.
173 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
174 identical IP addresses on different listening ports. Will also affect
177 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
178 add noisy ifdef guards to special-case this sillyness.
179 Patch from Bernd Kuhls.
181 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
182 than 255 are no longer allowed.
184 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
185 Disable the session-cache too, which might reduce our load. Since we
186 currrectly use a new context for every connection, both as server and
187 client, there is no benefit for these.
188 GnuTLS appears to not support tickets server-side by default (we don't
189 call gnutls_session_ticket_enable_server()) but client side is enabled
190 by default on recent versions (3.1.3 +) unless the PFS priority string
193 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
194 <https://reproducible-builds.org/specs/source-date-epoch/>.
196 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
197 the check for any unsuccessful recipients did not notice the limit, and
198 erroneously found still-pending ones.
200 JH/08 Pipeline CHUNKING command and data together, on kernels that support
201 MSG_MORE. Only in-clear (not on TLS connections).
203 JH/09 Avoid using a temporary file during transport using dkim. Unless a
204 transport-filter is involved we can buffer the headers in memory for
205 creating the signature, and read the spool data file once for the
206 signature and again for transmission.
208 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
209 4.77 as the kernel support then wasn't solid, having issues in 64bit
210 mode. Now, it's been long enough. Add support for FreeBSD also.
212 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
213 case where the routing stage had gathered several addresses to send to
214 a host before calling the transport for the first, we previously failed
215 to close down TLS in the old transport process before passing the TCP
216 connection to the new process. The new one sent a STARTTLS command
217 which naturally failed, giving a failed delivery and bloating the retry
218 database. Investigation and fix prototype from Wolfgang Breyha.
220 JH/12 Fix check on SMTP command input synchronisation. Previously there were
221 false-negatives in the check that the sender had not preempted a response
222 or prompt from Exim (running as a server), due to that code's lack of
223 awareness of the SMTP input buffering.
225 PP/04 Add commandline_checks_require_admin option.
226 Exim drops privileges sanely, various checks such as -be aren't a
227 security problem, as long as you trust local users with access to their
228 own account. When invoked by services which pass untrusted data to
229 Exim, this might be an issue. Set this option in main configuration
230 AND make fixes to the calling application, such as using `--` to stop
233 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
234 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
235 responses to those, into a single TLS record each way (this usually means
236 a single packet). As a side issue, smtp_enforce_sync now works on TLS
239 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
240 affects you only if you're dancing at the edge of the param size limits.
241 If you are, and this message makes sense to you, then: raise the
242 configured limit or use OpenSSL 1.1. Nothing we can do for older
245 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
246 cmdline element to get the documented default one. Previously it was
249 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
250 are unsuable for non-message-specific reasons - eg. connection timeout,
251 banner-time rejection.
253 JH/16 Fix logging of delivery remote port, when specified by router, under
256 PP/06 Repair manualroute's ability to take options in any order, even if one
257 is the name of a transport.
260 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
262 JH/17 Change the list-building routines interface to use the expanding-string
263 triplet model, for better allocation and copying behaviour.
265 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
266 Previously it was constructed the first time a possibly-matching string
267 was met in the configuration file input during startup; now it is done
270 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
271 compatible one, to avoid the (poorly documented) possibility of a config
272 file in the working directory redirecting the DB files, possibly correpting
273 some existing file. CVE-2017-10140 assigned for BDB.
275 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
276 cache-hot. Previously, although the result was properly cached, the
277 initial verify call returned a defer.
279 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
280 the main verify for receipient in uncached-mode.
282 JH/22 Retire historical build files to an "unsupported" subdir. These are
283 defined as "ones for which we have no current evidence of testing".
285 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
286 if present. Previously it was ignored.
288 JH/24 Start using specified-initialisers in C structure init coding. This is
289 a C99 feature (it's 2017, so now considered safe).
291 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
292 if was a fixed-sized field and bitmask ops via macros; it is now more
295 PP/07 GitHub PR 56: Apply MariaDB build fix.
296 Patch provided by Jaroslav Škarvada.
298 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
299 during Coverity cleanups [4.87 JH/47]
300 Diagnosis and fix provided by Michael Fischer v. Mollard.
302 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
303 the right size to place the terminating semicolon on its own folded
304 line, the header hash was calculated to an incorrect value thanks to
305 the (relaxed) space the fold became.
307 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
308 and confused the parent.
310 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
311 which could crash as a result. This could lead to undeliverable messages.
313 JH/28 Logging: "next input sent too soon" now shows where input was truncated
316 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
317 matters on fast-turnover and PID-randomising systems, which were getting
318 out-of-order delivery.
320 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
321 a possibly-overlapping copy. The symptom was that "Remote host closed
322 connection in response to HELO" was logged instead of the actual 4xx
325 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
326 Previously only that bufferd was discarded, resulting in SYMTP command
329 JH/32 DKIM: when a message has multiple signatures matching an identity given
330 in dkim_verify_signers, run the dkim acl once for each. Previously only
331 one run was done. Bug 2189.
333 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
334 "panic the current process" to "deliberately defer". The panic log is
335 still written with the problem list name; the mail and reject logs now
336 get a temp-reject line for the message that was being handled, saying
337 something like "domains check lookup or other defer". The SMTP 451
338 message is still "Temporary local problem".
340 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
341 A crafted sequence of BDAT commands could result in in-use memory beeing
342 freed. CVE-2017-16943.
344 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
345 from SMTP input. Previously it was always done; now only done for DATA
346 and not BDAT commands. CVE-2017-16944.
348 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
349 to the message (such as an overlong header line). Previously this was
350 not done and we did not exit BDAT mode. Followon from the previous item
351 though a different problem.
357 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
358 than -2003 did; needs libidn2 in addition to libidn.
360 JH/02 The path option on a pipe transport is now expanded before use.
362 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
363 Patch provided by "Björn", documentation fix added too.
365 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
366 missing a wire-to-host endian conversion.
368 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
369 close after a BDAT command line could be taken as a following command,
370 giving a synch failure. Fix by only checking for synch immediately
371 before acknowledging the chunk.
373 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
374 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
375 macro. Patches provided by Josh Soref.
377 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
378 Previously we did not; the RFC seems ambiguous and VRFY is not listed
379 by IANA as a service extension. However, John Klensin suggests that we
382 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
383 the dkim code may be unix-mode line endings rather than smtp wire-format
384 CRLF, so prepend a CR to any bare LF.
386 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
387 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
389 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
390 the first were themselves being wrongly included in the feed into dkim
391 processing; with most chunk sizes in use this resulted in an incorrect
392 body hash calculated value.
394 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
395 DKIM signature block, for verification. Although advised against by
396 standards it is specifically not ruled illegal.
398 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
400 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
401 missing a body hash (the bh= tag).
403 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
404 It seems that HAProxy sends the Proxy Protocol information in clear and
405 only then does a TLS startup, so do the same.
407 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
408 TCP connections (such as for Spamd) unless the daemon successfully set
409 Fast Open mode on its listening sockets. This fixes breakage seen on
410 too-old kernels or those not configured for Fast Open, at the cost of
411 requiring both directions being enabled for TFO, and TFO never being used
412 by non-daemon-related Exim processes.
414 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
415 endings, at least on the first header line. Try to canonify any that get
416 past that check, despite the cost.
418 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
419 now limited to an arbitrary five deep, while parsing addresses with the
420 strip_excess_angle_brackets option enabled.
422 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
423 instead leave the unprompted TLS handshake in socket buffer for the
424 TLS library to consume.
426 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
428 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
430 JH/16 Drop variables when they go out of scope. Memory management drops a whole
431 region in one operation, for speed, and this leaves assigned pointers
432 dangling. Add checks run only under the testsuite which checks all
433 variables at a store-reset and panics on a dangling pointer; add code
434 explicitly nulling out all the variables discovered. Fixes one known
435 bug: a transport crash, where a dangling pointer for $sending_ip_address
436 originally assigned in a verify callout, is re-used.
438 PP/06 Drop '.' from @INC in various Perl scripts.
440 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
442 PP/08 Reduce a number of compilation warnings under clang; building with
443 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
444 should be warning-free.
446 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
448 HS/01 Fix portability problems introduced by PP/08 for platforms where
449 realloc(NULL) is not equivalent to malloc() [SunOS et al].
451 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
452 chunk. This allows us to accept broken chunked messages. We need a more
453 general solution here.
455 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
456 already-broken messages in the queue.
458 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
460 JH/19 Fix reference counting bug in routing-generated-address tracking.
466 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
467 supports it and a size is available (ie. the sending peer gave us one).
469 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
470 years of being deprecated). The replacements are the ACLs
471 acl_smtp_mime and acl_not_smtp_mime.
473 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
474 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
475 or even an in-clear connection were permitted. Now, if the host lookup
476 was dnssec and dane was requested then the host is only used if the
477 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
478 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
479 if one fails this test.
480 This means that a poorly-configured remote DNS will make it incommunicado;
481 but it protects against a DNS-interception attack on it.
483 JH/04 Bug 1810: make continued-use of an open smtp transport connection
484 non-noisy when a race steals the message being considered.
486 JH/05 If main configuration option tls_certificate is unset, generate a
487 self-signed certificate for inbound TLS connections.
489 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
490 in rewrites and routers.
492 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
493 and logged a warning sing 4.83; now they are a configuration file error.
495 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
496 (lacking @domain). Apply the same qualification processing as RCPT.
498 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
500 JH/10 Support ${sha256:} applied to a string (as well as the previous
503 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
504 a cutthrough deliver is pending, as we always want to make a connection.
505 This also avoids re-routing the message when later placing the cutthrough
506 connection after a verify cache hit.
507 Do not update it with the verify result either.
509 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
510 when routing results in more than one destination address.
512 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
513 signing (which inhibits the cutthrough capability). Previously only
514 the presence of an option was tested; now an expansion evaluating as
515 empty is permissible (obviously it should depend only on data available
516 when the cutthrough connection is made).
518 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
519 the relevant preceding SMTP command did not note the pipelining mode.
521 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
522 Previously they were not counted.
524 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
525 as one having no matching records. Previously we deferred the message
526 that needed the lookup.
528 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
529 distinguished as "(=".
531 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
532 for missing MX records. Previously it only worked for missing A records.
534 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
536 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
537 after the data-go-ahead and data-ack. Patch from Jason Betts.
539 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
540 even for a "none" policy. Patch from Tony Meyer.
542 JH/22 Fix continued use of a connection for further deliveries. If a port was
543 specified by a router, it must also match for the delivery to be
546 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
547 When one of the recipients of a message was unsuitable for the connection
548 (has no matching addresses), we lost track of needing to mark it
549 deferred. As a result mail would be lost.
551 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
553 JH/25 Decoding ACL controls is now done using a binary search; the source code
554 takes up less space and should be simpler to maintain. Merge the ACL
555 condition decode tables also, with similar effect.
557 JH/26 Fix problem with one_time used on a redirect router which returned the
558 parent address unchanged. A retry would see the parent address marked as
559 delivered, so not attempt the (identical) child. As a result mail would
562 JH/27 Fix a possible security hole, wherein a process operating with the Exim
563 UID can gain a root shell. Credit to http://www.halfdog.net/ for
564 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
567 JH/28 Enable {spool,log} filesystem space and inode checks as default.
568 Main config options check_{log,spool}_{inodes,space} are now
569 100 inodes, 10MB unless set otherwise in the configuration.
571 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
572 Previously it only applied to the main-section connection policy
575 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
577 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
578 by me. Added RFC7919 DH primes as an alternative.
580 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
581 are not in the system include path.
583 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
584 GnuTLS, when a session startup failed (eg because the client disconnected)
585 Exim did stdio operations after fclose. This was exposed by a recent
586 change which nulled out the file handle after the fclose.
588 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
589 signed directly by the cert-signing cert, rather than an intermediate
590 OCSP-signing cert. This is the model used by LetsEncrypt.
592 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
594 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
595 an incoming connection.
597 HS/02 Bug 1802: Do not half-close the connection after sending a request
600 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
601 fallback to "prime256v1".
603 JH/34 SECURITY: Use proper copy of DATA command in error message.
604 Could leak key material. Remotely exploitable. CVE-2016-9963.
610 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
611 and 3.4.4 - once the server is enabled to respond to an OCSP request
612 it does even when not requested, resulting in a stapling non-aware
613 client dropping the TLS connection.
615 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
616 support variable-length bit vectors. No functional change.
618 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
619 The I= interface field on outgoing lines is now after the H= remote
620 host field, same as incoming lines. There is a separate
621 outgoing_interface log selector which allows you to disable the
624 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
625 If not running log_selector +smtp_connection the mainlog would be held
626 open indefinitely after a "too many connections" event, including to a
627 deleted file after a log rotate. Leave the per net connection logging
628 leaving it open for efficiency as that will be quickly detected by the
629 check on the next write.
631 HS/01 Bug 1671: Fix post transport crash.
632 Processing the wait-<transport> messages could crash the delivery
633 process if the message IDs didn't exist for some reason. When
634 using 'split_spool_directory=yes' the construction of the spool
635 file name failed already, exposing the same netto behaviour.
637 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
638 mime_regex ACL conditions.
640 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
641 to DSN fail messages (bounces): remote IP, remote greeting, remote response
642 to HELO, local diagnostic string.
644 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
645 log line to debug. Even when configured with a tls authenticator many
646 client connections are expected to not authenticate in this way, so
647 an authenticate fail is not an error.
649 HS/02 Add the Exim version string to the process info. This way exiwhat
650 gives some more detail about the running daemon.
652 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
653 matter for fast-change records such as DNSBLs.
655 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
656 retry record, even if constant. There may be multiple transports with
657 different interface settings and the retry behaviour needs to be kept
660 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
662 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
664 JH/10 Bug 840: fix log_defer_output option of pipe transport
666 JH/11 Bug 830: use same host for all RCPTS of a message, even under
667 hosts_randomize. This matters a lot when combined with mua_wrapper.
669 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
670 ${quote_pgsql:<string>} operator.
672 JH/13 Bug 1708: avoid misaligned access in cached lookup.
674 JH/14 Change header file name for freeradius-client. Relevant if compiling
675 with Radius support; from the Gentoo tree and checked under Fedora.
677 JH/15 Bug 1712: Introduce $prdr_requested flag variable
679 JH/16 Bug 1714: Permit an empty string as expansion result for transport
680 option transport_filter, meaning no filtering.
682 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
684 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
685 defaults to "*" (all hosts). The variable is now available when not built
686 with TLS, default unset, mainly to enable keeping the testsuite sane.
687 If a server certificate is not supplied (via tls_certificate) an error is
688 logged, and clients will find TLS connections fail on startup. Presumably
689 they will retry in-clear.
690 Packagers of Exim are strongly encouraged to create a server certificate
691 at installation time.
693 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
694 with the $config_file variable.
696 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
697 in transport context, after the attempt, and per-recipient. The latter type
698 is per host attempted. The event data is the error message, and the errno
699 information encodes the lookup type (A vs. MX) used for the (first) host,
700 and the trailing two digits of the smtp 4xx response.
702 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
703 to write to mainlog (or rejectlog, paniclog) in the window between file
704 creation and permissions/ownership being changed. Particularly affects
705 installations where exicyclog is run as root, rather than exim user;
706 result is that the running daemon panics and dies.
708 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
710 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
711 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
712 "pri" and "weight". Note that the previous implicit priority given by the
713 list order is no longer honoured.
715 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
718 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
719 by defining SUPPORT_SOCKS.
721 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
722 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
723 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
724 variables are renamed to proxy_{local,external}_{address,port}.
726 JH/27 Move Internationalisation support from Experimental to mainline, enabled
727 for a build by defining SUPPORT_I18N
729 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
730 of the query string, and make ${quote_redis:} do that quoting.
732 JH/29 Move Events support from Experimental to mainline, enabled by default
733 and removable for a build by defining DISABLE_EVENT.
735 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
737 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
738 cached by the daemon.
740 JH/32 Move Redis support from Experimental to mainline, enabled for a build
741 by defining LOOKUP_REDIS. The libhiredis library is required.
743 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
744 keys are given for lookup.
746 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
747 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
748 only supported when built with TLS support. The PolarSSL SHA routines
749 are still used when the TLS library is too old for convenient support.
751 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
752 openssl_options), for security. OpenSSL forces this from version 1.1.0
753 server-side so match that on older versions.
755 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
756 allocation for $value could be released as the expansion processing
757 concluded, but leaving the global pointer active for it.
759 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
760 and to use the domains and local_parts ACL conditions.
762 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
763 incorrectly not doubled on cutthrough transmission, hence seen as a
764 body-termination at the receiving system - resulting in truncated mails.
765 Commonly the sender saw a TCP-level error, and retransmitted the message
766 via the normal store-and-forward channel. This could result in duplicates
767 received - but deduplicating mailstores were liable to retain only the
768 initial truncated version.
770 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
772 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
774 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
775 we're in there, support oversigning also; bug 1309.
777 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
779 HS/04 Add support for keep_environment and add_environment options.
781 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
782 either intentional arithmetic overflow during PRNG, or testing config-
785 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
786 delivery resulted in actual delivery. Cancel cutthrough before DATA
789 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
790 rejects a recipient: pass the reject to the originator.
792 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
793 Many were false-positives and ignorable, but it's worth fixing the
796 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
797 for the new environment-manipulation done at startup. Move the routines
798 from being local to tls.c to being global via the os.c file.
800 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
801 an extract embedded as result-arg for a map, the first arg for extract
802 is unavailable so we cannot tell if this is a numbered or keyed
803 extraction. Accept either.
809 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
812 JH/02 The smtp transport option "multi_domain" is now expanded.
814 JH/03 The smtp transport now requests PRDR by default, if the server offers
817 JH/04 Certificate name checking on server certificates, when exim is a client,
818 is now done by default. The transport option tls_verify_cert_hostnames
819 can be used to disable this per-host. The build option
820 EXPERIMENTAL_CERTNAMES is withdrawn.
822 JH/05 The value of the tls_verify_certificates smtp transport and main options
823 default to the word "system" to access the system default CA bundle.
824 For GnuTLS, only version 3.0.20 or later.
826 JH/06 Verification of the server certificate for a TLS connection is now tried
827 (but not required) by default. The verification status is now logged by
828 default, for both outbound TLS and client-certificate supplying inbound
831 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
834 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
835 Status Notification (bounce) messages are now MIME format per RFC 3464.
836 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
837 under the control of the dsn_advertise_hosts option, and routers may
838 have a dsn_lasthop option.
840 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
841 default, modifiable by a malware= option. The list separator for
842 the options can now be changed in the usual way. Bug 68.
844 JH/10 The smtp_receive_timeout main option is now expanded before use.
846 JH/11 The incoming_interface log option now also enables logging of the
847 local interface on delivery outgoing connections.
849 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
850 if the interface and destination host and port all match.
852 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
855 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
856 Patch from Andrew Lewis.
858 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
859 now supports optional time-restrictions, weighting, and priority
860 modifiers per server. Patch originally by <rommer@active.by>.
862 JH/16 The spamd_address main option now supports a mixed list of local
863 and remote servers. Remote servers can be IPv6 addresses, and
864 specify a port-range.
866 JH/17 Bug 68: The spamd_address main option now supports an optional
867 timeout value per server.
869 JH/18 Bug 1581: Router and transport options headers_add/remove can
870 now have the list separator specified.
872 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
875 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
878 JH/21 Support for the A6 type of dns record is withdrawn.
880 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
881 rather than the verbs used.
883 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
884 from 255 to 1024 chars.
886 JH/24 Verification callouts now attempt to use TLS by default.
888 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
889 are generic router options now. The defaults didn't change.
891 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
892 Original patch from Alexander Shikoff, worked over by JH.
894 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
895 files if ZCAT_COMMAND is not executable.
897 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
899 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
901 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
902 Normally benign, it bites when the pair was led to by a CNAME;
903 modern usage is to not canonicalize the domain to a CNAME target
904 (and we were inconsistent anyway for A-only vs AAAA+A).
906 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
908 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
909 when evaluating $sender_host_dnssec.
911 JH/31 Check the HELO verification lookup for DNSSEC, adding new
912 $sender_helo_dnssec variable.
914 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
916 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
918 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
920 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
921 documented as working, but never had. Support all but $spam_report.
923 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
924 added for tls authenticator.
926 HS/03 Add perl_taintmode main config option
932 TL/01 When running the test suite, the README says that variables such as
933 no_msglog_check are global and can be placed anywhere in a specific
934 test's script, however it was observed that placement needed to be near
935 the beginning for it to behave that way. Changed the runtest perl
936 script to read through the entire script once to detect and set these
937 variables, reset to the beginning of the script, and then run through
938 the script parsing/test process like normal.
940 TL/02 The BSD's have an arc4random API. One of the functions to induce
941 adding randomness was arc4random_stir(), but it has been removed in
942 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
943 function when detected.
945 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
946 cause callback expansion.
948 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
949 syntax errors in an expansion can be treated as a string instead of
950 logging or causing an error, due to the internal use of bool_lax
951 instead of bool when processing it.
953 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
954 server certificates when making smtp deliveries.
956 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
958 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
960 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
962 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
963 Merged patch from Sebastian Wiedenroth.
965 JH/05 Fix results-pipe from transport process. Several recipients, combined
966 with certificate use, exposed issues where response data items split
967 over buffer boundaries were not parsed properly. This eventually
968 resulted in duplicates being sent. This issue only became common enough
969 to notice due to the introduction of connection certificate information,
970 the item size being so much larger. Found and fixed by Wolfgang Breyha.
972 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
973 size buffer was used, resulting in syntax errors when an expansion
976 JH/07 Add support for directories of certificates when compiled with a GnuTLS
977 version 3.3.6 or later.
979 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
980 is EXPERIMENTAL_EVENT, the main-configuration and transport options
981 both become "event_action", the variables become $event_name, $event_data
982 and $event_defer_errno. There is a new variable $verify_mode, usable in
983 routers, transports and related events. The tls:cert event is now also
984 raised for inbound connections, if the main configuration event_action
987 TL/06 In test suite, disable OCSP for old versions of openssl which contained
988 early OCSP support, but no stapling (appears to be less than 1.0.0).
990 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
991 server certificate names available under the smtp transport option
992 "tls_verify_cert_hostname" now do not permit multi-component wildcard
995 JH/10 Time-related extraction expansions from certificates now use the main
996 option "timezone" setting for output formatting, and are consistent
997 between OpenSSL and GnuTLS compilations. Bug 1541.
999 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1000 encoded parameter in the incoming message. Bug 1558.
1002 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1003 include certificate info, eximon was claiming there were spoolfile
1006 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1008 JH/14 Log delivery-related information more consistently, using the sequence
1009 "H=<name> [<ip>]" wherever possible.
1011 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1012 are problematic for Debian distribution, omit them from the release
1015 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1017 JH/16 Fix string representation of time values on 64bit time_t architectures.
1020 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1021 output list separator was used.
1026 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1027 checkers that were complaining about end of non-void function with no
1030 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1031 This was a regression introduced in 4.83 by another bugfix.
1033 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1035 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1036 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1042 TF/01 Correctly close the server side of TLS when forking for delivery.
1044 When a message was received over SMTP with TLS, Exim failed to clear up
1045 the incoming connection properly after forking off the child process to
1046 deliver the message. In some situations the subsequent outgoing
1047 delivery connection happened to have the same fd number as the incoming
1048 connection previously had. Exim would try to use TLS and fail, logging
1049 a "Bad file descriptor" error.
1051 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1052 utilities have not been installed.
1054 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1055 temporary space as the ACL may create new global variables.
1057 TL/01 LDAP support uses per connection or global context settings, depending
1058 upon the detected version of the libraries at build time.
1060 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1061 to extract and use the src ip:port in logging and expansions as if it
1062 were a direct connection from the outside internet. PPv2 support was
1063 updated based on HAProxy spec change in May 2014.
1065 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1067 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1068 Properly escape header and check for NULL return.
1070 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1073 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1075 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1076 characters in header names, implemented as a verify condition.
1077 Contributed by Michael Fischer v. Mollard.
1079 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1080 results permerror and temperror. Previous values are deprecated but
1081 still accepted. In a future release, err_perm and err_temp will be
1082 completely removed, which will be a backward incompatibility if the
1083 ACL tests for either of these two old results. Patch contributed by
1084 user bes-internal on the mailing list.
1086 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1088 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1089 selectors, in both main and reject logs.
1091 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1094 JH/07 Add malware type "sock" for talking to simple daemon.
1096 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1098 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1099 routers/transports under cutthrough routing.
1101 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1102 numbers. Touch up "bool" conditional to keep the same definition.
1104 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1106 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1108 JH/12 Expand items in router/transport headers_add or headers_remove lists
1109 individually rather than the list as a whole. Bug 1452.
1111 Required for reasonable handling of multiple headers_ options when
1112 they may be empty; requires that headers_remove items with embedded
1113 colons must have them doubled (or the list-separator changed).
1115 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1116 view the policy declared in the DMARC record. Currently, $dmarc_status
1117 is a combined value of both the record presence and the result of the
1120 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1122 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1123 dnslookup router and the smtp transport (applying to the forward
1126 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1127 of ldap servers used for a specific lookup. Patch provided by Heiko
1130 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1131 New variable $lookup_dnssec_authenticated for observability.
1133 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1134 Patch submitted by Lars Timman.
1136 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1138 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1139 Requires trusted mode and valid format message id, aborts otherwise.
1140 Patch contributed by Heiko Schlichting.
1142 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1143 certextract with support for various fields. Bug 1358.
1145 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1146 is requested by default, modifiable by smtp transport option
1149 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1150 operate on certificate variables to give certificate fingerprints
1151 Also new ${sha256:cert_variable}.
1153 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1155 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1156 Christian Aistleitner.
1158 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1160 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1161 file. Patch from Wolfgang Breyha.
1163 JH/25 Expand the coverage of the delivery $host and $host_address to
1164 client authenticators run in verify callout. Bug 1476.
1166 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1167 align with daemon_smtp_ports. Bug 72.
1169 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1170 support and error reporting did not work properly.
1172 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1173 and is readable. Patch from Andrew Colin Kissa.
1175 TL/14 Enhance documentation of ${run expansion and how it parses the
1176 commandline after expansion, particularly in the case when an
1177 unquoted variable expansion results in an empty value.
1179 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1181 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1182 by renaming away from it. Addresses GH issue 15, reported by
1185 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1187 TL/15 SECURITY: prevent double expansion in math comparison functions
1188 (can expand unsanitized data). Not remotely exploitable.
1195 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1197 PP/02 Make -n do something, by making it not do something.
1198 When combined with -bP, the name of an option is not output.
1200 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1203 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1204 $sender_host_name and config options to manage this, and basic check
1207 PP/05 DSCP support for outbound connections and control modifier for inbound.
1209 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1210 (Only plugin which currently uses this is kerberos4, which nobody should
1211 be using, but we should make it available and other future plugins might
1212 conceivably use it, even though it would break NAT; stuff *should* be
1213 using channel bindings instead).
1215 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1216 name; added for Sendmail compatibility; requires admin caller.
1217 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1218 just ignore it); requires trusted caller.
1219 Also parse but ignore: -Ac -Am -X<logfile>
1222 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1224 TL/02 Add +smtp_confirmation as a default logging option.
1226 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1227 Patch by Magnus Holmgren from 2007-02-20.
1229 TL/04 Bugzilla 1281 - Spec typo.
1230 Bugzilla 1283 - Spec typo.
1231 Bugzilla 1290 - Spec grammar fixes.
1233 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1235 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1237 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1238 mailing list by by Dmitry Isaikin.
1240 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1242 JH/02 Support "G" suffix to numbers in ${if comparisons.
1244 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1246 NM/01 Bugzilla 1197 - Spec typo
1247 Bugzilla 1196 - Spec examples corrections
1249 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1251 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1252 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1255 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1256 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1258 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1259 "acl {{name}{arg}...}", and optional args on acl condition
1262 JH/05 Permit multiple router/transport headers_add/remove lines.
1264 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1266 JH/07 Avoid using a waiting database for a single-message-only transport.
1267 Performance patch from Paul Fisher. Bugzilla 1262.
1269 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1272 JH/09 Add $headers_added variable, with content from use of ACL modifier
1273 add_header (but not yet added to the message). Bugzilla 199.
1275 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1276 Pulled from Bugzilla 817 by Wolfgang Breyha.
1278 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1280 (nb: this is the same fix as in Exim 4.80.1)
1282 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1285 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1288 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1290 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1291 advertise SMTP AUTH mechanism to us, instead of a generic
1292 protocol violation error. Also, make Exim more robust to bad
1293 data from the Dovecot auth socket.
1295 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1297 When a queue runner is handling a message, Exim first routes the
1298 recipient addresses, during which it prunes them based on the retry
1299 hints database. After that it attempts to deliver the message to
1300 any remaining recipients. It then updates the hints database using
1303 So if a recipient address works intermittently, it can get repeatedly
1304 deferred at routing time. The retry hints record remains fresh so the
1305 address never reaches the final cutoff time.
1307 This is a fairly common occurrence when a user is bumping up against
1308 their storage quota. Exim had some logic in its local delivery code
1309 to deal with this. However it did not apply to per-recipient defers
1310 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1312 This change adds a proper retry rule check during routing so that the
1313 final cutoff time is checked against the message's age. We only do
1314 this check if there is an address retry record and there is not a
1315 domain retry record; this implies that previous attempts to handle
1316 the address had the retry_use_local_parts option turned on. We use
1317 this as an approximation for the destination being like a local
1318 delivery, as in LMTP.
1320 I suspect this new check makes the old local delivery cutoff check
1321 redundant, but I have not verified this so I left the code in place.
1323 TF/02 Correct gecos expansion when From: is a prefix of the username.
1325 Test 0254 submits a message to Exim with the header
1329 When I ran the test suite under the user fanf2, Exim expanded
1330 the header to contain my full name, whereas it should have added
1331 a Resent-Sender: header. It erroneously treats any prefix of the
1332 username as equal to the username.
1334 This change corrects that bug.
1336 GF/01 DCC debug and logging tidyup
1337 Error conditions log to paniclog rather than rejectlog.
1338 Debug lines prefixed by "DCC: " to remove any ambiguity.
1340 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1342 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1343 Bug spotted by Jeremy Harris; was flawed since initial commit.
1344 Would have resulted in OCSP responses post-SNI triggering an Exim
1345 NULL dereference and crash.
1347 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1349 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1350 Bug detection, analysis and fix by Samuel Thibault.
1351 Bugzilla 1331, Debian bug #698092.
1353 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1355 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1356 Server implementation by Todd Lyons, client by JH.
1357 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1358 config variable "prdr_enable" controls whether the server
1359 advertises the facility. If the client requests PRDR a new
1360 acl_data_smtp_prdr ACL is called once for each recipient, after
1361 the body content is received and before the acl_smtp_data ACL.
1362 The client is controlled by both of: a hosts_try_prdr option
1363 on the smtp transport, and the server advertisement.
1364 Default client logging of deliveries and rejections involving
1365 PRDR are flagged with the string "PRDR".
1367 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1368 fclose(). Diagnosis by Todd Lyons.
1370 PP/17 Update configure.default to handle IPv6 localhost better.
1371 Patch by Alain Williams (plus minor tweaks).
1374 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1375 This is now consistent with GnuTLS, and is now documented: the
1376 previous undocumented portable approach to treating the option as
1377 unset was to force an expansion failure. That still works, and
1378 an empty string is now equivalent.
1380 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1381 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1382 not performing validation itself.
1384 PP/20 Added force_command boolean option to pipe transport.
1385 Patch from Nick Koston, of cPanel Inc.
1387 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1390 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1392 PP/21 Fix eximon continuous updating with timestamped log-files.
1393 Broken in a format-string cleanup in 4.80, missed when I repaired the
1394 other false fix of the same issue.
1395 Report and fix from Heiko Schlichting.
1398 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1399 Report from Prashanth Katuri.
1401 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1402 It's SecureTransport, so affects any MacOS clients which use the
1403 system-integrated TLS libraries, including email clients.
1405 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1406 using a MIME ACL for non-SMTP local injection.
1407 Report and assistance in diagnosis by Warren Baker.
1409 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1411 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1413 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1414 last id that failed so it may be referenced in subsequent ACL's.
1416 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1419 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1420 ldap library initialization, allowing self-signed CA's to be
1421 used. Also properly sets require_cert option later in code by
1422 using NULL (global ldap config) instead of ldap handle (per
1423 session). Bug diagnosis and testing by alxgomz.
1425 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1426 the src/util/ subdirectory.
1428 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1429 renamed to Transport Post Delivery Action by Jeremy Harris, as
1432 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1433 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1434 redis_servers = needs to be configured which will be used by the redis
1435 lookup. Patch from Warren Baker, of The Packet Hub.
1437 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1439 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1440 hostname or reverse DNS when processing a host list. Used suggestions
1441 from multiple comments on this bug.
1443 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1445 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1446 Missed a few lines, added it to make the runtest require no keyboard
1449 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1450 contains upper case chars. Make router use caseful_local_part.
1452 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1453 support when GnuTLS has been built with p11-kit.
1459 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1461 This, or similar/improved, will also be change PP/11 of 4.82.
1467 PP/01 Handle short writes when writing local log-files.
1468 In practice, only affects FreeBSD (8 onwards).
1469 Bugzilla 1053, with thanks to Dmitry Isaikin.
1471 NM/01 Bugzilla 949 - Documentation tweak
1473 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1476 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1478 PP/02 Implemented gsasl authenticator.
1480 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1482 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1483 `pkg-config foo` for cflags/libs.
1485 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1486 with rest of GSASL and with heimdal_gssapi.
1488 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1489 `pkg-config foo` for cflags/libs for the TLS implementation.
1491 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1492 properties get this fed in as external SSF. A number of robustness
1493 and debugging improvements to the cyrus_sasl authenticator.
1495 PP/08 cyrus_sasl server now expands the server_realm option.
1497 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1498 Patch by Jeremy Harris.
1500 PP/10 Added dbmjz lookup type.
1502 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1504 PP/12 MAIL args handles TAB as well as SP, for better interop with
1505 non-compliant senders.
1506 Analysis and variant patch by Todd Lyons.
1508 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1509 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1510 Patch from Dirk Mueller <dmueller@suse.com>
1512 PP/13 tls_peerdn now print-escaped for spool files.
1513 Observed some $tls_peerdn in wild which contained \n, which resulted
1514 in spool file corruption.
1516 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1517 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1518 or write after TLS renegotiation, which otherwise led to messages
1521 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1522 as a tracking header (ie: a signed header comes before the signature).
1523 Patch from Wolfgang Breyha.
1525 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1526 comma-sep list; embedded commas doubled.
1528 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1530 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1532 Report and patch from Dmitry Banschikov.
1534 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1535 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1536 failures. We appear to now support TLS1.1+ with Exim.
1538 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1539 lets Exim select keys and certificates based upon TLS SNI from client.
1540 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1541 before an outbound SMTP session. New log_selector, +tls_sni.
1543 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1544 NULL dereference. Report and patch from Alun Jones.
1546 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1547 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1548 Not seeing resolver debug output on NetBSD, but suspect this is a
1549 resolver implementation change.
1551 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1552 Left warnings. Added "eximon gdb" invocation mode.
1554 PP/21 Defaulting "accept_8bitmime" to true, not false.
1556 PP/22 Added -bw for inetd wait mode support.
1558 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1559 locate the relevant includes and libraries. Made this the default.
1561 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1562 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1564 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1565 This may cause build issues on older platforms.
1567 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1568 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1569 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1570 Added SNI support via GnuTLS too.
1571 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1573 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1575 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1576 Applied second patch from Janne, implementing suggestion to default
1577 multiple-strings-in-record handling to match SPF spec.
1579 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1581 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1582 read-only, out of scope).
1583 Patch from Wolfgang Breyha, report from Stuart Northfield.
1585 PP/29 Fix three issues highlighted by clang analyser static analysis.
1586 Only crash-plausible issue would require the Cambridge-specific
1587 iplookup router and a misconfiguration.
1588 Report from Marcin Mirosław.
1590 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1592 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1593 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1594 As part of this, removing so much warning spew let me fix some minor
1595 real issues in debug logging.
1597 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1598 assignment on my part. Fixed.
1600 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1601 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1602 Janne Snabb (who went above and beyond: thank you).
1604 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1605 string otherwise requires a connection and a bunch more work and it's
1606 relatively easy to get wrong. Should also expose TLS library linkage
1609 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1610 64-bit ${eval} (JH/03).
1612 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1613 GNU libc to support some of the 64-bit stuff, should not lead to
1614 conflicts. Defined before os.h is pulled in, so if a given platform
1615 needs to override this, it can.
1617 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1618 protection layer was required, which is not implemented.
1619 Bugzilla 1254, patch from Wolfgang Breyha.
1621 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1622 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1623 tls_dhparam take prime identifiers. Also unbreak combination of
1624 OpenSSL+DH_params+TLSSNI.
1626 PP/39 Disable SSLv2 by default in OpenSSL support.
1632 PP/01 Solaris build fix for Oracle's LDAP libraries.
1633 Bugzilla 1109, patch from Stephen Usher.
1635 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1637 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1640 TF/02 Fix a couple more cases where we did not log the error message
1641 when unlink() failed. See also change 4.74-TF/03.
1643 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1644 lock up or crash if it happened to be inside a call to libc when it
1645 got a SIGUSR1 from exiwhat.
1647 The SIGUSR1 handler appends the current process status to the process
1648 log which is later printed by exiwhat. It used to use the general
1649 purpose logging code to do this, but several functions it calls are
1650 not safe for signals.
1652 The new output code in the SIGUSR1 handler is specific to the process
1653 log, and simple enough that it's easy to inspect for signal safety.
1654 Removing some special cases also simplifies the general logging code.
1655 Removing the spurious timestamps from the process log simplifies
1658 TF/04 Improved ratelimit ACL condition.
1660 The /noupdate option has been deprecated in favour of /readonly which
1661 has clearer semantics. The /leaky, /strict, and /readonly update modes
1662 are mutually exclusive. The update mode is no longer included in the
1663 database key; it just determines when the database is updated. (This
1664 means that when you upgrade Exim will forget old rate measurements.)
1666 Exim now checks that the per_* options are used with an update mode that
1667 makes sense for the current ACL. For example, when Exim is processing a
1668 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1669 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1670 must specify per_mail/readonly. If you omit the update mode it defaults to
1671 /leaky where that makes sense (as before) or /readonly where required.
1673 The /noupdate option is now undocumented but still supported for
1674 backwards compatibility. It is equivalent to /readonly except that in
1675 ACLs where /readonly is required you may specify /leaky/noupdate or
1676 /strict/noupdate which are treated the same as /readonly.
1678 A useful new feature is the /count= option. This is a generalization
1679 of the per_byte option, so that you can measure the throughput of other
1680 aggregate values. For example, the per_byte option is now equivalent
1681 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1683 The per_rcpt option has been generalized using the /count= mechanism
1684 (though it's more complicated than the per_byte equivalence). When it is
1685 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1686 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1687 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1688 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1689 Note that using per_rcpt with a non-readonly update mode in more than
1690 one ACL will cause the recipients to be double-counted. (The per_mail
1691 and per_byte options don't have this problem.)
1693 The handling of very low rates has changed slightly. If the computed rate
1694 is less than the event's count (usually one) then this event is the first
1695 after a long gap. In this case the rate is set to the same as this event's
1696 count, so that the first message of a spam run is counted properly.
1698 The major new feature is a mechanism for counting the rate of unique
1699 events. The new per_addr option counts the number of different
1700 recipients that someone has sent messages to in the last time period. It
1701 behaves like per_rcpt if all the recipient addresses are different, but
1702 duplicate recipient addresses do not increase the measured rate. Like
1703 the /count= option this is a general mechanism, so the per_addr option
1704 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1705 example, measure the rate that a client uses different sender addresses
1706 with the options per_mail/unique=$sender_address. There are further
1707 details in the main documentation.
1709 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1711 TF/06 Removed a few PCRE remnants.
1713 TF/07 Automatically extract Exim's version number from tags in the git
1714 repository when doing development or release builds.
1716 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1717 Bugzilla 879. Patch from Paul Fisher.
1719 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1720 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1723 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1725 PP/05 Variable $av_failed, true if the AV scanner deferred.
1726 Bugzilla 1078. Patch from John Horne.
1728 PP/06 Stop make process more reliably on build failure.
1729 Bugzilla 1087. Patch from Heiko Schlittermann.
1731 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1732 Bugzilla 1089. Patch from Heiko Schlittermann.
1734 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1735 Bugzilla 1131. Patch from Holger Weiß.
1737 PP/09 Handle IPv6 addresses with SPF.
1738 Bugzilla 860. Patch from Wolfgang Breyha.
1740 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1742 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1745 PP/11 match_* no longer expand right-hand-side by default.
1746 New compile-time build option, EXPAND_LISTMATCH_RHS.
1747 New expansion conditions, "inlist", "inlisti".
1749 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1751 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1753 PP/14 fix log_write() format string regression from TF/03.
1754 Bugzilla 1152. Patch from Dmitry Isaikin.
1760 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1762 PP/02 Harmonised TLS library version reporting; only show if debugging.
1763 Layout now matches that introduced for other libraries in 4.74 PP/03.
1765 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1767 PP/04 New "dns_use_edns0" global option.
1769 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1772 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1773 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1775 TK/01 Updated PolarSSL code to 0.14.2.
1776 Bugzilla 1097. Patch from Andreas Metzler.
1778 PP/07 Catch divide-by-zero in ${eval:...}.
1779 Fixes bugzilla 1102.
1781 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1784 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1785 format-string attack -- SECURITY: remote arbitrary code execution.
1787 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1788 time unintentionally subject to list matching rules, letting the header
1789 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1790 arbitrary string expansion). This allowed for information disclosure.
1792 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1793 INT_MIN/-1 -- value coerced to INT_MAX.
1799 NM/01 Workaround for PCRE version dependency in version reporting
1802 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1803 This fixes portability to compilers other than gcc, notably
1804 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1806 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1807 makefiles for portability to HP-UX and POSIX correctness.
1809 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1810 Also via indented variable definition in the Makefile.
1811 (Debugging by Oliver Heesakkers).
1813 PP/02 Restore caching of spamd results with expanded spamd_address.
1814 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1816 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1817 Improves build reliability. Fix from: Frank Elsner
1819 NM/02 Fix wide character breakage in the rfc2047 coding
1820 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1822 NM/03 Allow underscore in dnslist lookups
1823 Fixes bug 1026. Patch from Graeme Fowler
1825 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1826 Code patches from Adam Ciarcinski of NetBSD.
1828 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1831 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1832 is logged, to avoid truncation. Patch from John Horne.
1834 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1835 Patch from Jakob Hirsch.
1837 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1838 SQL string expansion failure details.
1839 Patch from Andrey Oktyabrski.
1841 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1842 Patch from Simon Arlott.
1844 PP/09 New lookups functionality failed to compile on old gcc which rejects
1845 extern declarations in function scope.
1846 Patch from Oliver Fleischmann
1848 PP/10 Use sig_atomic_t for flags set from signal handlers.
1849 Check getgroups() return and improve debugging.
1850 Fixed developed for diagnosis in bug 927 (which turned out to be
1853 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1854 Patch from Mark Zealey.
1856 PP/12 Bugzilla 1056: Improved spamd server selection.
1857 Patch from Mark Zealey.
1859 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1860 Based on patch from Heiko Schlittermann.
1862 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1863 Patch from Uwe Doering, sign-off by Michael Haardt.
1865 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1866 variable declaration deep within a block. Bug and patch from
1869 PP/15 lookups-Makefile IRIX compatibility coercion.
1871 PP/16 Make DISABLE_DKIM build knob functional.
1873 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1874 Patch by Simon Arlott
1876 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1877 variable argument macros. Our copy now differs from upstream.
1883 TF/01 Failure to get a lock on a hints database can have serious
1884 consequences so log it to the panic log.
1886 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1887 controlled using the smtp_confirmation log selector.
1889 TF/03 Include the error message when we fail to unlink a spool file.
1891 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1892 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1893 for maintaining out-of-tree patches for some time.
1895 PP/01 Bugzilla 139: Documentation and portability issues.
1896 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1897 Handle per-OS dynamic-module compilation flags.
1899 PP/02 Let /dev/null have normal permissions.
1900 The 4.73 fixes were a little too stringent and complained about the
1901 permissions on /dev/null. Exempt it from some checks.
1902 Reported by Andreas M. Kirchwitz.
1904 PP/03 Report version information for many libraries, including
1905 Exim version information for dynamically loaded libraries. Created
1906 version.h, now support a version extension string for distributors
1907 who patch heavily. Dynamic module ABI change.
1909 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1910 privilege escalation vulnerability whereby the Exim run-time user
1911 can cause root to append content of the attacker's choosing to
1914 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1917 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1918 If dropping privileges for untrusted macros, we disabled normal logging
1919 on the basis that it would fail; for the Exim run-time user, this is not
1920 the case, and it resulted in successful deliveries going unlogged.
1921 Fixed. Reported by Andreas Metzler.
1927 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1928 only prepend for the Resent-* case. Fixes regression introduced in
1929 Exim 4.70 by NM/22 for Bugzilla 607.
1931 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1932 increasing numbers of administrators be bitten by this.
1934 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1936 PP/03 Bugzilla 994: added openssl_options main configuration option.
1938 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1940 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1942 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1944 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1945 without a peer certificate, leading to a segfault because of an
1946 assumption that peers always have certificates. Be a little more
1947 paranoid. Problem reported by Martin Tscholak.
1949 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1950 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1951 NB: ClamAV planning to remove STREAM in "middle of 2010".
1952 CL also introduces -bmalware, various -d+acl logging additions and
1953 more caution in buffer sizes.
1955 PP/09 Implemented reverse_ip expansion operator.
1957 PP/10 Bugzilla 937: provide a "debug" ACL control.
1959 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1961 PP/12 Bugzilla 973: Implement --version.
1963 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1965 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1967 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1969 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1970 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1971 ignore trailing whitespace.
1973 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1975 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1978 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1979 Notification from Dr Andrew Aitchison.
1981 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1982 ExtendedDetectionInfo response format.
1983 Notification from John Horne.
1985 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1988 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1989 XSL and documented dependency on system catalogs, with examples of how
1992 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1995 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1996 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1997 configuration file which is writeable by the Exim user or group.
1999 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2000 of configuration files to cover files specified with the -C option if
2001 they are going to be used with root privileges, not just the default
2004 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2005 option (effectively making it always true).
2007 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2008 files to be used while preserving root privileges.
2010 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2011 that rogue child processes cannot use them.
2013 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2014 run-time user, instead of root.
2016 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2017 Exim run-time user without dropping privileges.
2019 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2020 result string, instead of calling string_vformat() twice with the same
2023 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2024 for other users. Others should always drop root privileges if they use
2025 -C on the command line, even for a whitelisted configure file.
2027 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2029 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2035 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2036 $data_path, and $header_path variables; fixed documentation bugs and
2039 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2040 exipick to access non-standard spools, including the "frozen" queue
2043 NM/01 Bugzilla 965: Support mysql stored procedures.
2044 Patch from Alain Williams
2046 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2048 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2049 Patch from Andreas Metzler
2051 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2052 Patch from Kirill Miazine
2054 NM/05 Bugzilla 671: Added umask to procmail example.
2056 JJ/03 installed exipick 20100323.0, fixing doc bug
2058 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2059 directory. Notification and patch from Dan Rosenberg.
2061 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2063 TK/02 Improve log output when DKIM signing operation fails.
2065 MH/01 Treat the transport option dkim_domain as a colon separated
2066 list, not as a single string, and sign the message with each element,
2067 omitting multiple occurences of the same signer.
2069 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2070 Bugzilla 985, 986. Patch by Simon Arlott
2072 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2073 Patch by Simon Arlott
2075 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2076 MBX locking. Notification from Dan Rosenberg.
2082 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2084 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2086 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2088 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2090 NM/04 Bugzilla 915: Flag broken perl installation during build.
2096 TK/01 Added patch by Johannes Berg that expands the main option
2097 "spamd_address" if it starts with a dollar sign.
2099 TK/02 Write list of recipients to X-Envelope-Sender header when building
2100 the mbox-format spool file for content scanning (suggested by Jakob
2103 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2104 (http://www.dcc-servers.net/) support via dccifd. Activated by
2105 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2107 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2108 by Mark Daniel Reidel <mr@df.eu>.
2110 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2111 When building exim an external PCRE library is now needed -
2112 PCRE is a system library on the majority of modern systems.
2113 See entry on PCRE_LIBS in EDITME file.
2115 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2116 conversation. Added nologin parameter to request.
2117 Patch contributed by Kirill Miazine.
2119 TF/01 Do not log submission mode rewrites if they do not change the address.
2121 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2123 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2124 log files in place. Contributed by Roberto Lima.
2126 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2128 TF/03 Bugzilla 615: When checking the local_parts router precondition
2129 after a local_part_suffix or local_part_prefix option, Exim now
2130 does not use the address's named list lookup cache, since this
2131 contains cached lookups for the whole local part.
2133 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2134 Robert Millan. Documentation is in experimental-spec.txt.
2136 TF/04 Bugzilla 668: Fix parallel build (make -j).
2138 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2140 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2141 Patch provided by Jan Srzednicki.
2143 TF/05 Leading white space used to be stripped from $spam_report which
2144 wrecked the formatting. Now it is preserved.
2146 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2147 that they are available at delivery time.
2149 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2151 TF/08 TLS error reporting now respects the incoming_interface and
2152 incoming_port log selectors.
2154 TF/09 Produce a more useful error message if an SMTP transport's hosts
2155 setting expands to an empty string.
2157 NM/06 Bugzilla 744: EXPN did not work under TLS.
2158 Patch provided by Phil Pennock.
2160 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2161 Patch provided by Richard Godbee.
2163 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2164 acl_smtp_notquit, added index entry.
2166 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2167 Patch provided by Eugene Bujak.
2169 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2170 accept(). Patch provided by Maxim Dounin.
2172 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2173 Patch provided by Phil Pennock.
2175 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2177 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2178 Patch provided by Brad "anomie" Jorsch.
2180 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2181 Patch provided by Dean Brooks.
2183 TK/05 Add native DKIM support (does not depend on external libraries).
2185 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2186 Patch provided by Graeme Fowler.
2188 NM/16 Bugzilla 851: Documentation example syntax fix.
2190 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2192 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2195 NM/19 Bugzilla 745: TLS version reporting.
2196 Patch provided by Phil Pennock.
2198 NM/20 Bugzilla 167: bool: condition support.
2199 Patch provided by Phil Pennock.
2201 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2202 clients. Patch provided by Phil Pennock.
2204 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2205 Patch provided by Brad "anomie" Jorsch.
2207 NM/23 Bugzilla 687: Fix misparses in eximstats.
2208 Patch provided by Heiko Schlittermann.
2210 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2211 Patch provided by Heiko Schlittermann.
2213 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2214 plus update to original patch.
2216 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2218 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2219 Patch provided by David Brownlee.
2221 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2223 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2225 NM/30 Bugzilla 888: TLS documentation bugfixes.
2227 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2229 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2230 Unlike the original bugzilla I have changed all shell scripts in src tree.
2232 NM/33 Bugzilla 898: Transport filter timeout fix.
2233 Patch by Todd Rinaldo.
2235 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2236 Patch by Serge Demonchaux.
2238 NM/35 Bugzilla 39: Base64 decode bug fixes.
2239 Patch by Jakob Hirsch.
2241 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2243 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2245 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2247 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2253 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2254 ALT-N's libdkim that I have put here:
2255 http://duncanthrax.net/exim-experimental/
2257 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2258 were called 'true' and it seems that C99 defines that as a reserved
2259 keyword to be used with 'bool' variable types. That means you could
2260 not include C99-style headers which use bools without triggering
2261 build errors in sieve.c.
2263 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2264 as mailq or other aliases. Changed the --help handling significantly
2265 to do whats expected. exim_usage() emits usage/help information.
2267 SC/01 Added the -bylocaldomain option to eximstats.
2269 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2271 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2273 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2279 PH/01 Another patch from the Sieve maintainer.
2281 PH/02 When an IPv6 address is converted to a string for single-key lookup
2282 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2283 dots are used instead of colons so that keys in lsearch files need not
2284 contain colons. This was done some time before quoting was made available
2285 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2286 (notated using the quote facility) so as to distinguish them from IPv4
2287 keys. This meant that lookups for IP addresses in host lists did not work
2288 for iplsearch lookups.
2290 This has been fixed by arranging for IPv6 addresses to be expressed with
2291 colons if the lookup type is iplsearch. This is not incompatible, because
2292 previously such lookups could never work.
2294 The situation is now rather anomalous, since one *can* have colons in
2295 ordinary lsearch keys. However, making the change in all cases is
2296 incompatible and would probably break a number of configurations.
2298 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2301 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2302 conversion specification without a maximum field width, thereby enabling
2303 a rogue spamd server to cause a buffer overflow. While nobody in their
2304 right mind would setup Exim to query an untrusted spamd server, an
2305 attacker that gains access to a server running spamd could potentially
2306 exploit this vulnerability to run arbitrary code as the Exim user.
2308 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2309 $primary_hostname instead of what libspf2 thinks the hosts name is.
2311 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2312 a directory entry by the name of the lookup key. Previously, if a
2313 symlink pointed to a non-existing file or a file in a directory that
2314 Exim lacked permissions to read, a lookup for a key matching that
2315 symlink would fail. Now it is enough that a matching directory entry
2316 exists, symlink or not. (Bugzilla 503.)
2318 PH/03 The body_linecount and body_zerocount variables are now exported in the
2321 PH/04 Added the $dnslist_matched variable.
2323 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2324 This means they are set thereafter only if the connection becomes
2327 PH/06 Added the client_condition to authenticators so that some can be skipped
2328 by clients under certain conditions.
2330 PH/07 The error message for a badly-placed control=no_multiline_responses left
2331 "_responses" off the end of the name.
2333 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2335 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2336 (without spaces) instead of just copying the configuration text.
2338 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2340 PH/11 Added $max_received_linelength.
2342 PH/12 Added +ignore_defer and +include_defer to host lists.
2344 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2345 way in which PCRE > 7.0 is built.
2347 PH/14 Implemented queue_only_load_latch.
2349 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2350 MAIL command. The effect was to mangle the value on 64-bit systems.
2352 PH/16 Another patch from the Sieve maintainer.
2354 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2356 PH/18 If a system quota error occurred while trying to create the file for
2357 a maildir delivery, the message "Mailbox is full" was not appended to the
2358 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2359 only to a quota excession during the actual writing of the file.
2361 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2362 characters?) which causes problems in log lines. The DN values are now
2363 passed through string_printing() before being added to log lines.
2365 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2366 and InterBase are left for another time.)
2368 PH/21 Added message_body_newlines option.
2370 PH/22 Guard against possible overflow in moan_check_errorcopy().
2372 PH/23 POSIX allows open() to be a macro; guard against that.
2374 PH/24 If the recipient of an error message contained an @ in the local part
2375 (suitably quoted, of course), incorrect values were put in $domain and
2376 $local_part during the evaluation of errors_copy.
2382 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2383 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2386 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2387 issue a MAIL command.
2389 PH/02 In an ACL statement such as
2391 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2393 if a client was not listed at all, or was listed with a value other than
2394 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2395 the condition was not true (as it should be), so access was not denied.
2396 The bug was that the ! inversion was incorrectly passed on to the second
2397 item. This has been fixed.
2399 PH/03 Added additional dnslists conditions == and =& which are different from
2400 = and & when the dns lookup returns more than one IP address.
2402 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2403 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2405 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2406 FSYNC, which compiles an option called disable_fsync that allows for
2407 bypassing fsync(). The documentation is heavily laced with warnings.
2409 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2411 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2412 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2413 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2414 including adding "make clean"; (3) Added -fPIC when compiling the test
2415 dynamically loaded module, to get rid of a warning.
2417 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2418 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2419 = 0s. The bug is otherwise harmless.
2421 PH/07 There was a bug in the dovecot authenticator such that the value of
2422 $auth1 could be overwritten, and so not correctly preserved, after a
2423 successful authentication. This usually meant that the value preserved by
2424 the server_setid option was incorrect.
2426 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2428 PH/09 Installed PCRE release 7.0.
2430 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2431 run for batched SMTP input. It is now run at the start of every message
2432 in the batch. While fixing this I discovered that the process information
2433 (output by running exiwhat) was not always getting set for -bs and -bS
2434 input. This is fixed, and it now also says "batched" for BSMTP.
2436 PH/11 Added control=no_pipelining.
2438 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2439 patch, slightly modified), and move the expansion of helo_data till after
2440 the connection is made in the smtp transport (so it can use these
2443 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2445 PH/14 Added log_selector = +pid.
2447 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2449 PH/16 Add ${if forany and ${if forall.
2451 PH/17 Added dsn_from option to vary the From: line in DSNs.
2453 PH/18 Flush SMTP output before performing a callout, unless control =
2454 no_callout_flush is set.
2456 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2457 was true (the default) a successful delivery failed to delete the retry
2458 item, thus causing premature timeout of the address. The bug is now
2461 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2463 PH/21 Long custom messages for fakedefer and fakereject are now split up
2464 into multiline responses in the same way that messages for "deny" and
2465 other ACL rejections are.
2467 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2468 with slight modification.
2470 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2471 draft, changing the syntax and factoring some duplicate code".
2473 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2474 for deliveries of the second and subsequent messages over the same SMTP
2477 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2478 ${reduce, with only minor "tidies".
2480 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2482 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2483 expansion side effects.
2485 PH/27 When a message times out after an over-quota error from an Exim-imposed
2486 quota, the bounce message says "mailbox is full". This message was not
2487 being given when it was a system quota that was exceeded. It now should
2490 MH/03 Made $recipients available in local_scan(). local_scan() already has
2491 better access to the recipient list through recipients_list[], but
2492 $recipients can be useful in postmaster-provided expansion strings.
2494 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2495 in the case of a MAIL command with additional options following the
2496 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2497 were accidentally chopped off.
2499 PH/29 SMTP synchronization checks are implemented when a command is read -
2500 there is a check that no more input is waiting when there shouldn't be
2501 any. However, for some commands, a delay in an ACL can mean that it is
2502 some time before the response is written. In this time, more input might
2503 arrive, invalidly. So now there are extra checks after an ACL has run for
2504 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2505 pipelining has not been advertised.
2507 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2509 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2510 returned in the SMTP response when smtp_return_error_details was false.
2511 This has been fixed.
2513 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2514 instead of the C I/O that was originally supplied, because problems were
2515 reported on Solaris.
2517 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2518 Exim which did not show up earlier: it was assuming that a call to
2519 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2520 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2521 was a macro that became an assignment, so it seemed to work. This has
2522 changed to a proper function call with a void return, hence the compile
2523 error. Exim's code has been fixed.
2525 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2528 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2530 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2532 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2533 because it turns out to be too big or there is a timeout) there is no
2534 "Completed" line in the log. When some messages of this type were
2535 selected by exigrep, they were listed as "not completed". Others were
2536 picked up by some special patterns. I have improved the selection
2537 criteria to be more general.
2539 PH/38 The host_find_failed option in the manualroute router can now be set
2540 to "ignore", to completely ignore a host whose IP address cannot be
2541 found. If all hosts are ignored, the behaviour is controlled by the new
2542 host_all_ignored option.
2544 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2545 homing or because of multiple MX records with /mx) generated more than
2546 one IP address, and the following item turned out to be the local host,
2547 all the secondary addresses of the first item were incorrectly removed
2548 from the list, along with the local host and any following hosts (which
2549 is what is supposed to happen).
2551 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2552 whoever called Exim into the -H file. In the case of the daemon it was
2553 behaving confusingly. When first started, it used values for whoever
2554 started the daemon, but after a SIGHUP it used the Exim user (because it
2555 calls itself on a restart). I have changed the code so that it now always
2558 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2559 message are rejected with the same error (e.g. no authentication or bad
2560 sender address), and a DATA command is nevertheless sent (as can happen
2561 with PIPELINING or a stupid MUA), the error message that was given to the
2562 RCPT commands is included in the rejection of the DATA command. This is
2563 intended to be helpful for MUAs that show only the final error to their
2566 PH/42 Another patch from the Sieve maintainer.
2568 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2569 Eximstats - Fixed some broken HTML links and added missing column headers
2571 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2572 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2574 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2576 PH/43 Yet another patch from the Sieve maintainer.
2578 PH/44 I found a way to check for a TCP/IP connection going away before sending
2579 the response to the final '.' that terminates a message, but only in the
2580 case where the client has not sent further data following the '.'
2581 (unfortunately, this is allowed). However, in many cases there won't be
2582 any further data because there won't be any more messages to send. A call
2583 to select() can be used: if it shows that the input is "ready", there is
2584 either input waiting, or the socket has been closed. An attempt to read
2585 the next input character can distinguish the two cases. Previously, Exim
2586 would have sent an OK response which the client would never have see.
2587 This could lead to message repetition. This fix should cure that, at
2588 least in a lot of common cases.
2590 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2591 advertised in response to EHLO.
2597 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2598 fixed by 4.65/MH/01 (is this a record?) are fixed:
2600 (i) An empty string was always treated as zero by the numeric comparison
2601 operators. This behaviour has been restored.
2603 (ii) It is documented that the numeric comparison operators always treat
2604 their arguments as decimal numbers. This was broken in that numbers
2605 starting with 0 were being interpreted as octal.
2607 While fixing these problems I realized that there was another issue that
2608 hadn't been noticed. Values of message_size_limit (both the global option
2609 and the transport option) were treated as octal if they started with 0.
2610 The documentation was vague. These values are now always treated as
2611 decimal, and I will make that clear in the documentation.
2617 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2618 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2621 MH/01 Don't check that the operands of numeric comparison operators are
2622 integers when their expansion is in "skipping" mode (fixes bug
2623 introduced by 4.64-PH/07).
2625 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2626 child addresses, Exim now panics and dies. Previously, because the count
2627 is held in a short int, deliveries were likely to be lost. As such a
2628 large number of recipients for a single message is ridiculous
2629 (performance will be very, very poor), I have chosen to impose a limit
2630 rather than extend the field.
2636 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2637 leftover -K file (the existence of which was triggered by #402).
2638 While we were at it, introduced process PID as part of the -K
2639 filename. This should rule out race conditions when creating
2642 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2643 processing considerably. Previous code took too long for large mails,
2644 triggering a timeout which in turn triggers #401.
2646 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2647 in the DK code in transports.c. sendfile() is not really portable,
2648 hence the _LINUX specificness.
2650 TF/01 In the add_headers option to the mail command in an Exim filter,
2651 there was a bug that Exim would claim a syntax error in any
2652 header after the first one which had an odd number of characters
2655 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2656 callout verification, Exim cached a "reject" for the entire domain. This
2657 is correct for most verifications, but it is not correct for a recipient
2658 verification with use_sender or use_postmaster set, because in that case
2659 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2660 case of MAIL FROM:<> rejection from other early rejections (e.g.
2661 rejection of HELO). When verifying a recipient using a non-null MAIL
2662 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2663 Whatever the result of the callout, the value of the domain cache is
2664 left unchanged (for any other kind of callout, getting as far as trying
2665 RCPT means that the domain itself is ok).
2667 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2670 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2671 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2672 session, the daemon ignores ECONNECT errors and logs others; it now
2673 ignores EPIPE as well.
2675 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2676 (quoted-printable decoding).
2678 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2679 later the small subsequent patch to fix an introduced bug.
2681 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2683 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2685 PH/08 An error is now given if message_size_limit is specified negative.
2687 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2688 to be given (somewhat) arbitrary names.
2690 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2693 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2694 miscellaneous code fixes
2696 PH/10 Added the log_reject_target ACL modifier to specify where to log
2699 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2700 hostname. This is wrong, because it relates to the incoming message (and
2701 probably the interface on which it is arriving) and not to the outgoing
2702 callout (which could be using a different interface). This has been
2703 changed to use the value of the helo_data option from the smtp transport
2704 instead - this is what is used when a message is actually being sent. If
2705 there is no remote transport (possible with a router that sets up host
2706 addresses), $smtp_active_hostname is used.
2708 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2709 tweaks were necessary in order to get it to work (see also 21 below):
2710 (a) The code assumed that strncpy() returns a negative number on buffer
2711 overflow, which isn't the case. Replaced with Exim's string_format()
2713 (b) There were several signed/unsigned issues. I just did the minimum
2714 hacking in of casts. There is scope for a larger refactoring.
2715 (c) The code used strcasecmp() which is not a standard C function.
2716 Replaced with Exim's strcmpic() function.
2717 (d) The code set only $1; it now sets $auth1 as well.
2718 (e) A simple test gave the error "authentication client didn't specify
2719 service in request". It would seem that Dovecot has changed its
2720 interface. Fortunately there's a specification; I followed it and
2721 changed what the client sends and it appears to be working now.
2723 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2726 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2727 address A is aliased to B and C, where B exists and C does not. Without
2728 -v the output is "A verified" because verification stops after a
2729 successful redirection if more than one address is generated. However,
2730 with -v the child addresses are also verified. Exim was outputting "A
2731 failed to verify" and then showing the successful verification for C,
2732 with its parentage. It now outputs "B failed to verify", showing B's
2733 parentage before showing the successful verification of C.
2735 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2736 look up a TXT record in a specific list after matching in a combined
2739 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2740 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2741 they consult the DNS. I had assumed they would set it the way they
2742 wanted; and indeed my experiments on Linux seem to show that in some
2743 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2744 To be on the safe side, however, I have now made the interface to
2745 host_find_byname() similar to host_find_bydns(), with an argument
2746 containing the DNS resolver options. The host_find_byname() function now
2747 sets these options at its start, just as host_find_bydns() does. The smtp
2748 transport options dns_qualify_single and dns_search_parents are passed to
2749 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2750 of host_find_byname() use the default settings of RES_DEFNAMES
2751 (qualify_single) but not RES_DNSRCH (search_parents).
2753 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2754 spool_read_header() do less string testing, by means of a preliminary
2755 switch on the second character of optional "-foo" lines. (This is
2756 overdue, caused by the large number of possibilities that now exist.
2757 Originally there were few.) While I was there, I also converted the
2758 str(n)cmp tests so they don't re-test the leading "-" and the first
2759 character, in the hope this might squeeze out yet more improvement.
2761 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2762 flag allowing group syntax was set by the header_syntax check but not
2763 turned off, possible causing trouble later; (2) The flag was not being
2764 set at all for the header_verify test, causing "group"-style headers to
2765 be rejected. I have now set it in this case, and also caused header_
2766 verify to ignore an empty address taken from a group. While doing this, I
2767 came across some other cases where the code for allowing group syntax
2768 while scanning a header line wasn't quite right (mostly, not resetting
2769 the flag correctly in the right place). These bugs could have caused
2770 trouble for malformed header lines. I hope it is now all correct.
2772 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2773 with the "reply" argument non-NULL. The code, however (which originally
2774 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2775 but it didn't always do it. This confused somebody who was copying the
2776 code for some other use. I have removed all the tests.
2778 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2779 feature that was used to support insecure browsers during the U.S. crypto
2780 embargo. It requires special client support, and Exim is probably the
2781 only MTA that supported it -- and would never use it because real RSA is
2782 always available. This code has been removed, because it had the bad
2783 effect of slowing Exim down by computing (never used) parameters for the
2784 RSA_EXPORT functionality.
2786 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2787 authenticator to fail if there's a tab character in the incoming data
2788 (there should never be unless someone is messing about, as it's supposed
2789 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2790 if the connection is using TLS or if the remote IP is the same as the
2791 local IP, and the "valid-client-cert option" if a client certificate has
2794 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2795 authenticators. This can be used for authorization after authentication
2796 succeeds. (In the case of plaintext, it servers for both authentication
2799 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2800 if any retry times were supplied.
2802 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2803 connection, where there is no client IP address to check. In this
2804 situation, the verify now always succeeds.
2806 PH/25 Applied John Jetmore's -Mset patch.
2808 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2810 PH/27 In a string expansion for a processed (not raw) header when multiple
2811 headers of the same name were present, leading whitespace was being
2812 removed from all of them, but trailing whitespace was being removed only
2813 from the last one. Now trailing whitespace is removed from each header
2814 before concatenation. Completely empty headers in a concatenation (as
2815 before) are ignored.
2817 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2818 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2820 PH/29 [Removed. This was a change that I later backed out, and forgot to
2821 correct the ChangeLog entry (that I had efficiently created) before
2822 committing the later change.]
2824 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2825 address errors (4xx response to RCPT) over the same connection as other
2826 messages routed to the same hosts. Such deliveries are always "forced",
2827 so retry times are not inspected. This resulted in far too many retries
2828 for the affected addresses. The effect occurred only when there were more
2829 hosts than the hosts_max_try setting in the smtp transport when it had
2830 the 4xx errors. Those hosts that it had tried were not added to the list
2831 of hosts for which the message was waiting, so if all were tried, there
2832 was no problem. Two fixes have been applied:
2834 (i) If there are any address or message errors in an SMTP delivery, none
2835 of the hosts (tried or untried) are now added to the list of hosts
2836 for which the message is waiting, so the message should not be a
2837 candidate for sending over the same connection that was used for a
2838 successful delivery of some other message. This seems entirely
2839 reasonable: after all the message is NOT "waiting for some host".
2840 This is so "obvious" that I'm not sure why it wasn't done
2841 previously. Hope I haven't missed anything, but it can't do any
2842 harm, as the worst effect is to miss an optimization.
2844 (ii) If, despite (i), such a delivery is accidentally attempted, the
2845 routing retry time is respected, so at least it doesn't keep
2846 hammering the server.
2848 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2849 in ${readsocket because some servers need this prod.
2851 PH/32 Added some extra debug output when updating a wait-xxx database.
2853 PH/33 The hint "could be header name not terminated by colon", which has been
2854 given for certain expansion errors for a long time, was not being given
2855 for the ${if def:h_colon_omitted{... case.
2857 PH/34 The spec says: "With one important exception, whenever a domain list is
2858 being scanned, $domain contains the subject domain." There was at least
2859 one case where this was not true.
2861 PH/35 The error "getsockname() failed: connection reset by peer" was being
2862 written to the panic log as well as the main log, but it isn't really
2863 panic-worthy as it just means the connection died rather early on. I have
2864 removed the panic log writing for the ECONNRESET error when getsockname()
2867 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2868 runs only) independently of the message's sender address. This meant
2869 that, if the 4xx error was in fact related to the sender, a different
2870 message to the same recipient with a different sender could confuse
2871 things. In particular, this can happen when sending to a greylisting
2872 server, but other circumstances could also provoke similar problems.
2873 I have changed the default so that the retry time for these errors is now
2874 based a combination of the sender and recipient addresses. This change
2875 can be overridden by setting address_retry_include_sender=false in the
2878 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2879 remote server are returned as part of bounce messages. This was not
2880 happening for LMTP over a pipe (the lmtp transport), but now it is the
2881 same for both kinds of LMTP.
2883 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2884 in header lines that were in fact CNAMEs. This is no longer the case.
2886 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2887 and queue runs started by the daemon processed all messages. This has
2888 been fixed so that -R and -S can now usefully be given with -q<time>.
2890 PH/40 Import PCRE release 6.7 (fixes some bugs).
2892 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2894 PH/42 Give an error if -q is specified more than once.
2896 PH/43 Renamed the variables $interface_address and $interface_port as
2897 $received_ip_address and $received_port, to make it clear that these
2898 values apply to message reception, and not to the outgoing interface when
2899 a message is delivered. (The old names remain recognized, of course.)
2901 PH/44 There was no timeout on the connect() call when using a Unix domain
2902 socket in the ${readsocket expansion. There now is.
2904 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2905 be meaningful with "accept".
2907 SC/01 Eximstats V1.43
2908 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2910 SC/02 Eximstats V1.44
2911 Use a glob alias rather than an array ref in the generated
2912 parser. This improves both readability and performance.
2914 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2915 Collect SpamAssassin and rejection statistics.
2916 Don't display local sender or destination tables unless
2917 there is data to show.
2918 Added average volumes into the top table text output.
2920 SC/04 Eximstats V1.46
2921 Collect data on the number of addresses (recipients)
2922 as well as the number of messages.
2924 SC/05 Eximstats V1.47
2925 Added 'Message too big' to the list of mail rejection
2926 reasons (thanks to Marco Gaiarin).
2928 SC/06 Eximstats V1.48
2929 Mainlog lines which have GMT offsets and are too short to
2930 have a flag are now skipped.
2932 SC/07 Eximstats V1.49 (Alain Williams)
2933 Added the -emptyok flag.
2935 SC/08 Eximstats V1.50
2936 Fixes for obtaining the IP address from reject messages.
2938 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2939 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2940 whitespace changes from 4.64-PH/27
2942 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2945 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2946 are found, allow negative numbers in numeric criteria)
2948 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2950 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2951 to match changes made in 4.64-PH/43
2953 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2955 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2956 "domain" is actually an IP address, and force a failure. This locks out
2957 those revolvers/nameservers that support "A-for-A" lookups, in
2958 contravention of the specifications.
2960 PH/48 When a host name was looked up from an IP address, and the subsequent
2961 forward lookup of the name timed out, the host name was left in
2962 $sender_host_name, contrary to the specification.
2964 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2965 restricted to single-key lookups, Exim was not diagnosing an error if
2966 * or *@ was used with a query-style lookup.
2968 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2970 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2971 long ago, but noone interested enough thought of it. Let's just say that
2972 the "1.1" means that there are some new functions that weren't there at
2973 some point in the past.
2975 PH/51 Error processing for expansion failure of helo_data from an smtp
2976 transport during callout processing was broken.
2978 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2979 tested/used via the -bh/-bhc/-bs options.
2981 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2982 bug, fixed in subsequent PCRE releases).
2984 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2985 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2987 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2993 SC/01 Use a glob alias rather than an array ref in eximstats generated
2994 parser. This improves both readability and performance.
2996 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2997 Don't display local sender or destination tables in eximstats unless
2998 there is data to show.
2999 Added average volumes into the eximstats top table text output.
3001 SC/03 Collect data on the number of addresses (recipients) as well
3002 as the number of messages in eximstats.
3004 TF/01 Correct an error in the documentation for the redirect router. Exim
3005 does (usually) call initgroups() when daemonizing.
3007 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3008 with consistent privilege compared to when running as a daemon.
3010 TF/03 Note in the spec that $authenticated_id is not set for local
3011 submissions from trusted users.
3013 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3014 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3016 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3017 by adding some example configuration directives to the default
3018 configuration file. A little bit of work is required to uncomment the
3019 directives and define how usernames and passwords are checked, but
3020 there is now a framework to start from.
3022 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3023 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3024 without this. I don't know how relevant this is to other LDAP libraries.
3026 PH/02 Add the verb name to the "unknown ACL verb" error.
3028 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3030 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3032 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3033 directory not expanded when it should be if an expanded home directory
3034 was set for the address (which is overridden by the transport).
3036 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3039 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3040 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3041 because it is too late at that time, and has no effect.
3043 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3044 security issue with \' (bugzilla #107). I could not use the
3045 PQescapeStringConn() function, because it needs a PGconn value as one of
3048 PH/08 When testing addresses using -bt, indicate those final addresses that
3049 are duplicates that would not cause an additional delivery. At least one
3050 person was confused, thinking that -bt output corresponded to deliveries.
3051 (Suppressing duplicates isn't a good idea as you lose the information
3052 about possibly different redirections that led to the duplicates.)
3054 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3055 systems where poll() doesn't work, in particular OS X.
3057 PH/10 Added more information to debugging output for retry time not reached.
3059 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3060 operations in malware.c.
3062 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3065 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3066 syslog_duplication set false, log lines that would normally be written
3067 both the the main log and to the reject log were not written to syslog at
3070 PH/14 In the default configuration, change the use of "message" in ACL warn
3071 statements to "add_header".
3073 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3074 not followed by a command (e.g. "seen endif").
3076 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3077 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3080 PH/17 Added extra conditions to the default value of delay_warning_condition
3084 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3085 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3086 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3089 The Auto-Submitted: and various List- headers are standardised, whereas I
3090 don't think Precedence: ever was.
3092 PH/18 Refactored debugging code in route_finduser() to show more information,
3093 in particular, the error code if getpwnam() issues one.
3095 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3096 This is apparently needed in addition to the PH/07 change above to avoid
3097 any possible encoding problems.
3099 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3100 but not after initializing Perl.
3102 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3103 output them only if debugging. By default they are written stderr,
3104 apparently, which is not desirable.
3106 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3109 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3112 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3114 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3115 authenticated or an ident call has been made. Suppress the default
3116 values for $authenticated_id and $authenticated_sender (but permit -oMai
3117 and -oMas) when testing with -bh.
3119 PH/24 Re-jigged the order of the tests in the default configuration so that the
3120 tests for valid domains and recipients precede the DNS black list and CSA
3121 tests, on the grounds that those ones are more expensive.
3123 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3124 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3125 This bug exists in every version of Exim that I still have, right back to
3128 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3129 However, an attempt to turn on case-sensitivity in a regex key by
3130 including (?-i) didn't work because the subject string was already
3131 lowercased, and the effects were non-intuitive. It turns out that a
3132 one-line patch can be used to allow (?-i) to work as expected.
3138 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3139 other effects) broke the use of negated acl sub-conditions.
3141 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3144 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3145 "Deny" causes Exim to reject the incoming connection with a 554 error.
3146 Unfortunately, if there is a major crisis, such as a disk failure,
3147 tcp-wrappers gives "deny", whereas what one would like would be some
3148 kind of temporary error. A kludge has been added to help with this.
3149 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3150 554 error is used if errno is still zero or contains ENOENT (which occurs
3151 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3154 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3156 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3157 errors. Otherwise a message that provokes a temporary error (when other
3158 messages do not) can cause a whole host to time out.
3160 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3161 addresses were routed directly to files or pipes from a redirect router.
3162 File deliveries just didn't batch; pipe deliveries might have suffered
3165 PH/06 A failure to get a lock for a hints database would erroneously always say
3166 "Failed to get write lock", even when it was really a read lock.
3168 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3169 of 0600. This has been changed to use the value of the lockfile_mode
3170 option (which defaults to 0600).
3172 PH/08 Applied small patch from the Sieve maintainer.
3174 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3175 folder from quota calculations, a direct delivery into this folder messed
3176 up the contents of the maildirsize file. This was because the regex was
3177 used only to exclude .Trash (or whatever) when the size of the mailbox
3178 was calculated. There was no check that a delivery was happening into an
3179 excluded directory. This bug has been fixed by ignoring all quota
3180 processing for deliveries into excluded directories.
3182 PH/10 Added the maildirfolder_create_regex option to appendfile.
3188 PH/01 The code for finding all the local interface addresses on a FreeBSD
3189 system running IPv6 was broken. This may well have applied to all BSD
3190 systems, as well as to others that have similar system calls. The broken
3191 code found IPv4 interfaces correctly, but gave incorrect values for the
3192 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3193 that it would not match correctly against @[] and not recognize the IPv6
3196 PH/02 The ipliteral router was not recognizing addresses of the form user@
3197 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3199 PH/03 Added disable_ipv6.
3201 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3202 decoded form, because it is most often used to construct To: headers
3203 lines in autoreplies, and the decoded form may well be syntactically
3204 invalid. However, $reply_address has leading white space removed, and all
3205 newlines turned into spaces so that the autoreply transport does not
3208 PH/05 If group was specified without a user on a router, and no group or user
3209 was specified on a transport, the group from the router was ignored.
3211 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3212 for visible compile-time settings that can be used to change these
3213 numbers, for those that want even more. Backwards compatibility with old
3214 spool files has been maintained. However, going back to a previous Exim
3215 release will lost any variables that are in spool files.
3217 PH/07 Two small changes when running in the test harness: increase delay when
3218 passing a TCP/IP connection to a new process, in case the original
3219 process has to generate a bounce, and remove special handling of
3220 127.0.0.2 (sic), which is no longer necessary.
3222 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3223 be the same on different OS.
3225 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3228 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3229 whether --show-vars was specified or not
3231 JJ/02 exipick: Added support for new ACL variable spool format introduced
3234 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3235 syntactically invalid From: or Reply-to: line, and a filter used this to
3236 generate an autoreply, and therefore failed to obtain an address for the
3237 autoreply, Exim could try to deliver to a non-existent relative file
3238 name, causing unrelated and misleading errors. What now happens is that
3239 it logs this as a hard delivery error, but does not attempt to create a
3242 PH/11 The exinext utility has a -C option for testing purposes, but although
3243 the given file was scanned by exinext itself; it wasn't being passed on
3244 when Exim was called.
3246 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3247 an end-of-file indication when reading a command response.
3249 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3250 compiled. In many other places in Exim, IPv6 addresses are always
3251 recognized, so I have changed this. It also means that IPv4 domain
3252 literals of the form [IPV4:n.n.n.n] are now always recognized.
3254 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3255 used if the router is not running as root, for example, when verifying at
3256 ACL time, or when using -bh. The debugging output from this situation was
3257 non-existent - all you got was a failure to exec. I have made two
3260 (a) Failures to set uid/gid, the current directory, or a process leader
3261 in a subprocess such as that created by queryprogram now generate
3262 suitable debugging output when -d is set.
3264 (b) The queryprogram router detects when it is not running as root,
3265 outputs suitable debugging information if -d is set, and then runs
3266 the subprocess without attempting to change uid/gid.
3268 PH/15 Minor change to Makefile for building test_host (undocumented testing
3271 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3272 additional section of a DNS packet that returns MX or SRV records.
3273 Instead, it always explicitly searches for A/AAAA records. This avoids
3274 major problems that occur when a DNS server includes only records of one
3275 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3276 fixed another bug: if SRV records were looked up and the corresponding
3277 address records were *not* found in the additional section, the port
3278 values from the SRV records were lost.
3280 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3281 using the correct key (the original address) when searching the retry
3282 rules in order to find which one to use for generating the retry hint.
3284 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3285 adding the default one. Similarly, if it contains a Reply-To: header, the
3286 errors_reply_to option, if set, is not used.
3288 PH/19 When calculating a retry time, Exim used to measure the "time since
3289 failure" by looking at the "first failed" field in the retry record. Now
3290 it does not use this if it is later than than the arrival time of the
3291 message. Instead it uses the arrival time. This makes for better
3292 behaviour in cases where some deliveries succeed, thus re-setting the
3293 "first failed" field. An example is a quota failure for a huge message
3294 when small messages continue to be delivered. Without this change, the
3295 "time since failure" will always be short, possible causing more frequent
3296 delivery attempts for the huge message than are intended.
3297 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3299 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3300 $1, $2, $3) because the numerical variables can be reset during some
3301 expansion items (e.g. "match"), thereby losing the authentication data.
3303 PH/21 Make -bV show the size of off_t variables so that the test suite can
3304 decide whether to run tests for quotas > 2G.
3306 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3307 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3308 filecount value is greater than 2G or if a quota value is greater than 2G
3309 on a system where the size of off_t is not greater than 4, a panic error
3312 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3313 never match. The debug and -bh output now contains an explicit error
3314 message indicating a malformed IPv4 address or mask.
3316 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3317 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3318 PH/23 above applies.
3320 PH/25 Do not write to syslog when running in the test harness. The only
3321 occasion when this arises is a failure to open the main or panic logs
3322 (for which there is an explicit test).
3324 PH/26 Added the /no_tell option to "control=freeze".
3326 PH/27 If a host name lookup failed very early in a connection, for example, if
3327 the IP address matched host_lookup and the reverse lookup yielded a name
3328 that did not have a forward lookup, an error message of the form "no IP
3329 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3330 could be logged. Now it outputs the IP address instead of "NULL".
3332 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3333 allows the sender and the authenticated sender to be set when
3334 submitting a message from within Exim. Since child_open_exim() is
3335 documented for local_scan(), the new function should be too.
3337 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3338 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3339 results in an empty string is now treated as unset.
3341 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3343 PH/31 Added sender_verify_fail logging option.
3345 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3346 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3347 tidied the source and removed it altogether.
3349 PH/33 When a queue run was abandoned because the load average was too high, a
3350 log line was always written; now it is written only if the queue_run log
3351 selector is set. In addition, the log line for abandonment now contains
3352 information about the queue run such as the pid. This is always present
3353 in "start" and "stop" lines but was omitted from the "abandon" line.
3355 PH/34 Omit spaces between a header name and the colon in the error message that
3356 is given when verify = headers_syntax fails (if there are lots of them,
3357 the message gets confusing).
3359 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3360 names, as there are now some PTR records that contain slashes. This check
3361 is only to protect against broken name servers that fall over on strange
3362 characters, so the fact that it applies to all lookups doesn't matter.
3364 PH/36 Now that the new test suite is complete, we can remove some of the
3365 special code in Exim that was needed for the old test suite. For example,
3366 sorting DNS records because real resolvers return them in an arbitrary
3367 order. The new test suite's fake resolver always returns records in the
3370 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3371 bounces) except when queue_only is set, to avoid logging races between
3372 the different processes.
3374 PH/38 Panic-die if .include specifies a non-absolute path.
3376 PH/39 A tweak to the "H" retry rule from its user.
3378 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3379 a label. They prevented compilation on older perls.
3381 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3382 a warning to be raised on newish perls.
3384 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3385 on queue. Changes to match documented behaviour of showing count of
3386 messages matching specified criteria.
3388 PH/40 Changed the default ident timeout from 30s to 5s.
3390 PH/41 Added support for the use of login_cap features, on those BSD systems
3391 that have them, for controlling the resources used by pipe deliveries.
3393 PH/42 The content-scanning code uses fopen() to create files in which to put
3394 message data. Previously it was not paying any attention to the mode of
3395 the files. Exim runs with umask(0) because the rest of the code creates
3396 files with open(), and sets the required mode explicitly. Thus, these
3397 files were ending up world-writeable. This was not a big issue, because,
3398 being within the spool directory, they were not world-accessible. I have
3399 created a function called modefopen, which takes an additional mode
3400 argument. It sets umask(777), creates the file, chmods it to the required
3401 mode, then resets the umask. All the relevant calls to fopen() in the
3402 content scanning code have been changed to use this function.
3404 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3405 to 24 hours. This avoids potential overflow problems when processing G
3406 and H retry rules. I suspect nobody ever tinkers with this value.
3408 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3410 PH/45 When the plaintext authenticator is running as a client, the server's
3411 challenges are checked to ensure they are valid base64 strings. By
3412 default, the authentication attempt is cancelled if an invalid string is
3413 received. Setting client_ignore_invalid_base64 true ignores these errors.
3414 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3415 they are received. Thus, the responses can be made to depend on the
3416 challenges. If an invalid string is ignored, an empty string is placed in
3419 PH/46 Messages that are created by the autoreply transport now contains a
3420 References: header, in accordance with RFCs 2822 and 3834.
3422 PH/47 Added authenticated_sender_force to the smtp transport.
3424 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3426 PH/49 Installed latest patch from the Sieve maintainer.
3428 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3429 was also set, the appendfile transport was unnecessarily scanning a
3430 directory of message files (e.g. for maildir delivery) to find the count
3431 of files (along with the size), even though it did not need this
3432 information. It now does the scan only if it needs to find either the
3433 size of the count of files.
3435 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3437 PH/52 Two bugs concerned with error handling when the smtp transport is
3440 (i) Exim was not creating retry information for temporary errors given
3441 for individual recipients after the DATA command when the smtp transport
3442 was used in LMTP mode. This meant that they could be retried too
3443 frequently, and not timed out correctly.
3445 (ii) Exim was setting the flag that allows error details to be returned
3446 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3447 recipients that were returned after the DATA command.
3449 PH/53 This is related to PH/52, but is more general: for any failing address,
3450 when detailed error information was permitted to be returned to the
3451 sender, but the error was temporary, then after the final timeout, only
3452 "retry timeout exceeded" was returned. Now it returns the full error as
3453 well as "retry timeout exceeded".
3455 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3456 do this, and (what is worse) MTAs that accept it.
3458 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3459 will now be deprecated.
3461 PH/56 New os.c-cygwin from the Cygwin maintainer.
3463 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3464 formats (previously only available in exim formats via -bpr, -bpru,
3465 and -bpra. Now also available in native and exiqgrep formats)
3467 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3468 with very large, slow to parse queues
3470 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3472 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3474 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3475 responses to authentication challenges, though it was showing the
3476 challenges; (ii) I've removed the CR characters from the debug output for
3479 PH/58 Allow for the insertion of a newline as well as a space when a string
3480 is turned into more than one encoded-word during RFC 2047 encoding. The
3481 Sieve code now uses this.
3483 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3484 data_4xx, lost_connection, tls_required.
3486 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3487 message was being sent as an SMTP response.
3489 PH/61 Add -l and -k options to exicyclog.
3491 PH/62 When verifying, if an address was redirected to one new address, so that
3492 verification continued, and the new address failed or deferred after
3493 having set something in $address_data, the value of $address_data was not
3494 passed back to the ACL. This was different to the case when no
3495 redirection occurred. The value is now passed back in both cases.
3497 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3498 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3499 use login_cap.h, so on its own it isn't the distinguishing feature. The
3500 new name refers directly to the setclassresources() function.
3502 PH/65 Added configuration files for NetBSD3.
3504 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3506 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3507 is preferred over IPv4.
3509 PH/68 The bounce_return_message and bounce_return_body options were not being
3510 honoured for bounces generated during the reception of non-SMTP messages.
3511 In particular, this applied to messages rejected by the ACL. This bug has
3512 been fixed. However, if bounce_return_message is true and bounce_return_
3513 body is false, the headers that are returned for a non-SMTP message
3514 include only those that have been read before the error was detected.
3515 (In the case of an ACL rejection, they have all been read.)
3517 PH/69 The HTML version of the specification is now built in a directory called
3518 spec_html instead of spec.html, because the latter looks like a path with
3519 a MIME-type, and this confuses some software.
3521 PH/70 Catch two compiler warnings in sieve.c.
3523 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3524 function verify_get_ident() calls ip_connect() to connect a socket, but
3525 if the "connect()" function timed out, ip_connect() used to close the
3526 socket. However, verify_get_ident() also closes the socket later, and in
3527 between Exim writes to the log, which may get opened at this point. When
3528 the socket was closed in ip_connect(), the log could get the same file
3529 descriptor number as the socket. This naturally causes chaos. The fix is
3530 not to close the socket in ip_connect(); the socket should be closed by
3531 the function that creates it. There was only one place in the code where
3532 this was missing, in the iplookup router, which I don't think anybody now
3533 uses, but I've fixed it anyway.
3535 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3536 well as to direct DNS lookups. Otherwise the handling of names in host
3537 lists is inconsistent and therefore confusing.
3543 PH/01 Two changes to the default runtime configuration:
3545 (1) Move the checks for relay_from_hosts and authenticated clients from
3546 after to before the (commented out) DNS black list checks.
3548 (2) Add control=submission to the relay_from_hosts and authenticated
3549 clients checks, on the grounds that messages accepted by these
3550 statements are most likely to be submissions.
3552 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3554 (1) Generate an error if the third argument for the ${prvs expansion is
3557 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3560 (3) Reset the variables that are obtained from the first argument of
3561 ${prvscheck and used in the second argument before leaving the code,
3562 because their memory is reclaimed, so using them afterwards may do
3565 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3566 one (it's much easier than Tom thought :-).
3568 (5) Because of (4), we can now allow for the use of $prvscheck_result
3569 inside the third argument.
3571 PH/03 For some reason, the default setting of PATH when running a command from
3572 a pipe transport was just "/usr/bin". I have changed it to
3575 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3576 anything to be listed in the output from -bV.
3578 PH/05 When a filter generated an autoreply, the entire To: header line was
3579 quoted in the delivery log line, like this:
3581 => >A.N.Other <ano@some.domain> <original@ddress> ...
3583 This has been changed so that it extracts the operative address. There
3584 may be more than one such address. If so, they are comma-separated, like
3587 => >ano@some.domain,ona@other.domain <original@ddress> ...
3589 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3590 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3591 not being looked up in the rDNS to get a host name, Exim was showing the
3592 IP address twice in Received: lines, even though the IP addresses were
3593 identical. For example:
3595 Received: from [1.2.3.4] (helo=[1.2.3.4])
3597 However, if the real host name was known, it was omitting the HELO data
3598 if it matched the actual IP address. This has been tidied up so that it
3599 doesn't show the same IP address twice.
3601 PH/07 When both +timestamp and +memory debugging was on, the value given by
3602 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3603 called by the debug printing, thereby overwriting the timestamp buffer.
3604 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3606 PH/08 When the original message was included in an autoreply transport, it
3607 always said "this is a copy of the message, including all the headers",
3608 even if body_only or headers_only was set. It now gives an appropriate
3611 PH/09 Applied a patch from the Sieve maintainer which:
3613 o fixes some comments
3614 o adds the (disabled) notify extension core
3615 o adds some debug output for the result of if/elsif tests
3616 o points to the current vacation draft in the documentation
3617 and documents the missing references header update
3621 o fixes a bug in processing the envelope test (when testing
3622 multiple envelope elements, the last element determined the
3625 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3626 Electronic Mail") by including:
3628 Auto-submitted: auto-generated
3630 in the messages that it generates (bounce messages and others, such as
3631 warnings). In the case of bounce messages for non-SMTP messages, there was
3632 also a typo: it was using "Auto_submitted" (underscore instead of
3633 hyphen). Since every message generated by Exim is necessarily in response
3634 to another message, thes have all been changed to:
3636 Auto-Submitted: auto-replied
3638 in accordance with these statements in the RFC:
3640 The auto-replied keyword:
3642 - SHOULD be used on messages sent in direct response to another
3643 message by an automatic process,
3645 - MUST NOT be used on manually-generated messages,
3647 - MAY be used on Delivery Status Notifications (DSNs) and Message
3648 Disposition Notifications (MDNs),
3650 - MUST NOT be used on messages generated by automatic or periodic
3651 processes, except for messages which are automatic responses to
3654 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3655 to the default Received: header definition.
3657 PH/12 Added log selector acl_warn_skipped (default on).
3659 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3660 variables because (a) they are in the wrong storage pool and (b) even if
3661 they were copied, it wouldn't work properly because of the caching.
3663 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3664 checking when decoding. Apparently there are clients that generate
3665 overlong encoded strings. Why am I not surprised?
3667 PH/15 If the first argument of "${if match_address" was not empty, but did not
3668 contain an "@" character, Exim crashed. Now it writes a panic log message
3669 and treats the condition as false.
3671 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3673 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3674 extension "envelope-auth". The code is finished and in agreement with
3675 other implementations, but there is no documentation so far and in fact,
3676 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3677 not changing the active code.
3679 Print executed "if" and "elsif" statements when debugging is used. This
3680 helps a great deal to understand what a filter does.
3682 Document more things not specified clearly in RFC3028. I had all this
3683 sorted out, when out of a sudden new issues came to my mind. Oops."
3685 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3688 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3689 canonical form (as documented). However, after a host name lookup from
3690 the IP address, check_host() was doing a simple string comparison with
3691 addresses acquired from the DNS when checking that the found name did
3692 have the original IP as one of its addresses. Since any found IPv6
3693 addresses are likely to be in abbreviated form, the comparison could
3694 fail. Luckily, there already exists a function for doing the comparison
3695 by converting both addresses to binary, so now that is used instead of
3696 the text comparison.
3698 PH/20 There was another similar case to PH/19, when a complete host name was
3699 given in a host list; looking up its IP address could give an abbreviated
3700 form, whereas the current host's name might or might not be abbreviated.
3701 The same fix has been applied.
3707 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3708 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3711 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3712 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3714 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3716 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3717 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3718 Also, for "Auto-Submitted", treat anything other than "no" as
3719 non-personal, in accordance with RFC 3834. (Previously it treated
3720 anything starting "auto-" as non-personal.)
3722 TF/01 The control=submission/name=... option had a problem with syntax
3723 errors if the name included a slash character. The /name= option
3724 now slurps the rest of the string, so it can include any characters
3725 but it must come last in the list of options (after /sender_retain
3728 PH/05 Some modifications to the interface to the fake nameserver for the new
3736 TK/01 Added the "success_on_redirect" address verification option. See
3737 NewStuff for rationale and an example.
3739 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3741 PH/02 Patch to exigrep to allow it to work on syslog lines.
3743 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3744 fread() to skip over the body file's header line, because in Cygwin the
3745 header line is locked and is inaccessible.
3747 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3748 co-exist for some time) to make it clear that it is the Exim ID that is
3749 referenced, not the Message-ID: header line.
3751 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3752 string_format() function, because snprintf() does not exist on all
3755 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3756 ${stat: expansion item.
3758 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3759 protocol synchronization error", to keep the pedants happy.
3761 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3762 well as for IRIX systems, when gcc is being used. See the host.c source
3765 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3767 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3770 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3771 transport and to the smtp transport in LMTP mode.
3773 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3775 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3776 run a filter in a subprocess. This could lead to confusion in subsequent
3777 lookups in the parent process. There should also be a search_tidyup() at
3778 the end of the subprocess.
3780 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3781 only if the host matched helo_try_verify_hosts, which caused the
3782 verification to occur when the EHLO/HELO command was issued. The ACL just
3783 tested the remembered result. Now, if a previous verification attempt has
3784 not happened, "verify = helo" does it there and then.
3786 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3788 TK/03 Fix log output including CR from clamd.
3790 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3791 exist provoked a memory error which could cause a segfault.
3793 PH/15 Installed PCRE 6.2
3795 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3797 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3798 of the problem. Specifically, suggested +O2 rather than +O1 for the
3801 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3803 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3804 record, the port was not correctly logged when the outgoing_port log
3805 selector was set (it logged the transort's default port).
3807 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3808 fallback_hosts, and "hosts" in the smtp transport.
3810 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3811 host errors such as "Connection refused".
3813 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3814 authentication with radiusclient 0.4.9:
3816 - Error returned from rc_read_config was caught wrongly
3817 - Username/password not passed on to radius server due to wrong length.
3819 The presumption is that some radiusclient API changes for 4.51/PH/17
3820 were not taken care of correctly. The code is still untested by me (my
3821 Linux distribution still has 0.3.2 of radiusclient), but it was
3822 contributed by a Radius user.
3824 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3825 expanding the "port" option of the smtp transport.
3827 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3828 while reading a MIME header. Thanks to Tom Hughes for a patch.
3830 PH/24 Include config.h inside local_scan.h so that configuration settings are
3833 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3834 that in an ACL for RCPT (for example), you can examine exactly what was
3837 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3838 commands, but it was not correctly comparing the address with the actual
3839 client host address. Thus, it would show the EHLO address in Received:
3840 header lines when this was not necessary.
3842 PH/27 Added the % operator to ${eval:}.
3844 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3845 it should be ignoring failures (because with -C, for example, it has lost
3846 privilege). It wasn't ignoring creation failures other than "already
3849 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3852 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3853 because an input line was too long, either on its own, or by virtue of
3854 too many continuations, the temporary file was not being removed, and the
3855 return code was incorrect.
3857 PH/31 Missing "BOOL" in function definition in filtertest.c.
3859 PH/32 Applied Sieve patches from the maintainer.
3861 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3863 PH/33 Added "verify = not_blind".
3865 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3866 Local/Makefile (with some defaults set). These are used in built scripts
3867 such as exicyclog, but they have never been used in the exim_install
3868 script (though there are many overriding facilities there). I have
3869 arranged that the exim_install script now takes note of these two
3872 PH/35 Installed configuration files for Dragonfly.
3874 PH/36 When a locally submitted message by a trusted user did not contain a
3875 From: header, and the sender address was obtained from -f or from an SMTP
3876 MAIL command, and the trusted user did not use -F to supply a sender
3877 name, $originator_name was incorrectly used when constructing a From:
3878 header. Furthermore, $originator_name was used for submission mode
3879 messages from external hosts without From: headers in a similar way,
3880 which is clearly wrong.
3882 PH/37 Added control=suppress_local_fixups.
3884 PH/38 When log_selector = +received_sender was set, and the addition of the
3885 sender made the log line's construction buffer exactly full, or one byte
3886 less than full, an overflow happened when the terminating "\n" was
3889 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3890 when the result of a list match is failure because a DNS lookup failed.
3892 PH/40 RM_COMMAND is now used in the building process.
3894 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3895 the "build-* directories that it finds.
3897 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3898 address in a domain literal was a prefix of an interface address.
3900 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3901 when verifying a sender address, unless rewrite_headers is false.
3903 PH/44 Wrote a long comment about why errors_to addresses are verified as
3904 recipients, not senders.
3906 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3907 the ratelimit ACL was added.
3909 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3911 PH/46 Added extra information about PostgreSQL errors to the error string.
3913 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3914 avoiding the need to install special zones in a real server. This is
3915 backwards compatible; if it can't find the fake resolver, it drops back.
3916 Thus, both old and new test suites can be run.
3918 TF/02 Added util/ratelimit.pl
3920 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3921 clock is set back in time.
3923 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3924 Candler <B.Candler@pobox.com>.
3926 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3927 OK for addresses that are the result of redirections.
3929 PH/48 A number of further additions for the benefit of the new test suite,
3930 including a fake gethostbyname() that interfaces to the fake DNS resolver
3933 TF/06 The fix for widen_domains has also been applied to qualify_single and
3934 search_parents which are the other dnslookup options that can cause
3937 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3940 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3942 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3943 service" if it encountered an MX record with an empty target hostname.
3944 The message is now "an MX or SRV record indicated no SMTP service".
3946 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3947 if the DNS of the sending site is misconfigured. This is quite a
3948 common situation. This change restores the behaviour of treating a
3949 helo verification defer as a failure.
3951 PH/51 If self=fail was set on a router, the bounce message did not include the
3952 actual error message.
3958 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3960 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3961 command itself ended in error, the underlying message about the transport
3962 filter timeout was being overwritten with the pipe command error. Now the
3963 underlying error message should be appended to the second error message.
3965 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3967 PH/02 Reduce the amount of output that "make" produces by default. Full output
3968 can still be requested.
3970 PH/03 The warning log line about a condition test deferring for a "warn" verb
3971 was being output only once per connection, rather than after each
3972 occurrence (because it was using the same function as for successful
3973 "warn" verbs). This seems wrong, so I have changed it.
3975 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3976 it should not have, which might have caused a crash in the right
3977 circumstances, but probably never did.
3979 PH/04 Installed a modified version of Tony Finch's patch to make submission
3980 mode fix the return path as well as the Sender: header line, and to
3981 add a /name= option so that you can make the user's friendly name appear
3984 TF/03 Added the control = fakedefer ACL modifier.
3986 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3987 Mark Lowes for thorough testing.
3989 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3991 TK/03 Merged latest SRS patch from Miles Wilton.
3993 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3994 with the definition in sysexits.h (which is #included earlier).
3995 Fortunately, Exim does not actually use EX_OK. The code used to try to
3996 preserve the sysexits.h value, by assuming that macro definitions were
3997 scanned for macro replacements. I have been disabused of this notion,
3998 so now the code just undefines EX_OK before #including unistd.h.
4000 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4001 in the smtp transport. When a block could not be written in a single
4002 write() function, the timeout was being re-applied to each part-write.
4003 This seems wrong - if the receiver was accepting one byte at a time it
4004 would take for ever. The timeout is now adjusted when this happens. It
4005 doesn't have to be particularly precise.
4007 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4008 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4010 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4011 without a domain if the check to <postmaster@domain> fails.
4013 SC/01 Eximstats: added -xls and the ability to specify output files
4014 (patch written by Frank Heydlauf).
4016 SC/02 Eximstats: use FileHandles for outputting results.
4018 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4020 SC/04 Eximstats: fixed display of large numbers with -nvr option
4022 SC/05 Eximstats: fixed merging of reports with empty tables.
4024 SC/06 Eximstats: added the -include_original_destination flag
4026 SC/07 Eximstats: removed tabs and trailing whitespace.
4028 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4030 TK/06 MBOX spool code: Add real "From " MBOX separator line
4031 so the .eml file is really in mbox format (even though
4032 most programs do not really care). Patch from Alex Miller.
4034 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4035 The latter is generated from $received_to and is only set if the
4036 message has one envelope recipient. SA can use these headers,
4037 obviously out-of-the-box. Patch from Alex Miller.
4039 PH/08 The ${def test on a variable was returning false if the variable's
4040 value was "0", contrary to what the specification has always said!
4041 The result should be true unless the variable is empty.
4043 PH/09 The syntax error of a character other than { following "${if
4044 def:variable_name" (after optional whitespace) was not being diagnosed.
4045 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4046 accidental colon was present, for example, could give incorrect results.
4048 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4049 result is used (not including appendfile, where other changes are about
4052 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4053 This involved changing a lot of size variables from int to off_t. It
4054 should work with maildirs and everything.
4056 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4057 spamd dying while we are connected to it.
4059 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4062 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4063 function for BDB 4.3.
4065 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4067 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4068 the sources for PCRE have been re-arranged and more files are now
4071 PH/15 The code I had for printing potentially long long variables in PH/11
4072 above was not the best (it lost precision). The length of off_t variables
4073 is now inspected at build time, and an appropriate printing format (%ld
4074 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4075 to be "long long int" or "long int". This is needed for the internal
4076 formatting function string_vformat().
4078 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4079 the configuration file to be ":syslog", then the script "guesses" where
4080 the logs files are, rather than using the compiled in default. In our
4081 case the guess is not the same as the compiled default, so the script
4082 suddenly stopped working when I started to use syslog. The patch checks
4083 to see if log_file_path is "". If so, it attempts to read it from exim
4084 with no configuration file to get the compiled in version, before it
4085 falls back to the previous guessing code."
4087 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4088 implementing BATV in an Exim configuration. See NewStuff for the gory
4091 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4092 Makefile that are specific to HP-UX.
4094 PH/18 If the "use_postmaster" option was set for a recipient callout together
4095 with the "random" option, the postmaster address was used as the MAIL
4096 FROM address for the random test, but not for the subsequent recipient
4097 test. It is now used for both.
4099 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4100 patch removes a few documentation additions to RFC 3028, because the
4101 latest draft now contains them. It adds the new en;ascii-case comparator
4102 and a new error check for 8bit text in MIME parts. Comparator and
4103 require names are now matched exactly. I enabled the subaddress
4104 extension, but it is not well tested yet (read: it works for me)."
4106 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4107 rework some of the code of TK/09 above to avoid the hardwired use of
4108 "%lld" and "long long". Replaced the call to snprintf() with a call to
4111 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4112 records point to non-existent hosts", "retry timeout exceeded", and
4113 "retry time not reached for any host after a long failure period".
4115 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4116 experimental DomainKeys support:
4118 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4119 (2) On an error such as an illegally used "control", the wrong name for
4120 the control was given.
4122 These problems did NOT occur unless DomainKeys support was compiled.
4124 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4126 PH/24 Added ${if match_ip condition.
4128 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4129 databases so that it will be absolutely obvious if a crash occurs in the
4130 DB library. This is a regular occurrence (often caused by mis-matched
4133 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4134 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4135 picked up on a user's system that detects such things. There doesn't seem
4136 to be a gcc warning option for this - only an attribute that has to be
4137 put on the function's prototype. It seems that in Fedora Core 4 they have
4138 set this on a number of new functions. No doubt there will be more in due
4141 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4142 specify a transport. However, if an address that was verified by such a
4143 router was the subject of a callout, Exim crashed because it tried to
4144 read the rcpt_include_affixes from the non-existent transport. Now it
4145 just assumes that the setting of that option is false. This bug was
4146 introduced by 4.51/PH/31.
4148 PH/28 Changed -d+all to exclude +memory, because that information is very
4149 rarely of interest, but it makes the output a lot bigger. People tend to
4150 do -d+all out of habit.
4152 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4153 code in os-type was giving problems when libc.so lives in lib64, like on
4156 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4157 aren't the modern standard, and it seems that some systems' include files
4158 don't always have them. Exim was already checking for some of the newer
4159 ones like T_AAAA, and defining it itself. I've added checks for all the
4160 record types that Exim uses.
4162 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4163 not automatically generating a new one, as it is supposed to. This
4164 prevented TLS from working. If the file did exist, but contained invalid
4165 data, a new version was generated, as expected. It was only the case of a
4166 non-existent file that was broken.
4168 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4169 with a change in libdomainkeys > 0.64.
4171 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4172 from DNS. If the selector record carries the flag, it now has
4173 precedence over the domain-wide flag.
4175 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4177 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4178 the use of a port name that isn't defined in /etc/services) occurred, the
4179 message was deferred as in a normal delivery, and thus remained on the
4180 spool, instead of being failed because of the mua_wrapper setting. This
4181 is now fixed, and I tidied up some of the mua_wrapper messages at the
4184 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4185 the messages in a hash of arrays rather than using individual hashes.
4186 This is a bit cleaner and results in dramatic memory savings, albeit
4187 at a slight CPU cost.
4189 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4190 as requested by Marc Sherman.
4192 SC/10 Eximstats: added histograms for user specified patterns as requested
4195 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4197 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4198 fopen() in the content-scanning modules that did not already have it.
4204 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4205 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4207 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4209 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4211 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4212 to test Sieve filters that use "vacation".
4214 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4215 that changes the way the GnuTLS parameters are stored in the cache file.
4216 The new format can be generated externally. For backward compatibility,
4217 if the data in the cache doesn't make sense, Exim assumes it has read an
4218 old-format file, and it generates new data and writes a new file. This
4219 means that you can't go back to an older release without removing the
4222 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4223 work if there are any delivery delays because "one_time" forces the
4224 parent to be marked "delivered", so its unseen clone is never tried
4225 again. For this reason, Exim now forbids the simultaneous setting of
4228 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4229 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4230 entry if you want to know the details. The fix, however, neglected to
4231 consider the case when local delivery batching is involved. The test for
4232 "previously delivered" was not happening when checking to see if an
4233 address could be batched with a previous (undelivered) one; under
4234 certain circumstances this could lead to multiple deliveries to the same
4237 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4238 in its include files, and this causes problems building Exim.
4240 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4241 header_syntax) but Exim was just ignoring anything given after a slash.
4242 In particular, this caused confusion with an attempt to use "verify =
4243 reverse_host_lookup/defer_ok". An error is now given when options are
4244 supplied for verify items that do not have them. (Maybe reverse_host_
4245 lookup should have a defer_ok option, but that's a different point.)
4247 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4248 defined by RFC 821) to 2048, because there were problems with some AUTH
4249 commands, and RFC 1869 says the size should be increased for extended
4250 SMTP commands that take arguments.
4252 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4255 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4256 "unknown" error; now it says that the functionality isn't in the binary.
4258 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4259 an address' error message when a string expansion fails (syntax or
4260 whatever). Otherwise the password may appear in the log. Following change
4261 PH/42 below, there is no longer a chance of it appearing in a bounce
4264 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4266 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4267 of its addresses was ever tried. (Bugzilla bug #2.)
4269 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4270 the result incorrectly in the debug output. (It correctly added a newline
4271 to what was transported.)
4273 TF/01 Added $received_time.
4275 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4276 commented out examples of how to interface to a virus scanner and to
4277 SpamAssassin. Also added commented examples of av_scanner and
4278 spamd_address settings.
4280 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4281 and controls are allowed in which ACLs. There were a couple of minor
4282 errors. Some of the entries in the conditions table (which is a table of
4283 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4284 negation of where the condition IS allowed.
4286 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4288 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4289 header file does not have a version number, so I've had to invent a new
4290 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4291 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4292 radiusclient), but it was contributed by a Radius user.
4294 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4295 files or directories, for OpenSSL.
4297 PH/19 When an Exim process that is running as root has to create an Exim log
4298 file, it does so in a subprocess that runs as exim:exim so as to get the
4299 ownership right at creation (otherwise, other Exim processes might see
4300 the file with the wrong ownership). There was no test for failure of this
4301 fork() call, which would lead to the process getting stuck as it waited
4302 for a non-existent subprocess. Forks do occasionally fail when resources
4303 run out. I reviewed all the other calls to fork(); they all seem to check
4306 PH/20 When checking for unexpected SMTP input at connect time (before writing
4307 the banner), Exim was not dealing correctly with a non-positive return
4308 from the read() function. If the client had disconnected by this time,
4309 the result was a log entry for a synchronization error with an empty
4310 string after "input=" when read() returned zero. If read() returned -1
4311 (an event I could not check), uninitialized data bytes were printed.
4312 There were reports of junk text (parts of files, etc) appearing after
4315 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4317 PH/22 Added support for macro redefinition, and (re)definition in between
4318 driver and ACL definitions.
4320 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4321 forgetting to use the resulting value; it was using the unexpanded value.
4323 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4324 hadn't been configured. The fix is from Juergen Kreileder, who
4325 understands it better than I do:
4327 "Here's what I see happening with three configured cyrus_sasl
4328 authenticators configured (plain, login, cram-md5):
4330 On startup auth_cyrus_sasl_init() gets called for each of these.
4331 This means three calls to sasl_listmech() without a specified mech_list.
4332 => SASL tests which mechs of all available mechs actually work
4333 => three warnings about OTP not working
4334 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4336 With the patch, sasl_listmech() also gets called three times. But now
4337 SASL's mech_list option is set to the server_mech specified in the the
4338 authenticator. Or in other words, the answer from sasl_listmech()
4339 gets limited to just the mech you're testing for (which is different
4341 => the return list contains just 'plain' or 'login', 'cram-md5' or
4342 nothing depending on the value of ob->server_mech.
4344 I've just tested the patch: Authentication still works fine,
4345 unavailable mechs specified in the exim configuration are still
4346 caught, and the auth.log warnings about OTP are gone."
4348 PH/25 When debugging is enabled, the contents of the command line are added
4349 to the debugging output, even when log_selector=+arguments is not
4352 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4353 answer is "GNU", and only if the return is "GNU/something" is the answer
4356 PH/27 $acl_verify_message is now set immediately after the failure of a
4357 verification in an ACL, and so is available in subsequent modifiers. In
4358 particular, the message can be preserved by coding like this:
4360 warn !verify = sender
4361 set acl_m0 = $acl_verify_message
4363 Previously, $acl_verify_message was set only while expanding "message"
4364 and "log_message" when a very denied access.
4366 PH/28 Modified OS/os.c-Linux with
4368 -#ifndef OS_LOAD_AVERAGE
4369 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4371 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4372 nomenclature these days.)
4374 PH/29 Installed patch from the Sieve maintainer that adds the options
4375 sieve_useraddress and sieve_subaddress to the redirect router.
4377 PH/30 In these circumstances:
4378 . Two addresses routed to the same list of hosts;
4379 . First host does not offer TLS;
4380 . First host accepts first address;
4381 . First host gives temporary error to second address;
4382 . Second host offers TLS and a TLS session is established;
4383 . Second host accepts second address.
4384 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4385 and peerdn, if requested) that were in fact used only for the second
4388 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4389 attention to any local part prefix or suffix that was matched by the
4390 router that accepted the address. It now behaves in the same way as it
4391 does for delivery: the affixes are removed from the local part unless
4392 rcpt_include_affixes is set on the transport.
4394 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4395 timeout during the DATA phase of an incoming message.
4397 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4398 applied a patch sanctioned by the Sieve maintainer.
4400 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4401 the uid or gid is negative. A case of a negative gid caused this to be
4402 noticed. The fix allows for either to be negative.
4404 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4405 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4408 PH/36 The change PH/12 above was broken. Fixed it.
4410 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4411 the grounds that routing the same address twice would always produce the
4412 same answer. This might have been true once, but it is certainly no
4413 longer true now. Routing a child address may depend on the previous
4414 routing that produced that child. Some complicated redirection strategies
4415 went wrong when messages had multiple recipients, and made Exim's
4416 behaviour dependent on the order in which the addresses were given.
4418 I have moved the duplicate checking until after the routing is complete.
4419 Exim scans the addresses that are assigned to local and remote
4420 transports, and removes any duplicates. This means that more work will be
4421 done, as duplicates will always all be routed, but duplicates are
4422 presumably rare, so I don't expect this is of any significance.
4424 For deliveries to pipes, files, and autoreplies, the duplicate checking
4425 still happens during the routing process, since they are not going to be
4428 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4429 It corrects a timeout issue with spamd. This is Ian's comment: "The
4430 background is that sometimes spamd either never reads data from a
4431 connection it has accepted, or it never writes response data. The exiscan
4432 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4433 blindly assumes that writes won't block so it may never time out."
4435 PH/39 Allow G after quota size as well as K and M.
4437 PH/40 The value set for $authenticated_id in an authenticator may not contain
4438 binary zeroes or newlines because the value is written to log lines and
4439 to spool files. There was no check on this. Now the value is run through
4440 the string_printing() function so that such characters are converted to
4441 printable escape sequences.
4443 PH/41 $message_linecount is a new variable that contains the total number of
4444 lines in the message. Compare $body_linecount, which is the count for the
4447 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4448 bounce and delay warning messages, except in certain special cases, which
4451 (a) An SMTP error message from a remote host;
4452 (b) A message specified in a :fail: redirection;
4453 (c) A message specified in a "fail" command in a system filter;
4454 (d) A message specified in a FAIL return from the queryprogram router;
4455 (e) A message specified by the cannot_route_message router option.
4457 In these cases only, Exim does include the error details in bounce and
4458 warning messages. There are also a few cases where bland messages such
4459 as "unrouteable address" or "local delivery error" are given.
4461 PH/43 $value is now also set for the "else" part of a ${run expansion.
4463 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4464 being worked on, but at least Exim now implements the latest version to
4467 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4468 process to complete was treated as a delivery failure, a timeout while
4469 writing the message to the pipe was logged, but erroneously treated as a
4470 successful delivery. Such timeouts include transport filter timeouts. For
4471 consistency with the overall process timeout, these timeouts are now
4472 treated as errors, giving rise to delivery failures by default. However,
4473 there is now a new Boolean option for the pipe transport called
4474 timeout_defer, which, if set TRUE, converts the failures into defers for
4475 both kinds of timeout. A transport filter timeout is now identified in
4478 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4479 systems where "make" and "gmake" are different, calling "gmake" at top
4480 level broke things. I've arranged for the value of $(MAKE) to be passed
4481 from the Makefile to this script so that it can call the same version of
4485 A note about Exim versions 4.44 and 4.50
4486 ----------------------------------------
4488 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4489 changes of various kinds. As a consequence, a big documentation update was
4490 needed. This delayed the release for rather longer than seemed good, especially
4491 in the light of a couple of (minor) security issues. Therefore, the changes
4492 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4493 release. So 4.44 and 4.50 are in effect two different branches that both start
4496 I have left the 4.50 change log unchanged; it contains all the changes since
4497 4.43. The change log for 4.44 is below; many of its items are identical to
4498 those for 4.50. This seems to be the most sensible way to preserve the
4499 historical information.
4505 1. Minor wording change to the doc/README.SIEVE file.
4507 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4508 computation of the current number of files was incorrect.
4510 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4511 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4514 4. Give more explanation in the error message when the command for a transport
4515 filter fails to execute.
4517 5. There are several places where Exim runs a non-Exim command in a
4518 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4519 was being done only for the command run by the queryprogram router. It is
4520 now done for all such subprocesses. The other cases are: ${run, transport
4521 filters, and the commands run by the lmtp and pipe transports.
4523 6. Added CONFIGURE_GROUP build-time option.
4525 7. Some older OS have a limit of 256 on the maximum number of file
4526 descriptors. Exim was using setrlimit() to set 1000 as a large value
4527 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4528 systems. I've change it so that if it can't get 1000, it tries for 256.
4530 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4531 was an oversight, and furthermore, ever since the addition of extra
4532 controls (e.g. 4.43/32), the checks on when to allow different forms of
4533 "control" were broken. There should now be diagnostics for all cases when a
4534 control that does not make sense is encountered.
4536 9. Added the /retain_sender option to "control=submission".
4538 10. $recipients is now available in the predata ACL (oversight).
4540 11. Tidy the search cache before the fork to do a delivery from a message
4541 received from the command line. Otherwise the child will trigger a lookup
4542 failure and thereby defer the delivery if it tries to use (for example) a
4543 cached ldap connection that the parent has called unbind on.
4545 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4546 of $address_data from the recipient verification was clobbered by the
4547 sender verification.
4549 13. The value of address_data from a sender verification is now available in
4550 $sender_address_data in subsequent conditions in the ACL statement.
4552 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4554 15. Added a new option "connect=<time>" to callout options, to set a different
4557 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4558 was its contents. (It was OK if the option was not defined at all.)
4560 17. A "Completed" log line is now written for messages that are removed from
4561 the spool by the -Mrm option.
4563 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4564 information about exactly what failed.
4566 19. Added -dd to debug only the daemon process.
4568 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4569 handles timeouts, both on the server side and network timeouts. Renamed the
4570 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4572 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4573 It is now set to "smtps".
4575 22. $host_address is now set to the target address during the checking of
4576 ignore_target_hosts.
4578 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4579 being passed; this would have caused $sender_host_name to have been used if
4580 matching the list had actually called for a host name (not very likely,
4581 since this list is usually IP addresses). A host name is now passed as
4584 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4585 code that specifies a non-restarting handler (typically sigaction() in
4586 modern systems) in an attempt to fix a rare and obscure crash bug.
4588 25. Narrowed the window for a race in the daemon that could cause it to ignore
4589 SIGCHLD signals. This is not a major problem, because they are used only to
4590 wake it up if nothing else does.
4592 26. A malformed maildirsize file could cause Exim to calculate negative values
4593 for the mailbox size or file count. Odd effects could occur as a result.
4594 The maildirsize information is now recalculated if the size or filecount
4597 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4598 support for a long time. Removed HAVE_SYS_VFS_H.
4600 28. Installed the latest version of exipick from John Jetmore.
4602 29. In an address list, if the pattern was not a regular expression, an empty
4603 subject address (from a bounce message) matched only if the pattern was an
4604 empty string. Non-empty patterns were not even tested. This was the wrong
4605 because it is perfectly reasonable to use an empty address as part of a
4606 database query. An empty address is now tested by patterns that are
4607 lookups. However, all the other forms of pattern expect the subject to
4608 contain a local part and a domain, and therefore, for them, an empty
4609 address still always fails if the pattern is not itself empty.
4611 30. Exim went into a mad DNS loop when attempting to do a callout where the
4612 host was specified on an smtp transport, and looking it up yielded more
4613 than one IP address.
4615 31. Re-factored the code for checking spool and log partition space into a
4616 function that finds that data and another that does the check. The former
4617 is then used to implement four new variables: $spool_space, $log_space,
4618 $spool_inodes, and $log_inodes.
4620 32. The RFC2047 encoding function was originally intended for short strings
4621 such as real names; it was not keeping to the 75-character limit for
4622 encoded words that the RFC imposes. It now respects the limit, and
4623 generates multiple encoded words if necessary. To be on the safe side, I
4624 have increased the buffer size for the ${rfc2047: expansion operator from
4627 33. It is now permitted to omit both strings after an "if" condition; if the
4628 condition is true, the result is "true". As before, when the second string
4629 is omitted, a false condition yields an empty string. This makes it less
4630 cumbersome to write custom ACL and router conditions.
4632 34. Failure to deliver a bounce message always caused it to be frozen, even if
4633 there was an errors_to setting on the router. The errors_to setting is now
4636 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4637 canonical form (fully expanded) before being placed in
4638 $sender_host_address.
4640 36. The table in the code that translates DNS record types into text (T_A to
4641 "A" for instance) was missing entries for NS and CNAME. It is just possible
4642 that this could have caused confusion if both these types were looked up
4643 for the same domain, because the text type is used as part of Exim's
4644 per-process caching. But the chance of anyone hitting this buglet seems
4647 37. The dnsdb lookup has been extended in a number of ways.
4649 (1) There is a new type, "zns", which walks up the domain tree until it
4650 finds some nameserver records. It should be used with care.
4652 (2) There is a new type, "mxh", which is like "mx" except that it returns
4653 just the host names, not the priorities.
4655 (3) It is now possible to give a list of domains (or IP addresses) to be
4656 looked up. The behaviour when one of the lookups defers can be
4657 controlled by a keyword.
4659 (4) It is now possible to specify the separator character for use when
4660 multiple records are returned.
4662 38. The dnslists ACL condition has been extended: it is now possible to supply
4663 a list of IP addresses and/or domains to be looked up in a particular DNS
4666 39. Added log_selector=+queue_time_overall.
4668 40. When running the queue in the test harness, wait just a tad after forking a
4669 delivery process, to get repeatability of debugging output.
4671 41. Include certificate and key file names in error message when GnuTLS fails
4672 to set them up, because the GnuTLS error message doesn't include the name
4673 of the failing file when there is a problem reading it.
4675 42. Allow both -bf and -bF in the same test run.
4677 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4679 44. The "Exiscan patch" is now merged into the mainline Exim source.
4681 45. Sometimes the final signoff response after QUIT could fail to get
4682 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4683 < 0 before doing a fflush(). This bug looks as though it goes back to the
4684 introduction of TLS in release 3.20, but "sometimes" must have been rare
4685 because the tests only now provoked it.
4687 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4688 (this can affect the format of dates).
4690 47. exim_tidydb, when checking for the continued existence of a message for
4691 which it has found a message-specific retry record, was not finding
4692 messages that were in split spool directories. Consequently, it was
4693 deleting retry records that should have stayed in existence.
4695 48. Steve fixed some bugs in eximstats.
4697 49. The SPA authentication driver was not abandoning authentication and moving
4698 on to the next authenticator when an expansion was forced to fail,
4699 contradicting the general specification for all authenticators. Instead it
4700 was generating a temporary error. It now behaves as specified.
4702 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4703 (the order specifies the preference for clients). The order is now AES256,
4704 AES128, 3DES, ARCFOUR128.
4706 51. Small patch to Sieve code - explicitly set From: when generating an
4709 52. Exim crashed if a remote delivery caused a very long error message to be
4710 recorded - for instance if somebody sent an entire SpamAssassin report back
4711 as a large number of 550 error lines. This bug was coincidentally fixed by
4712 increasing the size of one of Exim's internal buffers (big_buffer) that
4713 happened as part of the Exiscan merge. However, to be on the safe side, I
4714 have made the code more robust (and fixed the comments that describe what
4717 53. Now that there can be additional text after "Completed" in log lines (if
4718 the queue_time_overall log selector is set), a one-byte patch to exigrep
4719 was needed to allow it to recognize "Completed" as not the last thing in
4722 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4723 patch that reportedly fixes this has been added. I am not expert enough to
4724 create a test for it. This is what the patch creator wrote:
4726 "I found a little strange behaviour of ldap code when working with
4727 Windows 2003 AD Domain, where users was placed in more than one
4728 Organization Units. When I tried to give exim partial DN, the exit code
4729 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4730 But simultaneously result of request was absolutely normal ldap result,
4731 so I produce this patch..."
4733 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4734 REFERENCE, so I have modified the code to exclude the patch when that macro
4737 55. Some experimental protocols are using DNS PTR records for new purposes. The
4738 keys for these records are domain names, not reversed IP addresses. The
4739 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4740 leaves it alone. Component reversal etc. now happens only for IP addresses.
4743 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4745 57. Double the size of the debug message buffer (to 2048) so that more of very
4746 long debug lines gets shown.
4748 58. The exicyclog utility now does better if the number of log files to keep
4749 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4751 59. Two changes related to the smtp_active_hostname option:
4753 (1) $smtp_active_hostname is now available as a variable.
4754 (2) The default for smtp_banner uses $smtp_active_hostname instead
4755 of $primary_hostname.
4757 60. The host_aton() function is supposed to be passed a string that is known
4758 to be a valid IP address. However, in the case of IPv6 addresses, it was
4759 not checking this. This is a hostage to fortune. Exim now panics and dies
4760 if the condition is not met. A case was found where this could be provoked
4761 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4762 components; fortuitously, this particular loophole had already been fixed
4763 by change 4.50/55 above.
4765 If there are any other similar loopholes, the new check in host_aton()
4766 itself should stop them being exploited. The report I received stated that
4767 data on the command line could provoke the exploit when Exim was running as
4768 exim, but did not say which command line option was involved. All I could
4769 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4770 running as the user.
4773 61. There was a buffer overflow vulnerability in the SPA authentication code
4774 (which came originally from the Samba project). I have added a test to the
4775 spa_base64_to_bits() function which I hope fixes it.
4778 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4779 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4781 63. The daemon start-up calls getloadavg() while still root for those OS that
4782 need the first call to be done as root, but it missed one case: when
4783 deliver_queue_load_max is set with deliver_drop_privilege. This is
4784 necessary for the benefit of the queue runner, because there is no re-exec
4785 when deliver_drop_privilege is set.
4787 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4788 This has been fixed.
4790 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4791 was in use, was not putting the data itself into the right store pool;
4792 consequently, it could be overwritten for a subsequent message in the same
4793 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4796 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4798 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4799 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4800 the function were treating the return as a boolean value, which happened to
4801 work because 0=false and not-0=true, but is not correct code.
4803 68. The host_aton() function was not handling scoped IPv6 addresses (those
4804 with, for example, "%eth0" on the end) correctly.
4806 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4807 negated items (that is, ~something) in unsigned ints. Some compilers
4808 apparently mutter when there is no cast.
4810 70. If an address verification called from an ACL failed, and did not produce a
4811 user-specific message (i.e. there was only a "system" message), nothing was
4812 put in $acl_verify_message. In this situation, it now puts the system
4815 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4816 session; change 4.31/43 added the unwanted input to the log line - except
4817 that it did not do this in the start of session case. It now does.
4819 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4820 This is wrong and can cause the other end to generate a synchronization
4821 error if it is another Exim or anything else that does the synchronization
4822 check. A QUIT command is no longer sent after a timeout.
4824 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4825 during host lookups.
4827 74. The defer_ok option of callout verification was not working if it was used
4828 when verifying addresses in header lines, that is, for this case:
4830 verify = header_sender/callout=defer_ok
4832 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4833 those file descriptors could be used for SMTP connections. If anything
4834 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4835 could be sent to the SMTP client, causing chaos. The daemon now opens
4836 stdin, stdout, and stderr to /dev/null when it puts itself into the
4839 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4840 log by default. The user can override this with suitable Perl magic.
4842 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4843 the log message when discard triggers, was not working for the DATA ACL or
4844 for the non-SMTP ACL.
4846 78. Error message wording change in sieve.c.
4848 79. If smtp_accept_max_per_host was set, the number of connections could be
4849 restricted to fewer than expected, because the daemon was trying to set up
4850 a new connection before checking whether the processes handling previous
4851 connections had finished. The check for completed processes is now done
4852 earlier. On busy systems, this bug wouldn't be noticed because something
4853 else would have woken the daemon, and it would have reaped the completed
4856 80. If a message was submitted locally by a user whose login name contained one
4857 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4858 It caused a spool format error. I have fixed the spool reading code. A
4859 related problem was that the "from" clause in the Received: line became
4860 illegal because of the space(s). It is now covered by ${quote_local_part.
4862 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4865 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4867 83. Patch from Sieve maintainer to fix unterminated string problem in
4868 "vacation" handling.
4870 84. Some minor changes to the Linux configuration files to help with other
4871 OS variants using glibc.
4873 85. One more patch for Sieve to update vacation handling to latest spec.
4876 ----------------------------------------------------
4877 See the note above about the 4.44 and 4.50 releases.
4878 ----------------------------------------------------
4884 1. Change 4.43/35 introduced a bug that caused file counts to be
4885 incorrectly computed when quota_filecount was set in an appendfile
4888 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4889 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4892 3. Give more explanation in the error message when the command for a transport
4893 filter fails to execute.
4895 4. There are several places where Exim runs a non-Exim command in a
4896 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4897 was being done only for the command run by the queryprogram router. It is
4898 now done for all such subprocesses. The other cases are: ${run, transport
4899 filters, and the commands run by the lmtp and pipe transports.
4901 5. Some older OS have a limit of 256 on the maximum number of file
4902 descriptors. Exim was using setrlimit() to set 1000 as a large value
4903 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4904 systems. I've change it so that if it can't get 1000, it tries for 256.
4906 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4907 was an oversight, and furthermore, ever since the addition of extra
4908 controls (e.g. 4.43/32), the checks on when to allow different forms of
4909 "control" were broken. There should now be diagnostics for all cases when a
4910 control that does not make sense is encountered.
4912 7. $recipients is now available in the predata ACL (oversight).
4914 8. Tidy the search cache before the fork to do a delivery from a message
4915 received from the command line. Otherwise the child will trigger a lookup
4916 failure and thereby defer the delivery if it tries to use (for example) a
4917 cached ldap connection that the parent has called unbind on.
4919 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4920 of $address_data from the recipient verification was clobbered by the
4921 sender verification.
4923 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4924 was its contents. (It was OK if the option was not defined at all.)
4926 11. A "Completed" log line is now written for messages that are removed from
4927 the spool by the -Mrm option.
4929 12. $host_address is now set to the target address during the checking of
4930 ignore_target_hosts.
4932 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4933 being passed; this would have caused $sender_host_name to have been used if
4934 matching the list had actually called for a host name (not very likely,
4935 since this list is usually IP addresses). A host name is now passed as
4938 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4939 code that specifies a non-restarting handler (typically sigaction() in
4940 modern systems) in an attempt to fix a rare and obscure crash bug.
4942 15. Narrowed the window for a race in the daemon that could cause it to ignore
4943 SIGCHLD signals. This is not a major problem, because they are used only to
4944 wake it up if nothing else does.
4946 16. A malformed maildirsize file could cause Exim to calculate negative values
4947 for the mailbox size or file count. Odd effects could occur as a result.
4948 The maildirsize information is now recalculated if the size or filecount
4951 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4952 support for a long time. Removed HAVE_SYS_VFS_H.
4954 18. Updated exipick to current release from John Jetmore.
4956 19. Allow an empty sender to be matched against a lookup in an address list.
4957 Previously the only cases considered were a regular expression, or an
4960 20. Exim went into a mad DNS lookup loop when doing a callout where the
4961 host was specified on the transport, if the DNS lookup yielded more than
4964 21. The RFC2047 encoding function was originally intended for short strings
4965 such as real names; it was not keeping to the 75-character limit for
4966 encoded words that the RFC imposes. It now respects the limit, and
4967 generates multiple encoded words if necessary. To be on the safe side, I
4968 have increased the buffer size for the ${rfc2047: expansion operator from
4971 22. Failure to deliver a bounce message always caused it to be frozen, even if
4972 there was an errors_to setting on the router. The errors_to setting is now
4975 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4976 canonical form (fully expanded) before being placed in
4977 $sender_host_address.
4979 24. Updated eximstats to version 1.33
4981 25. Include certificate and key file names in error message when GnuTLS fails
4982 to set them up, because the GnuTLS error message doesn't include the name
4983 of the failing file when there is a problem reading it.
4985 26. Expand error message when OpenSSL has problems setting up cert/key files.
4988 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4989 (this can affect the format of dates).
4991 28. exim_tidydb, when checking for the continued existence of a message for
4992 which it has found a message-specific retry record, was not finding
4993 messages that were in split spool directories. Consequently, it was
4994 deleting retry records that should have stayed in existence.
4996 29. eximstats updated to version 1.35
4997 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4998 1.35 - bugfix such that pie charts by volume are generated correctly
5000 30. The SPA authentication driver was not abandoning authentication and moving
5001 on to the next authenticator when an expansion was forced to fail,
5002 contradicting the general specification for all authenticators. Instead it
5003 was generating a temporary error. It now behaves as specified.
5005 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5006 (the order specifies the preference for clients). The order is now AES256,
5007 AES128, 3DES, ARCFOUR128.
5009 31. Small patch to Sieve code - explicitly set From: when generating an
5012 32. Exim crashed if a remote delivery caused a very long error message to be
5013 recorded - for instance if somebody sent an entire SpamAssassin report back
5014 as a large number of 550 error lines. This bug was coincidentally fixed by
5015 increasing the size of one of Exim's internal buffers (big_buffer) that
5016 happened as part of the Exiscan merge. However, to be on the safe side, I
5017 have made the code more robust (and fixed the comments that describe what
5020 33. Some experimental protocols are using DNS PTR records for new purposes. The
5021 keys for these records are domain names, not reversed IP addresses. The
5022 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5023 leaves it alone. Component reversal etc. now happens only for IP addresses.
5026 34. The host_aton() function is supposed to be passed a string that is known
5027 to be a valid IP address. However, in the case of IPv6 addresses, it was
5028 not checking this. This is a hostage to fortune. Exim now panics and dies
5029 if the condition is not met. A case was found where this could be provoked
5030 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5031 components; fortuitously, this particular loophole had already been fixed
5032 by change 4.50/55 or 4.44/33 above.
5034 If there are any other similar loopholes, the new check in host_aton()
5035 itself should stop them being exploited. The report I received stated that
5036 data on the command line could provoke the exploit when Exim was running as
5037 exim, but did not say which command line option was involved. All I could
5038 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5039 running as the user.
5042 35. There was a buffer overflow vulnerability in the SPA authentication code
5043 (which came originally from the Samba project). I have added a test to the
5044 spa_base64_to_bits() function which I hope fixes it.
5047 36. The daemon start-up calls getloadavg() while still root for those OS that
5048 need the first call to be done as root, but it missed one case: when
5049 deliver_queue_load_max is set with deliver_drop_privilege. This is
5050 necessary for the benefit of the queue runner, because there is no re-exec
5051 when deliver_drop_privilege is set.
5053 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5054 was in use, was not putting the data itself into the right store pool;
5055 consequently, it could be overwritten for a subsequent message in the same
5056 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5059 38. Sometimes the final signoff response after QUIT could fail to get
5060 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5061 < 0 before doing a fflush(). This bug looks as though it goes back to the
5062 introduction of TLS in release 3.20, but "sometimes" must have been rare
5063 because the tests only now provoked it.
5069 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5070 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5071 Now it returns an integer. A number of calls were still expecting a T/F
5072 return. Fortuitously, in all cases, the tests worked in OK situations,
5073 which is the norm. However, things would have gone wrong on any write
5074 failures on the smtp file descriptor. This function is used when sending
5075 messages over SMTP and also when doing verify callouts.
5077 2. When Exim is called to do synchronous delivery of a locally submitted
5078 message (the -odf or -odi options), it no longer closes stderr before doing
5081 3. Implemented the mua_wrapper option.
5083 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5085 5. Implemented the functions header_remove(), header_testname(),
5086 header_add_at_position(), and receive_remove_recipient(), and exported them
5089 6. If an ACL "warn" statement specified the addition of headers, Exim already
5090 inserted X-ACL-Warn: at the start if there was no header name. However, it
5091 was not making this test for the second and subsequent header lines if
5092 there were newlines in the string. This meant that an invalid header could
5093 be inserted if Exim was badly configured.
5095 7. Allow an ACL "warn" statement to add header lines at the start or after all
5096 the Received: headers, as well as at the end.
5098 8. Added the rcpt_4xx retry error code.
5100 9. Added postmaster_mailfrom=xxx to callout verification option.
5102 10. Added mailfrom=xxxx to the callout verification option, for verify=
5105 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5106 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5108 12. Inserted some casts to stop certain compilers warning when using pointer
5109 differences as field lengths or precisions in printf-type calls (mostly
5110 affecting debugging statements).
5112 13. Added optional readline() support for -be (dynamically loaded).
5114 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5115 same clock tick as a message's arrival, so that its received time was the
5116 same as the "first fail" time on the retry record, and that message
5117 remained on the queue past the ultimate address timeout, every queue runner
5118 would try a delivery (because it was past the ultimate address timeout) but
5119 after another failure, the ultimate address timeout, which should have then
5120 bounced the address, did not kick in. This was a "< instead of <=" error;
5121 in most cases the first failure would have been in the next clock tick
5122 after the received time, and all would be well.
5124 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5125 being recognized when the domain list was tested by the match_domain
5126 condition in an expansion string.
5128 16. Added the ${str2b64: operator.
5130 17. Exim was always calling setrlimit() to set a large limit for the number of
5131 processes, without checking whether the existing limit was already
5132 adequate. (It did check for the limit on file descriptors.) Furthermore,
5133 errors from getrlimit() and setrlimit() were being ignored. Now they are
5134 logged to the main and panic logs, but Exim does carry on, to try to do its
5135 job under whatever limits there are.
5137 18. Imported PCRE 5.0.
5139 19. Trivial typo in log message " temporarily refused connection" (the leading
5142 20. If the log selector return_path_on_delivery was set and an address was
5143 redirected to /dev/null, the delivery process crashed because it assumed
5144 that a return path would always be set for a "successful" delivery. In this
5145 case, the whole delivery is bypassed as an optimization, and therefore no
5148 21. Internal re-arrangement: the function for sending a challenge and reading
5149 a response while authentication was assuming a zero-terminated challenge
5150 string. It's now changed to take a pointer and a length, to allow for
5151 binary data in such strings.
5153 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5155 23. Exim was not respecting finduser_retries when seeking the login of the
5156 uid under which it was called; it was always trying 10 times. (The default
5157 setting of finduser_retries is zero.) Also, it was sleeping after the final
5158 failure, which is pointless.
5160 24. Implemented tls_on_connect_ports.
5162 25. Implemented acl_smtp_predata.
5164 26. If the domain in control=submission is set empty, Exim assumes that the
5165 authenticated id is a complete email address when it generates From: or
5166 Sender: header lines.
5168 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5169 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5170 chown and chgrp in /bin and hostname in /usr/bin.
5172 28. Exim was keeping the "process log" file open after each use, just as it
5173 does for the main log. This opens the possibility of it remaining open for
5174 long periods when the USR1 signal hits a daemon. Occasional processlog
5175 errors were reported, that could have been caused by this. Anyway, it seems
5176 much more sensible not to leave this file open at all, so that is what now
5179 29. The long-running daemon process does not normally write to the log once it
5180 has entered its main loop, and it closes the log before doing so. This is
5181 so that log files can straightforwardly be renamed and moved. However,
5182 there are a couple of unusual error situations where the daemon does write
5183 log entries, and I had neglected to close the log afterwards.
5185 30. The text of an SMTP error response that was received during a remote
5186 delivery was being truncated at 512 bytes. This is too short for some of
5187 the long messages that one sometimes sees. I've increased the limit to
5190 31. It is now possible to make retry rules that apply only when a message has a
5191 specific sender, in particular, an empty sender.
5193 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5194 it possible to be selective about when SMTP synchronization is enforced.
5196 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5198 32. Implemented hosts_connection_nolog.
5200 33. Added an ACL for QUIT.
5202 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5205 35. Added mailbox_size and mailbox_filecount to appendfile.
5207 36. Added control = no_multiline_responses to ACLs.
5209 37. There was a bug in the logic of the code that waits for the clock to tick
5210 in the case where the clock went backwards by a substantial amount such
5211 that the microsecond fraction of "now" was more than the microsecond
5212 fraction of "then" (but the whole seconds number was less).
5214 38. Added support for the libradius Radius client library this is found on
5215 FreeBSD (previously only the radiusclient library was supported).
5221 1. When certain lookups returned multiple values in the form name=value, the
5222 quoting of the values was not always being done properly. Specifically:
5223 (a) If the value started with a double quote, but contained no whitespace,
5225 (b) If the value contained whitespace other than a space character (i.e.
5226 tabs or newlines or carriage returns) it was not quoted.
5227 This fix has been applied to the mysql and pgsql lookups by writing a
5228 separate quoting function and calling it from the lookup code. The fix
5229 should probably also be applied to nisplus, ibase and oracle lookups, but
5230 since I cannot test any of those, I have not disturbed their existing code.
5232 2. A hit in the callout cache for a specific address caused a log line with no
5233 reason for rejecting RCPT. Now it says "Previous (cached) callout
5234 verification failure".
5236 3. There was an off-by-one bug in the queryprogram router. An over-long
5237 return line was truncated at 256 instead of 255 characters, thereby
5238 overflowing its buffer with the terminating zero. As well as fixing this, I
5239 have increased the buffer size to 1024 (and made a note to document this).
5241 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5242 when Exim is waiting for an SMTP response from a remote server, Exim
5243 restarts its select() call on the socket, thereby resetting its timeout.
5244 This is not a problem when such interrupts are rare. Somebody set up a cron
5245 job to run exiwhat every 2 minutes, which is less than the normal select()
5246 timeout (5 or 10 minutes). This meant that the select() timeout never
5247 kicked in because it was always reset. I have fixed this by comparing the
5248 time when an interrupt arrives with the time at the start of the first call
5249 to select(). If more time than the timeout has elapsed, the interrupt is
5250 treated as a timeout.
5252 5. Some internal re-factoring in preparation for the addition of Sieve
5253 extensions (by MH). In particular, the "personal" test is moved to a
5254 separate function, and given an option for scanning Cc: and Bcc: (which is
5255 not set for Exim filters).
5257 6. When Exim created an email address using the login of the caller as the
5258 local part (e.g. when creating a From: or Sender: header line), it was not
5259 quoting the local part when it contained special characters such as @.
5261 7. Installed new OpenBSD configuration files.
5263 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5264 try to make them clearer.
5266 9. Callout options, other than the timeout value, were being ignored when
5267 verifying sender addresses in header lines. For example, when using
5269 verify = header_sender/callout=no_cache
5271 the cache was (incorrectly) being used.
5273 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5274 only the Cygwin environment.
5276 11. When return_path_on_delivery was set as a log selector, if different remote
5277 addresses in the same message used different return paths and parallel
5278 remote delivery occurred, the wrong values would sometimes be logged.
5279 (Whenever a remote delivery process finished, the return path value from
5280 the most recently started remote delivery process was logged.)
5282 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5283 lines when AUTH and/or TLS are in use. This is the "received protocol"
5284 field. Exim used to use "asmtp" for authenticated SMTP, without any
5285 indication (in the protocol name) for TLS use. Now it follows the RFC and
5286 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5287 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5288 names appear in log lines as well as in Received: header lines.
5290 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5291 extensions, and comparison tests, and to fix some bugs.
5293 14. Changes to the "personal" filter test:
5295 (1) The test was buggy in that it was just doing the equivalent of
5296 "contains" tests on header lines. For example, if a user's address was
5297 anne@some.where, the "personal" test would incorrectly be true for
5299 To: susanne@some.where
5301 This test is now done by extracting each address from the header in turn,
5302 and checking the entire address. Other tests that are part of "personal"
5303 are now done using regular expressions (for example, to check local parts
5304 of addresses in From: header lines).
5306 (2) The list of non-personal local parts in From: addresses has been
5307 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5308 taken from the Sieve specification recommendations.
5310 (3) If the message contains any header line starting with "List-" it is
5311 treated as non-personal.
5313 (4) The test for "circular" in the Subject: header line has been removed
5314 because it now seems ill-conceived.
5316 15. Minor typos in src/EDITME comments corrected.
5318 16. Installed latest exipick from John Jetmore.
5320 17. If headers_add on a router specified a text string that was too long for
5321 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5322 of string_sprintf() is now avoided.
5324 18. $message_body_size was not set (it was always zero) when running the DATA
5325 ACL and the local_scan() function.
5327 19. For the "mail" command in an Exim filter, no default was being set for
5328 the once_repeat time, causing a random time value to be used if "once" was
5329 specified. (If the value happened to be <= 0, no repeat happened.) The
5330 default is now 0s, meaning "never repeat". The "vacation" command was OK
5331 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5332 (I found it when inspecting the code).
5334 20. There is now an overall timeout for performing a callout verification. It
5335 defaults to 4 times the callout timeout, which applies to individual SMTP
5336 commands during the callout. The overall timeout applies when there is more
5337 than one host that can be tried. The timeout is checked before trying the
5338 next host. This prevents very long delays if there are a large number of
5339 hosts and all are timing out (e.g. when the network connections are timing
5340 out). The value of the overall timeout can be changed by specifying an
5341 additional sub-option for "callout", called "maxwait". For example:
5343 verify = sender/callout=5s,maxwait=20s
5345 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5346 to the end before writing, but this should make it even safer).
5348 22. Exim was forgetting that it had advertised PIPELINING for the second and
5349 subsequent messages on an SMTP connection. It was also not resetting its
5350 memory on STARTTLS and an internal HELO.
5352 23. When Exim logs an SMTP synchronization error within a session, it now
5353 records whether PIPELINING has been advertised or not.
5355 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5356 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5357 rather than long int.
5359 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5361 26. Added the never_mail option to autoreply.
5367 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5368 crash if the getsockname() call failed; this can happen if a connection is
5369 closed very soon after it is established. The problem was simply in the
5370 order in which certain operations were done, causing Exim to try to write
5371 to the SMTP stream before it had set up the file descriptor. The bug has
5372 been fixed by making things happen in the correct order.
5378 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5379 before the connection was closed, thus losing the rejection response.
5381 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5382 some early Solaris releases, but causes trouble in current releases where
5383 socklen_t is defined.
5385 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5388 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5391 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5392 that releasing the top of it at the end releases what was used for sub-
5393 expansions (unless the block got too big). However, discard this block if
5394 the first thing is a variable or header, so that we can use its block when
5395 it is dynamic (useful for very large $message_headers, for example).
5397 6. Lookups now cache *every* query, not just the most recent. A new, separate
5398 store pool is used for this. It can be recovered when all lookup caches are
5399 flushed. Lookups now release memory at the end of their result strings.
5400 This has involved some general refactoring of the lookup sources.
5402 7. Some code has been added to the store_xxx() functions to reduce the amount
5403 of flapping under certain conditions.
5405 8. log_incoming_interface used to affect only the <= reception log lines. Now
5406 it causes the local interface and port to be added to several more SMTP log
5407 lines, for example "SMTP connection from", and rejection lines.
5409 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5411 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5413 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5414 could be overwritten at the end of the current message (or the start of a
5415 new message if it was set in a HELO ACL). The value is now preserved for
5416 the duration of the SMTP connection.
5418 12. If a transport had a headers_rewrite setting, and a matching header line
5419 contained an unqualified address, that address was qualified, even if it
5420 did not match any rewriting rules. The underlying bug was that the values
5421 of the flags that permit the existence of unqualified sender and recipient
5422 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5423 non-local messages, and by -bnq for local messages) were not being
5424 preserved with the message after it was received.
5426 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5427 "next input=" as part of the text comprising the host identity instead of
5428 the correct text. The code was using the same buffer for two different
5429 strings. However, depending on which order the printing function evaluated
5430 its arguments, the bug did not always show up. Under Linux, for example, my
5431 test suite worked just fine.
5433 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5434 doesn't work with some older versions of Perl. It has been changed to "my",
5435 which in any case is probably the better facility to use.
5437 15. A really picky compiler found some instances of statements for creating
5438 error messages that either had too many or two few arguments for the format
5441 16. The size of the buffer for calls to the DNS resolver has been increased
5442 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5443 for addresses that have a lot of PTR records. This alleviates a problem; it
5444 does not fully solve it.
5446 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5447 buffer now truncates the list and logs the incident, which is the same
5448 action as happens when Exim is looking up a host name and its aliases.
5449 Previously in this situation something unpredictable would happen;
5450 sometimes it was "internal error: store_reset failed".
5452 18. If a server dropped the connection unexpectedly when an Exim client was
5453 using GnuTLS and trying to read a response, the client delivery process
5454 crashed while trying to generate an error log message.
5456 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5457 string, for example:
5459 warn message = H1: something\nH2: something
5461 the text was added as a single header line from Exim's point of view
5462 though it ended up OK in the delivered message. However, searching for the
5463 second and subsequent header lines using $h_h2: did not work. This has been
5464 fixed. Similarly, if a system filter added multiple headers in this way,
5465 the routers could not see them.
5467 20. Expanded the error message when iplsearch is called with an invalid key to
5468 suggest using net-iplsearch in a host list.
5470 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5471 ACLs are no longer actually imposed (and a message to that effect is
5474 22. If a "gecos" field in a passwd entry contained escaped characters, in
5475 particular, if it contained a \" sequence, Exim got it wrong when building
5476 a From: or a Sender: header from that name. A second bug also caused
5477 incorrect handling when an unquoted " was present following a character
5478 that needed quoting.
5480 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5481 was not being matched caselessly.
5483 24. Arranged for all hyphens in the exim.8 source to be escaped with
5486 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5487 properly. Recipient callout cache records were still being keyed to include
5488 the sender, even when use_sender was set false. This led to far more
5489 callouts that were necessary. The sender is no longer included in the key
5490 when use_sender is false.
5492 26. Added "control = submission" modifier to ACLs.
5494 27. Added the ${base62d: operator to decode base 62 numbers.
5496 28. dnsdb lookups can now access SRV records.
5498 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5499 the configuration file.
5501 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5502 (-v) mode. This makes the output for a verbose queue run more intelligible.
5504 31. Added a use_postmaster feature to recipient callouts.
5506 32. Added the $body_zerocount variable, containing the number of binary zero
5507 bytes in the message body.
5509 33. The time of last modification of the "new" subdirectory is now used as the
5510 "mailbox time last read" when there is a quota error for a maildir
5513 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5515 35. Added +ignore_unknown as a special item in host lists.
5517 36. Code for decoding IPv6 addresses in host lists is now included, even if
5518 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5519 address was recognized as an IP address, but was then not correctly decoded
5520 into binary, causing unexpected and incorrect effects when compared with
5527 1. Very minor rewording of debugging text in manualroute to say "list of
5528 hosts" instead of "hostlist".
5530 2. If verify=header_syntax was set, and a header line with an unqualified
5531 address (no domain) and a large number of spaces between the end of the
5532 name and the colon was received, the reception process suffered a buffer
5533 overflow, and (when I tested it) crashed. This was caused by some obsolete
5534 code that should have been removed. The fix is to remove it!
5536 3. When running in the test harness, delay a bit after writing a bounce
5537 message to get a bit more predictability in the log output.
5539 4. Added a call to search_tidyup() just before forking a reception process. In
5540 theory, someone could use a lookup in the expansion of smtp_accept_max_
5541 per_host which, without the tidyup, could leave open a database connection.
5543 5. Added the variables $recipient_data and $sender_data which get set from a
5544 lookup success in an ACL "recipients" or "senders" condition, or a router
5545 "senders" option, similar to $domain_data and $local_part_data.
5547 6. Moved the writing of debug_print from before to after the "senders" test
5550 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5551 problems for message scanning, either using a data ACL, or using
5552 local_scan() because the Received: header was not generated till after they
5553 were called (in order to set the time as the time of reception completion).
5554 I have revised the way this works. The header is now generated after the
5555 body is received, but before the ACL or local_scan() are called. After they
5556 are run, the timestamp in the header is updated.
5562 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5563 before starting a queue runner without re-exec. This happened only when
5564 deliver_drop_privilege was set or when the Exim user was set to root. The
5565 effect of the bug was that timeouts during subsequent deliveries caused
5566 crashes instead of being properly handled. The handler is now left at its
5567 default (and expected) setting.
5569 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5570 message, again when deliver_drop_privilege is set or Exim is run as root.
5571 The bug described in (1) was not present in this case, but the tidying up
5572 of the other signals was missing. I have made the two cases consistent.
5574 3. The ignore_target_hosts setting on a manualroute router was being ignored
5575 for hosts that were looked up using the /MX notation.
5577 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5580 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5581 operated on the sender address. After changing the $sender_address to <>
5582 for the sender address verify, Exim was re-instated it as the original
5583 (before rewriting) address, but remembering that it had rewritten it, so it
5584 wasn't rewriting it again. This bug also had the effect of breaking the
5585 sender address verification caching when the sender address was rewritten.
5587 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5588 This has been changed so that if the ip literal address matches
5589 ignore_target_hosts, the router declines.
5591 7. Added expansion conditions match_domain, match_address, and match_local_
5592 part (NOT match_host).
5594 8. The placeholder for the Received: header didn't have a length field set.
5596 9. Added code to Exim itself and to exim_lock to test for a specific race
5597 condition that could lead to file corruption when using MBX delivery. The
5598 issue is with the lockfile that is created in /tmp. If this file is removed
5599 after a process has opened it but before that process has acquired a lock,
5600 there is the potential for a second process to recreate the file and also
5601 acquire a lock. This could lead to two Exim processes writing to the file
5602 at the same time. The added code performs the same test as UW imapd; it
5603 checks after acquiring the lock that its file descriptor still refers to
5604 the same named file.
5606 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5607 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5610 11. Added the smtp_active_hostname option. If used, this will typically be made
5611 to depend on the incoming interface address. Because $interface_address is
5612 not set up until the daemon has forked a reception process, error responses
5613 that can happen earlier (such as "too many connections") no longer contain
5616 12. If an expansion in a condition on a "warn" statement fails because a lookup
5617 defers, the "warn" statement is abandoned, and the next ACL statement is
5618 processed. Previously this caused the whole ACL to be aborted.
5620 13. Added the iplsearch lookup type.
5622 14. Added ident_timeout as a log selector.
5624 15. Added tls_certificate_verified as a log selector.
5626 16. Added a global option tls_require_ciphers (compare the smtp transport
5627 option of the same name). This controls incoming TLS connections.
5629 17. I finally figured out how to make tls_require_ciphers do a similar thing
5630 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5631 before starting the TLS session.
5633 18. Tabs are now shown as \t in -bP output.
5635 19. If the log selector return_path_on_delivery was set, Exim crashed when
5636 bouncing a message because it had too many Received: header lines.
5638 20. If two routers both had headers_remove settings, and the first one included
5639 a superfluous trailing colon, the final name in the first list and the
5640 first name in the second list were incorrectly joined into one item (with a
5641 colon in the middle).
5647 1. Added -C and -D options to the exinext utility, mainly to make it easier
5648 to include in the automated testing, but these could be helpful when
5649 multiple configurations are in use.
5651 2. The exinext utility was not formatting the output nicely when there was
5652 an alternate port involved in the retry record key, nor when there was a
5653 message id as well (for retries that were specific to a specific message
5654 and a specific host). It was also confused by IPv6 addresses, because of
5655 the additional colons they contain. I have fixed the IPv4 problem, and
5656 patched it up to do a reasonable job for IPv6.
5658 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5659 delivery, the log line now contains "pipelined" if PIPELINING was used.
5661 4. An SMTP transport process used to panic and die if the bind() call to set
5662 an explicit outgoing interface failed. This has been changed; it is now
5663 treated in the same way as a connect() failure.
5665 5. A reference to $sender_host_name in the part of a conditional expansion
5666 that was being skipped was still causing a DNS lookup. This no longer
5669 6. The def: expansion condition was not recognizing references to header lines
5670 that used bh_ and bheader_.
5672 7. Added the _cache feature to named lists.
5674 8. The code for checking quota_filecount in the appendfile transport was
5675 allowing one more file than it should have been.
5677 9. For compatibility with Sendmail, the command line option
5685 and sets the incoming protocol and host name (for trusted callers). The
5686 host name and its colon can be omitted when only the protocol is to be set.
5687 Note the Exim already has two private options, -pd and -ps, that refer to
5688 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5689 "s", but I don't think that's a major issue.
5691 10. A number of refactoring changes to the code, none of which should affect
5694 (a) The number of logging options was getting close to filling up the
5695 32-bit word that was used as a bit map. I have split them into two classes:
5696 those that are passed in the argument to log_write(), and those that are
5697 only ever tested independently outside of that function. These are now in
5698 separate 32-bit words, so there is plenty of room for expansion again.
5699 There is no change in the user interface or the logging behaviour.
5701 (b) When building, for example, log lines, the code previously used a
5702 macro that called string_cat() twice, in order to add two strings. This is
5703 not really sufficiently general. Furthermore, there was one instance where
5704 it was actually wrong because one of the argument was used twice, and in
5705 one call a function was used. (As it happened, calling the function twice
5706 did not affect the overall behaviour.) The macro has been replaced by a
5707 function that can join an arbitrary number of extra strings onto a growing
5710 (c) The code for expansion conditions now uses a table and a binary chop
5711 instead of a serial search (which was left over from when there were very
5712 few conditions). Also, it now recognizes conditions like "pam" even when
5713 the relevant support is not compiled in: a suitably worded error message is
5714 given if an attempt is made to use such a condition.
5716 11. Added ${time_interval:xxxxx}.
5718 12. A bug was causing one of the ddress fields not to be passed back correctly
5719 from remote delivery subprocesses. The field in question was not being
5720 subsequently used, so this caused to problems in practice.
5722 13. Added new log selectors queue_time and deliver_time.
5724 14. Might have fixed a bug in maildirsizefile handling that threw up
5725 "unexpected character" debug warnings, and recalculated the data
5726 unnecessarily. In any case, I expanded the warning message to give more
5729 15. Added the message "Restricted characters in address" to the statements in
5730 the default ACL that block characters like @ and % in local parts.
5732 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5733 Three changes have been made:
5735 (a) There was a serious bug; a negative response to MAIL caused the whole
5736 recipient domain to be cached as invalid, thereby blocking all messages
5737 to all local parts at the same domain, from all senders. This bug has
5738 been fixed. The domain is no longer cached after a negative response to
5739 MAIL if the sender used is not empty.
5741 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5744 (c) A new callout option, "use_sender" has been added for people who want
5745 the modified behaviour.
5751 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5754 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5755 indeed breaks things for older releases.
5757 3. Added additional logging to the case where there is a problem reading data
5758 from a filter that is running in a subprocess using a pipe, in order to
5759 try to track down a specific problem.
5761 4. Testing facility fudge: when running in the test harness and attempting
5762 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5763 getting "No route to host". Convert this to a timeout.
5765 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5768 6. Some OS don't have socklen_t but use size_t instead. This affects the
5769 fifth argument of getsockopt() amongst other things. This is now
5770 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5771 can be set for individual OS. I have set it for SunOS5, OSF1, and
5772 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5773 some earlier ones do not.
5775 7. Change 4.30/15 was not doing the test caselessly.
5777 8. The standard form for an IPv6 address literal was being rejected by address
5778 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5779 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5780 this, as well as the form without the "IPv6" on the front (but only when
5781 address literals are enabled, of course).
5783 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5785 10. Exim crashed if a message with an empty sender address specified by -f
5786 encountered a router with an errors_to setting. This could be provoked only
5787 by a command such as
5791 where an empty string was supplied; "<>" did not hit this bug.
5793 11. Installed PCRE release 4.5.
5795 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5796 remained set. It is now erased.
5798 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5799 times from message ids (which are base 36 rather than the normal 62).
5801 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5802 were being counted as actual protocol errors, and logged if the log
5803 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5804 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5805 and DATA following a set of rejected RCPTs do not count as protocol errors.
5806 In other words, Exim assumes they were pipelined, though this may not
5807 actually be the case. Of course, in all cases the client gets an
5808 appropriate error code.
5810 15. If a lookup fails in an ACL condition, a message about the failure may
5811 be available; it is used if testing the ACL cannot continue, because most
5812 such messages specify what the cause of the deferral is. However, some
5813 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5814 that caused an old message to be retained and used if a later statement
5815 caused a defer, replacing the real cause of the deferral.
5817 16. If an IP address had so many PTR records that the DNS lookup buffer
5818 was not large enough to hold them, Exim could crash while trying to process
5819 the truncated data. It now detects and logs this case.
5821 17. Further to 4.21/58, another change has been made: if (and only if) the
5822 first line of a message (the first header line) ends with CRLF, a bare LF
5823 in a subsequent header line has a space inserted after it, so as not to
5824 terminate the header.
5826 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5827 unnecessarily, used atoi() instead of strtol(), and didn't check the
5828 termination when getting file sizes from file names by regex.
5830 19. Completely re-implemented the support for maildirsize files, in the light
5831 of a number of problems with the previous contributed implementation
5832 (4.30/29). In particular:
5834 . If the quota is zero, the maildirsize file is maintained, but no quota is
5837 . If the maildir directory does not exist, it is created before any attempt
5838 to write a maildirsize file.
5840 . The quota value in the file is just a cache; if the quota is changed in
5841 the transport, the new value overrides.
5843 . A regular expression is available for excluding directories from the
5846 20. The autoreply transport checks the characters in options that define the
5847 message's headers; it allows continued headers, but it was checking with
5848 isspace() after an embedded newline instead of explicitly looking for a
5851 21. If all the "regular" hosts to which an address was routed had passed their
5852 expiry times, and had not reached their retry times, the address was
5853 bounced, even if fallback hosts were defined. Now Exim should go on to try
5856 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5857 equivalent code in the SMTP transport. Some hosts send humungous responses
5858 to HELO/EHLO, more than 1024 it seems.
5860 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5861 gives compiler warnings in some environments. I've now done it "properly",
5864 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5865 (because of problems with the built-in one) was declared to return uschar *
5866 instead of char *, causing compiler failure.
5868 25. Fixed a file descriptor leak when processing alias/forward files.
5870 26. Fixed a minor format string issue in dbfn.c.
5872 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5874 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5875 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5878 29. When a sender address is verified, it is cached, to save repeating the test
5879 when there is more than one recipient in a message. However, when the
5880 verification involves a callout, it is possible for different callout
5881 options to be set for different recipients. It is too complicated to keep
5882 track of this in the cache, so now Exim always runs a verification when a
5883 callout is required, relying on the callout cache for the optimization.
5884 The overhead is duplication of the address routing, but this should not be
5887 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5888 to be verified with callout=postmaster, and the main callout worked but the
5889 postmaster check failed, the verification correctly failed. However, if a
5890 subsequent RCPT command asked for sender verification *without* the
5891 postmaster check, incorrect caching caused this verification also to fail,
5894 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5895 it was not caching the DNS options (qualify_single, search_parents) that
5896 were used when the lookup failed. A subsequent lookup with different
5897 options therefore always gave the same answer, though there were cases
5898 where it should not have. (Example: a "domains = !$mx_any" option on a
5899 dnslookup router: the "domains" option is always processed without any
5900 widening, but the router might have qualify_single set.) Now Exim uses the
5901 cached value only when the same options are set.
5903 32. Added John Jetmore's "exipick" utility to the distribution.
5905 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5906 than a timeout (e.g. a certificate is required, and is not provided), an
5907 Exim server now closes the connection immediately. Previously it waited for
5908 the client to close - but if the client is SSL, it seems that they each
5909 wait for each other, leading to a delay before one of them times out.
5911 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5912 maintained 0.8.x compatibility because I don't think many are using it, and
5913 it is clearly obsolete.
5915 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5918 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5919 client certificate was expired. A simple patch fixes this, though I don't
5920 understand the full logic of why the verify callback is called multiple
5923 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5924 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5925 which causes problems with some clients (such as the Certicom SSL Plus
5926 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5927 disables the coutermeasure allowing Eudora to connect."
5929 38. Exim was not checking that a write() to a log file succeeded. This could
5930 lead to Bad Things if a log got too big, in particular if it hit a file
5931 size limit. Exim now panics and dies if it cannot write to a log file, just
5932 as it does if it cannot open a log file.
5934 39. Modified OS/Makefile-Linux so that it now contains
5936 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5938 The two -D definitions ensure that Exim is compiled with large file
5939 support, which makes it possible to handle log files that are bigger than
5942 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5943 instance) a domain was checked against a named list that involved a lookup,
5944 causing $domain_data to be set, then another domain was checked against the
5945 same list, then the first domain was re-checked, the value of $domain_data
5946 after the final check could be wrong. In particular, if the second check
5947 failed, it could be set empty. This bug probably also applied to
5950 41. The strip_trailing_dot option was not being applied to the address given
5951 with the -f command-line option.
5953 42. The code for reading a message's header from the spool was incrementing
5954 $received_count, but never initializing it. This meant that the value was
5955 incorrect (doubled) while delivering a message in the same process in which
5956 it was received. In the most common configuration of Exim, this never
5957 happens - a fresh exec is done - but it can happen when
5958 deliver_drop_privilege is set.
5960 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5961 it now includes up to 150 characters of the unexpected data in the log
5964 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5965 and building data strings. The size of both of these buffers was 10 000
5966 bytes - far larger than anybody would *ever* want, thought I. Needless to
5967 say, somebody hit the limit. I have increased the maximum line length to
5968 20 000 and the maximum data length of concatenated lines to 100 000. I have
5969 also fixed two bugs, because there was no checking on these buffers. Tsk,
5970 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5971 buffer is too small.
5973 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5974 lsearch lookups. Now it does.
5976 46. When parsing a route_list item in a manualroute router, a fixed-length
5977 buffer was used for the list of hosts. I made this 1024 bytes long,
5978 thinking that nobody would ever have a list of hosts that long. Wrong.
5979 Somebody had a whole pile of complicated expansion conditions, and the
5980 string was silently truncated, leading to an expansion error. It turns out
5981 that it is easier to change to an unlimited length (owing to other changes
5982 that have happened since this code was originally written) than to build
5983 structure for giving a limitation error. The length of the item that
5984 expands into the list of hosts is now unlimited.
5986 47. The lsearch lookup could not handle data where the length of text line was
5987 more than 4095 characters. Such lines were truncated, leading to shortened
5988 data being returned. It should now handle lines of any length.
5990 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5991 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5994 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5995 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5996 no longer happens in comment lines. A list of replacements is now placed
5997 at the head of all of the source files, except those whose only change is
5998 to replace PERL_COMMAND in the very first #! line.
6000 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6001 messages on the queue, with a bottom-up merge sort, using code contributed
6002 by Michael Haardt. This should make operations like -bp somewhat faster on
6003 large queues. It won't affect queue runners, except when queue_run_in_order
6006 51. Installed eximstats 1.31 in the distribution.
6008 52. Added support for SRV lookups to the dnslookup router.
6010 53. If an ACL referred to $message_body or $message_body_end, the value was not
6011 reset for any messages that followed in the same SMTP session.
6013 54. The store-handling optimization for building very long strings was not
6014 differentiating between the different store pools. I don't think this
6015 actually made any difference in practice, but I've tidied it.
6017 55. While running the routers to verify a sender address, $sender_address
6018 was still set to the sender address. This is wrong, because when routing to
6019 send a bounce to the sender, it would be empty. Therefore, I have changed
6020 it so that, while verifying a sender address, $sender_address is set to <>.
6021 (There is no change to what happens when verifying a recipient address.)
6023 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6024 target A or AAAA records (if not already returned) without resetting the
6025 qualify_single or search_parents options of the DNS resolver. These are
6026 inappropriate in this case because the targets of MX and SRV records must
6027 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6028 target that, when qualified, matched something in the local domain. These
6029 two options are now turned off when doing these lookups.
6031 57. It seems that at least some releases of Reiserfs (which does not have the
6032 concept of a fixed number of inodes) returns zero and not -1 for the
6033 number of available inodes. This interacted badly with check_spool_inodes,
6034 which assumed that -1 was the "no such thing" setting. What I have done is
6035 to check that the total number of inodes is greater than zero before doing
6036 the test of how many are available.
6038 58. When a "warn" ACL statement has a log_message modifier, the message is
6039 remembered, and not repeated. This is to avoid a lot of repetition when a
6040 message has many recipients that cause the same warning to be written.
6041 However, Exim was preserving the list of already written lines for an
6042 entire SMTP session, which doesn't seem right. The memory is now reset if a
6043 new message is started.
6045 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6046 debugging output unless log_rewrite was also set.
6048 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6049 of (int)(handle) when we know that handle contains (void *)(-1).
6051 61. The Exim daemon panic-logs an error return when it closes the incoming
6052 connection. However "connection reset by peer" seems to be common, and
6053 isn't really an error worthy of noting specially, so that particular error
6056 62. When Exim is trying to find all the local interfaces, it used to panic and
6057 die if the ioctl to get the interface flags failed. However, it seems that
6058 on at least one OS (Solaris 9) it is possible to have an interface that is
6059 included in the list of interfaces, but for which you get a failure error
6060 for this call. This happens when the interface is not "plumbed" into a
6061 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6062 failure of the "get flags" call assumes that the interface is down.
6064 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6065 makes life easier for people who are doing arithmetic on fields extracted
6066 from dates, where you often get leading zeros that should not be
6067 interpreted as octal.
6069 64. Added qualify_domain to the redirect router, to override the global
6072 65. If a pathologically long header line contained very many addresses (the
6073 report of this problem mentioned 10 000) and each of them was rewritten,
6074 Exim could use up a very large amount of memory. (It kept on making new
6075 copies of the header line as it rewrote, and never released the old ones.)
6076 At the expense of a bit more processing, the header rewriting function has
6077 been changed so that it no longer eats memory in this way.
6079 66. The generation of the Received: header has been moved from the time that a
6080 message starts to be received, to the time that it finishes. The timestamp
6081 in the Received: header should now be very close to that of the <= log
6082 line. There are two side-effects of this change:
6084 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6085 logged header lines no longer include the local Received: line, because
6086 it has not yet been created. The same applies to a copy of the message
6087 that is returned to a non-SMTP sender when a message is rejected.
6089 (b) When a filter file is tested using -bf, no additional Received: header
6090 is added to the test message. After some thought, I decided that this
6093 This change does not affect the value of $received_for. It is still set
6094 after address rewriting, but before local_scan() is called.
6096 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6098 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6099 gave an unhelpful panic error message, and a defer error. I have managed to
6100 change this behaviour so that it now rejects any supplied certificate,
6101 which seems right, as the list of acceptable certificates is empty.
6103 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6104 gave an unhelpful defer error. I have not managed to make this reject any
6105 supplied certificates, but the error message it gives is "no certificate
6106 supplied", which is not helpful.
6108 70. exigrep's output now also includes lines that are not associated with any
6109 message, but which match the given pattern. Implemented by a patch from
6110 Martin Sluka, which also tidied up the Perl a bit.
6112 71. Recipient callout verification, like sender verification, was using <> in
6113 the MAIL FROM command. This isn't really the right thing, since the actual
6114 sender may affect whether the remote host accepts the recipient or not. I
6115 have changed it to use the actual sender in the callout; this means that
6116 the cache record is now keyed on a recipient/sender pair, not just the
6117 recipient address. There doesn't seem to be a real danger of callout loops,
6118 since a callout by the remote host to check the sender would use <>.
6119 [SEE ABOVE: changed after hitting problems.]
6121 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6122 temporary errors. However, in the case of such a code being given after
6123 the end of a data transmission (i.e. after ".") Exim was failing to write
6124 a retry record for the message. (Yes, there was some broken host that was
6125 actually sending 8xx at this point.)
6127 73. An unknown lookup type in a host list could cause Exim to panic-die when
6128 the list was checked. (An example that provoked this was putting <; in the
6129 middle of a list instead of at the start.) If this happened during a DATA
6130 ACL check, a -D file could be left lying around. This kind of configuration
6131 error no longer causes Exim to die; instead it causes a defer error. The
6132 incident is still logged to the main and panic logs.
6134 74. Buglet left over from Exim 3 conversion. The message "too many messages
6135 in one connection" was written to the rejectlog but not the mainlog, except
6136 when address rewriting (yes!) was being logged.
6138 75. Added write_rejectlog option.
6140 76. When a system filter was run not as root (that is, when system_filter_user
6141 was set), the values of the $n variables were not being returned to the
6142 main process; thus, they were not subsequently available in the $sn
6145 77. Added +return_path_on_delivery log selector.
6147 78. A connection timeout was being treated differently from recipients deferred
6148 when testing hosts_max_try with a message that was older than the host's
6149 retry timeout. (The host should not be counted, thus allowing all hosts to
6150 be tried at least once before bouncing.) This may have been the cause of an
6151 occasionally reported bug whereby a message would remain on the queue
6152 longer than the retry timeout, but would be bounced if a delivery was
6153 forced. I say "may" because I never totally pinned down the problem;
6154 setting up timeout/retry tests is difficult. See also the next item.
6156 79. The ultimate address timeout was not being applied to errors that involved
6157 a combination of host plus message (for example, a timeout on a MAIL
6158 command). When an address resolved to a number of possible hosts, and they
6159 were not all tried for each delivery (e.g. because of hosts_max_try), a
6160 message could remain on the queue longer than the retry timeout.
6162 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6165 81. Fixed an obscure SMTP outgoing bug which required at least the following
6166 conditions: (a) there was another message waiting for the same server;
6167 (b) the server returned 5xx to all RCPT commands in the first message so
6168 that the message was not completed; (c) the server dropped the connection
6169 or gave a negative response to the RSET that Exim sends to abort the
6170 transaction. The observed case was a dropped connection after DATA that had
6171 been sent in pipelining mode. That is, the server had advertised PIPELINING
6172 but was not implementing it correctly. The effect of the bug was incorrect
6173 behaviour, such as trying another host, and this could lead to a crash.
6179 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6180 and daemon.c were passed as pointers to ints; they should have been
6181 pointers to socklen_t variables (which are typically unsigned ints).
6183 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6186 3. Fixed a really odd bug that affected only the testing scheme; patching a
6187 certain fixed string in the binary changed the value of another string that
6188 happened to be identical to the end of the original first string.
6190 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6191 name", it returns that address as the IP address. On some operating
6192 systems (e.g. Solaris), it also passes back the IP address string as the
6193 "host name". However, on others (e.g. Linux), it passes back an empty
6194 string. Exim wasn't checking for this, and was changing the host name to an
6195 empty string, assuming it had been canonicalized.
6197 5. Although rare, it is permitted to have more than one PTR record for a given
6198 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6199 all the names associated with an address, because they do in Solaris.
6200 However, it seems that they do not in Linux for data that comes from the
6201 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6202 I found this out when I moved to a new Linux workstation and tried to run
6203 the Exim test suite.
6205 To get round this problem I have changed the code so that it now does its
6206 own call to the DNS to look up PTR records when searching for a host name.
6207 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6208 addresses that are only in /etc/hosts are still found.
6210 This behaviour is, however, controlled by an option called host_lookup_
6211 order, which defaults to "bydns:byaddr". If people want to use the other
6212 order, or indeed, just use one or the other means of lookup, they can
6213 specify it in this variable.
6215 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6216 some operating systems, this comes back from gethostbyaddr() as an empty
6217 string, and this is what Exim used to test for. However, it seems that in
6218 other systems, "." is yielded. Exim now tests for this case too.
6220 7. The values of check_spool_space and check_log_space are now held internally
6221 as a number of kilobytes instead of an absolute number of bytes. If a
6222 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6223 kilobyte. This means that much larger values can be stored.
6225 8. Exim monitor: an attempt to get the action menu when not actually pointing
6226 at a message produces an empty menu entitled "No message selected". This
6227 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6228 no entries in it ("Shell widget menu has zero width and/or height"). So I
6229 have added a single, blank menu entry in this case.
6231 9. Added ${quote_local_part.
6233 10. MIME decoding is now applied to the contents of Subject: header lines when
6236 11. Now that a reference to $sender_host_address automatically causes a reverse
6237 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6238 host lookup before query-style lookups in lists that might use this
6239 variable. This has therefore been abolished, and the "net-" prefix is no
6240 longer necessary for query-style lookups.
6242 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6243 have been a typo for LFLAGS, so it has been changed.
6245 13. The install script calls Exim with "-C /dev/null" in order to find the
6246 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6247 to be output. However, since Exim outputs its version number before the
6248 error, it didn't break the script. It just looked ugly. I fixed this by
6249 always allowing "-C /dev/null" if the caller is root.
6251 14. Ignore overlarge ACL variable number when reading spool file - insurance
6252 against a later release with more variables having written the file.
6254 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6255 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6256 this, as well as the form without the "IPv6" on the front.
6258 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6259 OS/Makefile-Darwin file.
6261 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6262 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6264 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6265 variable. However, some OS use ints for inodes. I've added cast to long int
6266 to get rid of the compiler warning.
6268 19. I had forgotten to lock out "/../" in configuration file names when
6269 ALT_CONFIG_PREFIX was set.
6271 20. Routers used for verification do not need to specify transports. However,
6272 if such a router generated a host list, and callout was configured, Exim
6273 crashed, because it could not find a port number from the (non-existent)
6274 transport. It now assumes port 25 in this circumstance.
6276 21. Added the -t option to exigrep.
6278 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6279 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6280 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6281 them is set, LOOKUP_LSEARCH is forced.
6283 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6285 24. Added sender and host information to the "rejected by local_scan()" log
6286 line; previously there was no indication of these.
6288 25. Added .include_if_exists.
6290 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6291 Linux. It turns out that not all file systems support this. Apparently some
6292 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6293 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6294 not supported on the file descriptor, is now ignored when Exim is trying to
6295 sync a directory. This applies only to Linux.
6297 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6299 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6300 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6303 29. Added support for maildirsize files from supplied patch (modified a bit).
6305 30. The use of :fail: followed by an empty string could lead Exim to respond to
6306 sender verification failures with (e.g.):
6308 550 Verification failed for <xxx>
6309 550 Sender verify failed
6311 where the first response line was missing the '-' that indicates it is not
6312 the final line of the response.
6314 31. The loop for finding the name of the user that called Exim had a hardwired
6315 limit of 10; it now uses the value of finduser_retries, which is used for
6316 all other user lookups.
6318 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6321 33. Exim was neglecting to zero errno before one call of strtol() when
6322 expanding a string and expecting an integer value. On some systems this
6323 resulted in spurious "integer overflow" errors. Also, it was casting the
6324 result into an int without checking.
6326 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6327 did not work. The code looks as if it has *never* worked, though it appears
6328 to have been documented since at least release 1.62. I have made it work.
6330 35. The "timeout_DNS" error in retry rules, also documented since at least
6331 1.62, also never worked. As it isn't clear exactly what this means, and
6332 clearly it isn't a major issue, I have abolished the feature by treating it
6333 as "timeout", and writing a warning to the main and panic logs.
6335 36. The display of retry rules for -brt wasn't always showing the error code
6338 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6339 timeout_connect_A, timeout_connect_MX.
6341 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6342 to the empty sender.
6344 39. The daemon was not analysing the content of -oX till after it had closed
6345 stderr and disconnected from the controlling terminal. This meant that any
6346 syntax errors were only noted on the panic log, and the return code from
6347 the command was 0. By re-arranging the code a little, I've made the
6348 decoding happen first, so such errors now appear on stderr, and the return
6349 code is 1. However, the actual setting up of the sockets still happens in
6350 the disconnected process, so errors there are still only recorded on the
6353 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6354 connections (as happens on some IP stacks) was logged at start up time as
6355 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6356 it from "IPv6 and IPv4", which means that two separate sockets are being
6359 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6360 says whether AF_INET or AF_INET6 was passed as an argument.
6362 42. Exiwhat output was messed up when time zones were included in log
6365 43. Exiwhat now gives more information about the daemon's listening ports,
6366 and whether -tls-on-connect was used.
6368 44. The "port" option of the smtp transport is now expanded.
6370 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6371 silently ignored. Now an error message is written to the main and panic
6374 46. There's a new ACL modifier called "logwrite" which writes to a log file
6375 as soon as it is encountered.
6377 47. Added $local_user_uid and $local_user_gid at routing time.
6379 48. Exim crashed when trying to verify a sender address that was being
6382 49. Exim was recognizing only a space character after ".include". It now also
6383 recognizes a tab character.
6385 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6386 extracting the relevant information from the specification. The man page no
6387 longer contains scrambled data for the -d option, and I've added a section
6388 at the front about calling Exim under different names.
6390 51. Added "extra_headers" argument to the "mail" command in filter files.
6392 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6395 53. Installed eximstats 1.29.
6397 54. Added transport_filter_timeout as a generic transport option.
6399 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6400 Cc: header lines. This was required by RFC 822, but it not required by RFC
6403 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6404 incoming messages that did not have them. Now it does so only if the
6405 message originates locally, that is, if there is no associated remote host
6406 address. When Resent- header lines are present, this applies to the Resent-
6407 lines rather than the non-Resent- lines.
6409 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6410 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6412 58. Messages for configuration errors now include the name of the main
6413 configuration file - useful now that there may be more than one file in a
6414 list (.included file names were always shown).
6416 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6417 for those rare installations that do not start the daemon as root or run it
6418 setuid root. I've cut out the call to initgroups() if the daemon is not
6421 60. The Exim user and group can now be bound into the binary as text strings
6422 that are looked up at the start of Exim's processing.
6424 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6426 62. Added $mailstore_basename variable.
6428 63. Installed patch to sieve.c from Michael Haardt.
6430 64. When Exim failed to open the panic log after failing to open the main log,
6431 the original message it was trying to log was written to stderr and debug
6432 output, but if they were not available (the usual case in production), it
6433 was lost. Now it is written to syslog before the two lines that record the
6434 failures to open the logs.
6436 65. Users' Exim filters run in subprocesses under the user's uid. It is
6437 possible for a "deliver" command or an alias in a "personal" command to
6438 provoke an address rewrite. If logging of address rewriting is configured,
6439 this fails because the process is not running as root or exim. There may be
6440 a better way of dealing with this, but for the moment (because 4.30 needs
6441 to be released), I have disabled address rewrite logging when running a
6442 filter in a non-root, non-exim process.
6448 1. The buildconfig auxiliary program wasn't quoting the value set for
6449 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6450 not defined. This bug was masked in 4.22 by the effect that was fixed in
6453 2. Some messages that were rejected after a message id was allocated were
6454 shown as "incomplete" by exigrep. It no longer does this for messages that
6455 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6457 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6458 have allow_domain_literals set, the ID did not get logged in the <= line.
6459 Domain literals are now always recognized in Message-ID: header lines.
6461 4. The first argument for a ${extract expansion item is the key name or field
6462 number. Leading and trailing spaces in this item were not being ignored,
6463 causing some misleading effects.
6465 5. When deliver_drop_privilege was set, single queue runner processes started
6466 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6467 same command in the process it spins off) were not dropping privilege.
6469 6. When the daemon running as "exim" started a queue runner, it always
6470 re-executed Exim in the spun-off process. This is a waste of effort when
6471 deliver_drop_privilege is set. The new process now just calls the
6472 queue-runner function directly.
6478 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6481 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6482 ignored. Though the use of -oP was forcing the writing of a pid file, it
6483 was always written to the default place.
6485 3. If the message "no IP address found for host xxxx" is generated during
6486 incoming verification, it is now followed by identification of the incoming
6487 connection (so you can more easily find what provoked it).
6489 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6491 5. Added some features to "harden" Exim a bit more against certain attacks:
6493 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6494 be put in Local/Makefile. This is like the never_users runtime option,
6495 but it cannot be overridden. The default setting is "root".
6497 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6498 prefix string with which any file named in a -C command line option
6501 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6502 is retained for -C and -D only if the caller of Exim is root. Without
6503 it, the exim user may also use -C and -D and retain privilege.
6505 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6506 command line option is disabled.
6508 6. Macro names set by the -D option must start with an upper case letter, just
6509 like macro names defined in the configuration file.
6511 7. Added "dereference=" facility to LDAP.
6513 8. Two instances of the typo "uknown" in the source files are fixed.
6515 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6516 the Configure-Makefile script screwed up while processing it.
6518 10. Incorporated PCRE 4.4.
6520 11. The SMTP synchronization check was not operating right at the start of an
6521 SMTP session. For example, it could not catch a HELO sent before the client
6522 waited for the greeting. There is now a check for outstanding input at the
6523 point when the greeting is written. Because of the duplex, asynchronous
6524 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6525 way, but not yet received, when the check is performed.
6527 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6528 on TCP/IP sockets, because this apparently causes some broken clients to
6531 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6532 unchanged) from the Cygwin maintainer.
6534 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6535 was supported instead of testing for mbx. Effectively a typo.
6537 15. The spa authenticator server code was not checking that the input it
6538 received was valid base64.
6540 16. The debug output line for the "set" modifier in ACLs was not showing the
6541 name of the variable that was being set.
6543 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6545 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6546 lookup on its own. Something else was needed to trigger the lookup. For
6547 example, a match in host_lookup or the need for a host name in a host list.
6548 Now, if $sender_host_name is referenced and the host name has not yet been
6549 looked up, a lookup is performed. If the lookup fails, the variable remains
6550 empty, and $host_lookup_failed is set to "1".
6552 19. Added "eqi" as a case-independent comparison operator.
6554 20. The saslauthd authentication condition could segfault if neither service
6555 nor realm was specified.
6557 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6558 error message that was logged was misleading, and incoming SMTP
6559 connections were dropped. The message is now more accurate, and temporary
6560 errors are given to SMTP connections.
6562 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6563 (except RSET) with a 421 error, until QUIT is received. However, it was
6564 failing to send a response to QUIT.
6566 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6567 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6568 ACL is now run after the helo_try_verify_hosts code.
6570 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6571 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6572 case-independent, so other case variants are also recognized). Apparently
6573 some systems use these upper case variants.
6575 25. If more than two messages were waiting for the same host, and a transport
6576 filter was specified for the transport, Exim sent two messages over the
6577 same TCP/IP connection, and then failed with "socket operation on non-
6578 socket" when it tried to send the third.
6580 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6582 27. The extern definition of crypt16() in expand.c was not being excluded when
6583 the OS had its own crypt16() function.
6585 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6586 as a preferred synonym for return_size_limit, both as an option and as an
6589 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6591 30. Changed the default configuration ACL to relax the local part checking rule
6592 for addresses that are not in any local domains. For these addresses,
6593 slashes and pipe symbols are allowed within local parts, but the sequence
6594 /../ is explicitly forbidden.
6596 31. SPA server authentication was not clearing the challenge buffer before
6599 32. log_message in a "warn" ACL statement was writing to the reject log as
6600 well as to the main log, which contradicts the documentation and doesn't
6601 seem right (because no rejection is happening). So I have stopped it.
6603 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6604 However, I am unable to do any testing of this.
6606 34. Fixed an infelicity in the appendfile transport. When checking directories
6607 for a mailbox, to see if any needed to be created, it was accidentally
6608 using path names with one or more superfluous leading slashes; tracing
6609 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6611 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6612 contents are added to the log line that is written for every discarded
6613 recipient. (Previously a log_message setting was ignored.)
6615 36. The ${quote: operator now quotes the string if it is empty.
6617 37. The install script runs exim in order to find its version number. If for
6618 some reason other than non-existence or emptiness, which it checks, it
6619 could not run './exim', it was installing it with an empty version number,
6620 i.e. as "exim-". This error state is now caught, and the installation is
6623 38. An argument was missing from the function that creates an error message
6624 when Exim fails to connect to the socket for saslauthd authentication.
6625 This could cause Exim to crash, or give a corrupted message.
6627 39. Added isip, isip4, and isip6 to ${if conditions.
6629 40. The ACL variables $acl_xx are now saved with the message, and can be
6630 accessed later in routers, transports, and filters.
6632 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6633 strings in the file are not string-expanded.
6635 42. If a MAIL command specified a SIZE value that was too large to fit into an
6636 int variable, the check against message_size_limit failed. Such values are
6637 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6638 day this will have to be increased, but I don't think I want to be around
6639 when emails are that large.
6646 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6647 iconv() is not standard in FreeBSD.
6649 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6650 IPv6 enabled. The observed symptom was a segmentation fault on return from
6651 the function os_common_find_running_interfaces() in src/os.c.
6653 3. In the check_special_case() function in daemon.c I had used "errno" as an
6654 argument name, which causes warnings on some systems. This was basically a
6655 typo, since it was named "eno" in the comments!
6657 4. The code that waits for the clock to tick (at a resolution of some fraction
6658 of a second) so as to ensure message-id uniqueness was always waiting for
6659 at least one whole tick, when it could have waited for less. [This is
6660 almost certainly not relevant at current processor speeds, where it is
6661 unlikely to ever wait at all. But we try to future-proof.]
6663 5. The function that sleeps for a time interval that includes fractions of a
6664 second contained a race. It did not block SIGALRM between setting the
6665 timer, and suspending (a couple of lines later). If the interval was short
6666 and the sigsuspend() was delayed until after it had expired, the suspension
6667 never ended. On busy systems this could lead to processes getting stuck for
6670 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6671 process, before it forks any delivery processes. The open lookup caching
6672 mechanism meant that the open file or database connection was passed into
6673 the delivery process. The problem was that delivery processes always tidy
6674 up cached lookup data. This could cause a problem for the next delivery
6675 process started by the queue runner, because the external queue runner
6676 process does not know about the closure. So the next delivery process
6677 still has data in the lookup cache. In the case of a file lookup, there was
6678 no problem because closing a file descriptor in a subprocess doesn't affect
6679 the parent. However, if the lookup was caching a connection to a database,
6680 the connection was closed, and the second delivery process was likely to
6681 see errors such as "PGSQL: query failed: server closed the connection
6682 unexpectedly". The problem has been fixed by closing all cached lookups
6683 in a queue runner before running a delivery process.
6685 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6686 seem to have the "const" qualifier which it has on other OS. I've
6689 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6690 *of the same type* (client or server) with the same public name that an
6691 error should be diagnosed.
6693 9. When Exim looked up a host name for an IP address, but failed to find the
6694 original IP address when looking up the host name (a safety check), it
6695 output the message "<ip address> does not match any IP for NULL", which was
6696 confusing, to say the least. The bug was that the host name should have
6697 appeared instead of "NULL".
6699 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6700 user that is built into the binary, and the -C or -D options is used, root
6701 privilege is dropped before the configuration file is read. In addition,
6702 logging is switched to stderr instead of the normal log files. If the
6703 configuration then re-defines the Exim user, the unprivileged environment
6704 is probably not what is expected, so Exim logs a panic warning message (but
6707 However, if deliver_drop_privilege is set, the unprivileged state may well
6708 be exactly what is intended, so the warning has been cut out in that case,
6709 and Exim is allowed to try to write to its normal log files.
6715 1. smtp_return_error_details was not giving details for temporary sender
6716 or receiver verification errors.
6718 2. Diagnose a configuration error if two authenticators have the same public
6721 3. Exim used not to create the message log file for a message until the first
6722 delivery attempt. This could be confusing when incoming messages were held
6723 for policy or load reasons. The message log file is now created at the time
6724 the message is received, and an initial "Received" line is written to it.
6726 4. The automatically generated man page for command line options had a minor
6727 bug that caused no ill effects; however, a more serious problem was that
6728 the procedure for building the man page automatically didn't always
6729 operate. Consequently, release 4.20 contains an out-of-date version. This
6730 shouldn't happen again.
6732 5. When building Exim with embedded Perl support, the script that builds the
6733 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6734 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6736 6. The freeze_tell option was not being used for messages that were frozen on
6737 arrival, either by an ACL or by local_scan().
6739 7. Added the smtp_incomplete_transaction log selector.
6741 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6742 was accepting AUTH without a new EHLO.
6744 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6745 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6748 10. Logging of TCP/IP connections (when configured) now happens in the main
6749 daemon process instead of the child process, so that the TCP/IP connection
6750 count is more accurate (but it can never be perfect).
6752 11. The use of "drop" in a nested ACL was not being handled correctly in the
6753 outer ACL. Now, if condition failure induced by the nested "drop" causes
6754 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6755 or "require"), the connection is dropped.
6757 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6758 that yield "discard" can now be used with an "accept" or a "discard" verb,
6759 but an error is generated for any others (because I can't see a useful way
6760 to define what should happen).
6762 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6763 are now processed in the same way as lines in the Exim configuration file.
6764 In particular, continuation lines are supported.
6766 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6768 15. Added -ti meaning -t -i.
6770 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6771 domains, and warn by logging if others are found.
6773 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6774 returned from GIFCONF when Exim is trying to find the list of interfaces on
6775 a host. The code in os.c has been modified to copy each ifreq to an aligned
6776 structure in all cases.
6778 Also, in some cases, the returned ifreq's were being copied to a 'struct
6779 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6780 means the last couple of bytes of an IPv6 address could be chopped if the
6781 ifreq contained only a normal sockaddr (14 bytes storage).
6783 18. Named domain lists were not supported in the hosts_treat_as_local option.
6784 An entry such as +xxxx was not recognized, and was treated as a literal
6787 19. Ensure that header lines added by a DATA ACL are included in the reject log
6788 if the ACL subsequently rejects the message.
6790 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6791 MD5 (which is deprecated).
6793 21. When testing a filter file using -bf, Exim was writing a message when it
6794 took the sender from a "From " line in the message, but it was not doing so
6795 when it took $return_path from a Return-Path: header line. It now does.
6797 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6798 with a valid header line field name (a series of printing characters
6799 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6801 23. Changed "disc" in the source to "disk" to conform to the documentation and
6802 the book and for uniformity.
6804 24. Ignore Sendmail's -Ooption=value command line item.
6806 25. When execve() failed while trying to run a command in a pipe transport,
6807 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6808 could be confused with a return value of 69 from the command itself. This
6809 has been changed to 127, the value the shell returns if it is asked to run
6810 a non-existent command. The wording for the related log line suggests a
6811 non-existent command as the problem.
6813 26. If received_header_text expands to an empty string, do not add a Received:
6814 header line to the message. (Well, it adds a token one on the spool, but
6815 marks it "old" so that it doesn't get used or transmitted.)
6817 27. Installed eximstats 1.28 (addition of -nt option).
6819 28. There was no check for failure on the call to getsockname() in the daemon
6820 code. This can fail if there is a shortage of resources on the system, with
6821 ENOMEM, for example. A temporary error is now given on failure.
6823 29. Contrary to the C standard, it seems that in some environments, the
6824 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6825 program. Exim now does this explicitly; it affects the formatting of
6826 timestamps using strftime().
6828 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6829 complaints. I've now initialized all the variables, to avoid this.
6831 32. Header lines added by a system filter were not being "seen" during
6832 transport-time rewrites.
6834 33. The info_callback() function passed to OpenSSL is set up with type void
6835 (*)(SSL *, int, int), as described somewhere. However, when calling the
6836 function (actually a macro) that sets it up, the type void(*)() is
6837 expected. I've put in a cast to prevent warnings from picky compilers.
6839 34. If a DNS black list lookup found a CNAME record, but there were no A
6840 records associated with the domain it pointed at, Exim crashed.
6842 35. If a DNS black list lookup returned more than one A record, Exim ignored
6843 all but the first. It now scans all returned addresses if a particular IP
6844 value is being sought. In this situation, the contents of the
6845 $dnslist_value variable are a list of all the addresses, separated by a
6848 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6849 to accept a host name and all its aliases if the forward lookup for any of
6850 them yielded the IP address of the incoming connection. Now it accepts only
6851 those names whose forward lookup yields the correct IP address. Any other
6852 names are discarded. This closes a loophole whereby a rogue DNS
6853 administrator could create reverse DNS records to break through a
6854 wildcarded host restriction in an ACL.
6856 37. If a user filter or a system filter that ran in a subprocess used any of
6857 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6858 the wrong values were passed to the pipe command ($thisaddress had the
6859 value of $0, $0 had the value of $1, etc). This bug was introduced by
6860 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6862 38. Improved the line breaking for long SMTP error messages from ACLs.
6863 Previously, if there was no break point between 40 and 75 characters, Exim
6864 left the rest of the message alone. Two changes have been made: (a) I've
6865 reduced the minimum length to 35 characters; (b) if it can't find a break
6866 point between 35 and 75 characters, it looks ahead and uses the first one
6867 that it finds. This may give the occasional overlong line, but at least the
6868 remaining text gets split now.
6870 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6871 file descriptors might be low, and that setting 1000 would always raise it.
6872 It turns out that in some environments, the limit is already over 1000 and
6873 that lowering it causes trouble. So now Exim takes care not to decrease it.
6875 40. When delivering a message, the value of $return_path is set to $sender_
6876 address at the start of routing (routers may change the value). By an
6877 oversight, this default was not being set up when an address was tested by
6878 -bt or -bv, which affected the outcome if any router or filter referred to
6881 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6882 log only when "message" or "log_message" are set. However, if one of the
6883 conditions was an address verification, or a call to a nested ACL, the
6884 messages generated by the underlying test were being passed through. This
6885 no longer happens. The underlying message is available in $acl_verify_
6886 message for both "message" and "log_message" expansions, so it can be
6887 passed through if needed.
6889 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6890 new expansion $bh_ to give the encoded byte string without charset
6891 translation. Translation happens only if iconv() is available; HAVE_ICONV
6892 indicates this at build time. HEADERS_CHARSET gives the charset to
6893 translate to; headers_charset can change it in the configuration, and
6894 "headers charset" can change it in an individual filter file.
6896 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6897 that creates RFC 2047 encoded "words" labels them as that charset instead
6898 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6899 expansion operator; (ii) when Exim creates a From: line for a local
6900 message; (iii) when a header line is rewritten to include a "phrase" part.
6902 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6903 buggy, causing it to skip the first lines of messages whose message ID
6904 ended in 'D'. This would not have bitten before Exim release 4.14, because
6905 message IDs were unlikely to end in 'D' before then. The effect was to have
6906 incorrect size information for certain domains.
6908 45. #include "config.h" was missing at the start of the crypt16.c module. This
6909 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6912 46. If there was a timeout during a "random" callout check, Exim treated it as
6913 a failure of the random address, and carried on sending RSET and the real
6914 address. If the delay was just some slowness somewhere, the response to the
6915 original RCPT would be taken as a response to RSET and so on, causing
6916 mayhem of various kinds.
6918 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6919 when I implemented it. It didn't allow for the fact that some option values
6920 may legitimately be negative (e.g. size_addition), and it didn't even do
6921 the right test for positive values.
6923 48. Domain names in DNS records are case-independent. Exim always looks them up
6924 in lower case. Some resolvers return domain names in exactly the case they
6925 appear in the zone file, that is, they may contain uppercase letters. Not
6926 all resolvers do this - some return always lower case. Exim was treating a
6927 change of case by a resolver as a change of domain, similar to a widening
6928 of a domain abbreviation. This triggered its re-routing code and so it was
6929 trying to route what was effectively the same domain again. This normally
6930 caused routing to fail (because the router wouldn't handle the domain
6931 twice). Now Exim checks for this case specially, and just changes the
6932 casing of the domain that it ultimately uses when it transmits the message
6935 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6938 50. If a filter generated a file delivery with a non-absolute name (possible if
6939 no home directory exists for the router), the forbid_file option was not
6942 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6943 the existing equality matching.
6945 52. Exim was using ints instead of ino_t variables in some places where it was
6946 dealing with inode numbers.
6948 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6949 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6950 called TMPDIR, and if it finds it is different, it changes its value.
6952 54. The smtp_printf() function is now made available to local_scan() so
6953 additional output lines can be written before returning. There is also an
6954 smtp_fflush() function to enable the detection of a dropped connection.
6955 The variables smtp_input and smtp_batched_input are exported to
6958 55. Changed the default runtime configuration: the message "Unknown user"
6959 has been removed from the ACL, and instead placed on the localuser router,
6960 using the cannot_route_message feature. This means that any verification
6961 failures that generate their own messages won't get overridden. Similarly,
6962 the "Unrouteable address" message that was in the ACL for unverifiable
6963 relay addresses has also been removed.
6965 56. Added hosts_avoid_esmtp to the smtp transport.
6967 57. The exicyclog script was not checking for the esoteric option
6968 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6969 will work only if exicyclog is run under the appropriate euid.
6971 58. Following a discussion on the list, the rules by which Exim recognises line
6972 endings on incoming messages have been changed. The -dropcr and drop_cr
6973 options are now no-ops, retained only for backwards compatibility. The
6974 following line terminators are recognized: LF CRLF CR. However, special
6975 processing applies to CR:
6977 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6978 nor a local message in the state where . is a terminator.
6980 (ii) If a bare CR is encountered in a header line, an extra space is added
6981 after the line terminator so as not to end the header. The reasoning
6982 behind this is that bare CRs in header lines are most likely either
6983 to be mistakes, or people trying to play silly games.
6985 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6986 was being incorrectly given as 18 bytes larger than it should have been.
6987 This is a VOB (very old bug).
6989 60. This may never have affected anything current, but just in case it has:
6990 When the local host is found other than at the start of a list of hosts,
6991 the local host, those with the same MX, and any that follow, are discarded.
6992 When the list in question was part of a longer list of hosts, the following
6993 hosts (not currently being processed) were also being discarded. This no
6994 longer happens. I'm not sure if this situation could ever has previously
6997 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7000 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7001 header. This is something that is recommended in a new Internet Draft, and
7002 is something that is documented as being done by Sendmail. There are two
7003 possible values. For messages generated by the autoreply transport, Exim
7006 Auto-Submitted: auto-replied
7008 whereas for all other generated messages (e.g. bounces) it adds
7010 Auto-Submitted: auto-generated
7012 63. The "personal" condition in filters now includes a test for the
7013 Auto-Submitted: header. If it contains the string "auto-" the message it
7014 not considered personal.
7016 64. Added rcpt_include_affixes as a generic transport option.
7018 65. Added queue_only_override (default true).
7020 66. Added the syslog_duplication option.
7022 67. If what should have been the first header line of a message consisted of
7023 a space followed by a colon, Exim was mis-interpreting it as a header line.
7024 It isn't of course - it is syntactically invalid and should therefore be
7025 treated as the start of the message body. The misbehaviour could have
7026 caused a number of strange effects, including loss of data in subsequent
7027 header lines, and spool format errors.
7029 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7030 client host had authenticated. This control can now be exercised by an ACL
7031 for more flexibility.
7033 69. By default, callouts do not happen when testing with -bh. There is now a
7034 variant, -bhc, which does actually run the callout code, including
7035 consulting and updating the callout cache.
7037 70. Added support for saslauthd authentication, courtesy of Alexander
7040 71. If statvfs() failed on the spool or log directories while checking their
7041 size for availability, Exim confusingly gave the error "space shortage".
7042 Furthermore, in debugging mode it crashed with a floating point exception.
7043 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7044 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7045 serious problem, Exim now writes to the main and panic logs when this
7046 happens, with details of the failure. It then refuses to accept the
7047 incoming message, giving the message "spool directory problem" or "log
7048 directory problem" with a 421 code for SMTP messages.
7050 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7051 0, 1, and 2 exist, because some OS complain for execs without them (see
7052 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7053 descriptors. However, the code omitted to check that the open succeeded,
7054 causing mysterious errors if for some reason the permissions on /dev/null
7055 got screwed. Now Exim writes a message to the main and panic logs, and
7056 bombs out if it can't open /dev/null.
7058 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7059 interact so that it is all more flexible. It is supposed to remain
7060 backwards compatible. Also added extra_local_interfaces.
7062 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7063 to bomb out with an assertion failure - to the client this appears as a
7064 connection drop. This problem occurs in the part of the code that was taken
7065 from the Samba project. Fortunately, the assertion is in a very simple
7066 function, so I have fixed this by reproducing the function inline in the
7067 one place where it is called, and arranging for authentication to fail
7068 instead of killing the process with assert().
7070 75. The SPA client code was not working when the server requested OEM rather
7071 than Unicode encoding.
7073 76. Added code to make require_files with a specific uid setting more usable in
7074 the case where statting the file as root fails - usually a non-root-mounted
7075 NFS file system. When this happens and the failure is EACCES, Exim now
7076 forks a subprocess and does the per-uid checking as the relevant uid.
7078 77. Added process_log_path.
7080 78. If log_file_path was not explicitly set, a setting of check_log_space or
7081 check_log_inodes was ignored.
7083 79. If a space check for the spool or log partitions fails, the incident is now
7084 logged. Of course, in the latter case the data may get lost...
7086 80. Added the %p formatting code to string_format() so that it can be used to
7087 print addresses in debug_print(). Adjusted all the address printing in the
7088 debugging in store.c to use %p rather than %d.
7090 81. There was a concern that a line of code in smtp_in.c could overflow a
7091 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7092 initially expressed, the concern was not well-founded, because trailing
7093 spaces are removed early. However, if the trailing spaces were followed by
7094 a NULL, they did not get removed, so the overflow was possible. Two fixes
7097 (a) I re-wrote the offending code in a cleaner fashion.
7098 (b) If an incoming SMTP command contains a NULL character, it is rejected
7101 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7102 runs initgroups(), so that if the Exim user is in any additional groups,
7103 they will be used during message reception.
7109 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob