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 GF/02 Catch and remove uninitialized value warning in exiqsumm
19 Check for existence of @ARGV before looking at $ARGV[0]
21 JH/01 Replace the store_release() internal interface with store_newblock(),
22 which internalises the check required to safely use the old one, plus
23 the allocate and data copy operations duplicated in both (!) of the
26 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
27 modifier. This matches the restriction on the commandline.
29 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
30 Previously only the last row was returned.
32 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
33 we assumed that tags in the header were well-formed, and parsed the
34 element content after inspecting only the first char of the tag.
35 Assumptions at that stage could crash the receive process on malformed
38 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
39 While running the DKIM ACL we operate on the Permanent memory pool so that
40 variables created with "set" persist to the DATA ACL. Also (at any time)
41 DNS lookups that fail create cache records using the Permanent pool. But
42 expansions release any allocations made on the current pool - so a dnsdb
43 lookup expansion done in the DKIM ACL releases the memory used for the
44 DNS negative-cache, and bad things result. Solution is to switch to the
45 Main pool for expansions.
46 While we're in that code, add checks on the DNS cache during store_reset,
47 active in the testsuite.
48 Problem spotted, and debugging aided, by Wolfgang Breyha.
50 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
51 When none of the hosts presented to a transport match an already-open
52 connection, close it and proceed with the list. Previously we would
53 queue the message. Spotted by Lena with Yahoo, probably involving
56 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
57 Previously a spurious "250 OK id=" response was appended to the proper
60 JH/08 The "support for" informational output now, which built with Content
61 Scanning support, has a line for the malware scanner interfaces compiled
62 in. Interface can be individually included or not at build time.
64 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
65 by the template makefile "src/EDITME". The "STREAM" support for an older
66 ClamAV interface method is removed.
68 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
69 rows affected is given instead).
71 JH/11 The runtime Berkeley DB library version is now additionally output by
72 "exim -d -bV". Previously only the compile-time version was shown.
74 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
75 SMTP connection. Previously, when one had more receipients than the
76 first, an abortive onward connection was made. Move to full support for
77 multiple onward connections in sequence, handling cutthrough connection
78 for all multi-message initiating connections.
80 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
81 routers. Previously, a multi-recipient message would fail to match the
82 onward-connection opened for the first recipient, and cause its closure.
84 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
85 a timeout on read on a GnuTLS initiating connection, resulting in the
86 initiating connection being dropped. This mattered most when the callout
87 was marked defer_ok. Fix to keep the two timeout-detection methods
90 JH/15 Relax results from ACL control request to enable cutthrough, in
91 unsupported situations, from error to silently (except under debug)
92 ignoring. This covers use with PRDR, frozen messages, queue-only and
95 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
97 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
98 metadata, resulting in a crash in free().
100 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
101 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
102 Broken also in d185889f4, with init system revamp.
104 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
105 Previously we abruptly closed the connection after reading a malware-
106 found indication; now we go on to read the "scan ok" response line,
109 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
110 ACL. Previously, a crash would result.
112 JH/19 Speed up macro lookups during configuration file read, by skipping non-
113 macro text after a replacement (previously it was only once per line) and
114 by skipping builtin macros when searching for an uppercase lead character.
116 JH/20 DANE support moved from Experimental to mainline. The Makefile control
117 for the build is renamed.
119 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
120 was allocated for every new TLS startup, meaning one per message. Fix
121 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
123 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
124 reported the original. Fix to report (as far as possible) the ACL
125 result replacing the original.
127 JH/23 Fix memory leak during multi-message connections using STARTTLS under
128 OpenSSL. Certificate information is loaded for every new TLS startup,
129 and the resources needed to be freed.
131 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
133 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
136 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
137 DATA response info to the (existing) per-recipient response info for
138 the "C=" log element. It can have useful tracking info from the
139 destination system. Patch from Simon Arlott.
141 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
142 length value. Previously this would segfault.
144 HS/02 Support Avast multiline protoocol, this allows passing flags to
145 newer versions of the scanner.
147 JH/28 Ensure that variables possibly set during message acceptance are marked
148 dead before release of memory in the daemon loop. This stops complaints
149 about them when the debug_store option is enabled. Discovered specifically
150 for sender_rate_period, but applies to a whole set of variables.
151 Do the same for the queue-runner and queue-list loops, for variables set
152 from spool message files. Do the same for the SMTP per-message loop, for
153 certain variables indirectly set in ACL operations.
155 JH/29 Bug 2250: Fix a longstanding bug in heavily-pipelined SMTP input (such
156 as a multi-recipient message from a mailinglist manager). The coding had
157 an arbitrary cutoff number of characters while checking for more input;
158 enforced by writing a NUL into the buffer. This corrupted long / fast
159 input. The problem was exposed more widely when more pipelineing of SMTP
160 responses was introduced, and one Exim system was feeding another.
161 The symptom is log complaints of SMTP syntax error (NUL chars) on the
162 receiving system, and refused recipients seen by the sending system
163 (propating to people being dropped from mailing lists).
164 Discovered and pinpointed by David Carter.
166 JH/30 The (EXPERIMENTAL_DMARC) variable $dmarc_ar_header is withdrawn, being
167 replaced by the ${authresults } expansion.
169 JH/31 Bug 2257: Fix pipe transport to not use a socket-only syscall.
171 HS/03 Set a handler for SIGTERM and call exit(3) if running as PID 1. This
172 allows proper process termination in container environments.
174 JH/32 Bug 2258: Fix spool_wireformat in combination with LMTP transport.
175 Previously the "final dot" had a newline after it; ensure it is CR,LF.
177 JH/33 SPF: remove support for the "spf" ACL condition outcome values "err_temp"
178 and "err_perm", deprecated since 4.83 when the RFC-defined words
179 "temperror" and "permerror" were introduced.
181 JH/34 Re-introduce enforcement of no cutthrough delivery on transports having
182 transport-filters or DKIM-signing. The restriction was lost in the
183 consolidation of verify-callout and delivery SMTP handling.
189 JH/01 Rework error string handling in TLS interface so that the caller in
190 more cases is responsible for logging. This permits library-sourced
191 string to be attached to addresses during delivery, and collapses
192 pairs of long lines into single ones.
194 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
195 during configuration. Wildcards are allowed and expanded.
197 JH/02 Rework error string handling in DKIM to pass more info back to callers.
198 This permits better logging.
200 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
201 do not close it down and have the child transport start it up again on
202 the passed-on TCP connection. Instead, proxy the child (and any
203 subsequent ones) for TLS via a unix-domain socket channel. Logging is
204 affected: the continued delivery log lines do not have any DNSSEC, TLS
205 Certificate or OCSP information. TLS cipher information is still logged.
207 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
208 identical IP addresses on different listening ports. Will also affect
211 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
212 add noisy ifdef guards to special-case this sillyness.
213 Patch from Bernd Kuhls.
215 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
216 than 255 are no longer allowed.
218 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
219 Disable the session-cache too, which might reduce our load. Since we
220 currrectly use a new context for every connection, both as server and
221 client, there is no benefit for these.
222 GnuTLS appears to not support tickets server-side by default (we don't
223 call gnutls_session_ticket_enable_server()) but client side is enabled
224 by default on recent versions (3.1.3 +) unless the PFS priority string
227 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
228 <https://reproducible-builds.org/specs/source-date-epoch/>.
230 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
231 the check for any unsuccessful recipients did not notice the limit, and
232 erroneously found still-pending ones.
234 JH/08 Pipeline CHUNKING command and data together, on kernels that support
235 MSG_MORE. Only in-clear (not on TLS connections).
237 JH/09 Avoid using a temporary file during transport using dkim. Unless a
238 transport-filter is involved we can buffer the headers in memory for
239 creating the signature, and read the spool data file once for the
240 signature and again for transmission.
242 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
243 4.77 as the kernel support then wasn't solid, having issues in 64bit
244 mode. Now, it's been long enough. Add support for FreeBSD also.
246 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
247 case where the routing stage had gathered several addresses to send to
248 a host before calling the transport for the first, we previously failed
249 to close down TLS in the old transport process before passing the TCP
250 connection to the new process. The new one sent a STARTTLS command
251 which naturally failed, giving a failed delivery and bloating the retry
252 database. Investigation and fix prototype from Wolfgang Breyha.
254 JH/12 Fix check on SMTP command input synchronisation. Previously there were
255 false-negatives in the check that the sender had not preempted a response
256 or prompt from Exim (running as a server), due to that code's lack of
257 awareness of the SMTP input buffering.
259 PP/04 Add commandline_checks_require_admin option.
260 Exim drops privileges sanely, various checks such as -be aren't a
261 security problem, as long as you trust local users with access to their
262 own account. When invoked by services which pass untrusted data to
263 Exim, this might be an issue. Set this option in main configuration
264 AND make fixes to the calling application, such as using `--` to stop
267 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
268 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
269 responses to those, into a single TLS record each way (this usually means
270 a single packet). As a side issue, smtp_enforce_sync now works on TLS
273 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
274 affects you only if you're dancing at the edge of the param size limits.
275 If you are, and this message makes sense to you, then: raise the
276 configured limit or use OpenSSL 1.1. Nothing we can do for older
279 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
280 cmdline element to get the documented default one. Previously it was
283 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
284 are unsuable for non-message-specific reasons - eg. connection timeout,
285 banner-time rejection.
287 JH/16 Fix logging of delivery remote port, when specified by router, under
290 PP/06 Repair manualroute's ability to take options in any order, even if one
291 is the name of a transport.
294 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
296 JH/17 Change the list-building routines interface to use the expanding-string
297 triplet model, for better allocation and copying behaviour.
299 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
300 Previously it was constructed the first time a possibly-matching string
301 was met in the configuration file input during startup; now it is done
304 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
305 compatible one, to avoid the (poorly documented) possibility of a config
306 file in the working directory redirecting the DB files, possibly correpting
307 some existing file. CVE-2017-10140 assigned for BDB.
309 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
310 cache-hot. Previously, although the result was properly cached, the
311 initial verify call returned a defer.
313 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
314 the main verify for receipient in uncached-mode.
316 JH/22 Retire historical build files to an "unsupported" subdir. These are
317 defined as "ones for which we have no current evidence of testing".
319 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
320 if present. Previously it was ignored.
322 JH/24 Start using specified-initialisers in C structure init coding. This is
323 a C99 feature (it's 2017, so now considered safe).
325 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
326 if was a fixed-sized field and bitmask ops via macros; it is now more
329 PP/07 GitHub PR 56: Apply MariaDB build fix.
330 Patch provided by Jaroslav Škarvada.
332 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
333 during Coverity cleanups [4.87 JH/47]
334 Diagnosis and fix provided by Michael Fischer v. Mollard.
336 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
337 the right size to place the terminating semicolon on its own folded
338 line, the header hash was calculated to an incorrect value thanks to
339 the (relaxed) space the fold became.
341 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
342 and confused the parent.
344 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
345 which could crash as a result. This could lead to undeliverable messages.
347 JH/28 Logging: "next input sent too soon" now shows where input was truncated
350 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
351 matters on fast-turnover and PID-randomising systems, which were getting
352 out-of-order delivery.
354 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
355 a possibly-overlapping copy. The symptom was that "Remote host closed
356 connection in response to HELO" was logged instead of the actual 4xx
359 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
360 Previously only that bufferd was discarded, resulting in SYMTP command
363 JH/32 DKIM: when a message has multiple signatures matching an identity given
364 in dkim_verify_signers, run the dkim acl once for each. Previously only
365 one run was done. Bug 2189.
367 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
368 "panic the current process" to "deliberately defer". The panic log is
369 still written with the problem list name; the mail and reject logs now
370 get a temp-reject line for the message that was being handled, saying
371 something like "domains check lookup or other defer". The SMTP 451
372 message is still "Temporary local problem".
374 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
375 A crafted sequence of BDAT commands could result in in-use memory beeing
376 freed. CVE-2017-16943.
378 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
379 from SMTP input. Previously it was always done; now only done for DATA
380 and not BDAT commands. CVE-2017-16944.
382 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
383 to the message (such as an overlong header line). Previously this was
384 not done and we did not exit BDAT mode. Followon from the previous item
385 though a different problem.
391 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
392 than -2003 did; needs libidn2 in addition to libidn.
394 JH/02 The path option on a pipe transport is now expanded before use.
396 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
397 Patch provided by "Björn", documentation fix added too.
399 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
400 missing a wire-to-host endian conversion.
402 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
403 close after a BDAT command line could be taken as a following command,
404 giving a synch failure. Fix by only checking for synch immediately
405 before acknowledging the chunk.
407 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
408 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
409 macro. Patches provided by Josh Soref.
411 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
412 Previously we did not; the RFC seems ambiguous and VRFY is not listed
413 by IANA as a service extension. However, John Klensin suggests that we
416 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
417 the dkim code may be unix-mode line endings rather than smtp wire-format
418 CRLF, so prepend a CR to any bare LF.
420 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
421 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
423 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
424 the first were themselves being wrongly included in the feed into dkim
425 processing; with most chunk sizes in use this resulted in an incorrect
426 body hash calculated value.
428 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
429 DKIM signature block, for verification. Although advised against by
430 standards it is specifically not ruled illegal.
432 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
434 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
435 missing a body hash (the bh= tag).
437 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
438 It seems that HAProxy sends the Proxy Protocol information in clear and
439 only then does a TLS startup, so do the same.
441 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
442 TCP connections (such as for Spamd) unless the daemon successfully set
443 Fast Open mode on its listening sockets. This fixes breakage seen on
444 too-old kernels or those not configured for Fast Open, at the cost of
445 requiring both directions being enabled for TFO, and TFO never being used
446 by non-daemon-related Exim processes.
448 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
449 endings, at least on the first header line. Try to canonify any that get
450 past that check, despite the cost.
452 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
453 now limited to an arbitrary five deep, while parsing addresses with the
454 strip_excess_angle_brackets option enabled.
456 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
457 instead leave the unprompted TLS handshake in socket buffer for the
458 TLS library to consume.
460 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
462 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
464 JH/16 Drop variables when they go out of scope. Memory management drops a whole
465 region in one operation, for speed, and this leaves assigned pointers
466 dangling. Add checks run only under the testsuite which checks all
467 variables at a store-reset and panics on a dangling pointer; add code
468 explicitly nulling out all the variables discovered. Fixes one known
469 bug: a transport crash, where a dangling pointer for $sending_ip_address
470 originally assigned in a verify callout, is re-used.
472 PP/06 Drop '.' from @INC in various Perl scripts.
474 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
476 PP/08 Reduce a number of compilation warnings under clang; building with
477 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
478 should be warning-free.
480 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
482 HS/01 Fix portability problems introduced by PP/08 for platforms where
483 realloc(NULL) is not equivalent to malloc() [SunOS et al].
485 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
486 chunk. This allows us to accept broken chunked messages. We need a more
487 general solution here.
489 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
490 already-broken messages in the queue.
492 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
494 JH/19 Fix reference counting bug in routing-generated-address tracking.
500 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
501 supports it and a size is available (ie. the sending peer gave us one).
503 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
504 years of being deprecated). The replacements are the ACLs
505 acl_smtp_mime and acl_not_smtp_mime.
507 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
508 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
509 or even an in-clear connection were permitted. Now, if the host lookup
510 was dnssec and dane was requested then the host is only used if the
511 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
512 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
513 if one fails this test.
514 This means that a poorly-configured remote DNS will make it incommunicado;
515 but it protects against a DNS-interception attack on it.
517 JH/04 Bug 1810: make continued-use of an open smtp transport connection
518 non-noisy when a race steals the message being considered.
520 JH/05 If main configuration option tls_certificate is unset, generate a
521 self-signed certificate for inbound TLS connections.
523 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
524 in rewrites and routers.
526 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
527 and logged a warning sing 4.83; now they are a configuration file error.
529 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
530 (lacking @domain). Apply the same qualification processing as RCPT.
532 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
534 JH/10 Support ${sha256:} applied to a string (as well as the previous
537 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
538 a cutthrough deliver is pending, as we always want to make a connection.
539 This also avoids re-routing the message when later placing the cutthrough
540 connection after a verify cache hit.
541 Do not update it with the verify result either.
543 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
544 when routing results in more than one destination address.
546 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
547 signing (which inhibits the cutthrough capability). Previously only
548 the presence of an option was tested; now an expansion evaluating as
549 empty is permissible (obviously it should depend only on data available
550 when the cutthrough connection is made).
552 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
553 the relevant preceding SMTP command did not note the pipelining mode.
555 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
556 Previously they were not counted.
558 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
559 as one having no matching records. Previously we deferred the message
560 that needed the lookup.
562 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
563 distinguished as "(=".
565 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
566 for missing MX records. Previously it only worked for missing A records.
568 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
570 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
571 after the data-go-ahead and data-ack. Patch from Jason Betts.
573 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
574 even for a "none" policy. Patch from Tony Meyer.
576 JH/22 Fix continued use of a connection for further deliveries. If a port was
577 specified by a router, it must also match for the delivery to be
580 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
581 When one of the recipients of a message was unsuitable for the connection
582 (has no matching addresses), we lost track of needing to mark it
583 deferred. As a result mail would be lost.
585 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
587 JH/25 Decoding ACL controls is now done using a binary search; the source code
588 takes up less space and should be simpler to maintain. Merge the ACL
589 condition decode tables also, with similar effect.
591 JH/26 Fix problem with one_time used on a redirect router which returned the
592 parent address unchanged. A retry would see the parent address marked as
593 delivered, so not attempt the (identical) child. As a result mail would
596 JH/27 Fix a possible security hole, wherein a process operating with the Exim
597 UID can gain a root shell. Credit to http://www.halfdog.net/ for
598 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
601 JH/28 Enable {spool,log} filesystem space and inode checks as default.
602 Main config options check_{log,spool}_{inodes,space} are now
603 100 inodes, 10MB unless set otherwise in the configuration.
605 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
606 Previously it only applied to the main-section connection policy
609 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
611 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
612 by me. Added RFC7919 DH primes as an alternative.
614 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
615 are not in the system include path.
617 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
618 GnuTLS, when a session startup failed (eg because the client disconnected)
619 Exim did stdio operations after fclose. This was exposed by a recent
620 change which nulled out the file handle after the fclose.
622 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
623 signed directly by the cert-signing cert, rather than an intermediate
624 OCSP-signing cert. This is the model used by LetsEncrypt.
626 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
628 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
629 an incoming connection.
631 HS/02 Bug 1802: Do not half-close the connection after sending a request
634 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
635 fallback to "prime256v1".
637 JH/34 SECURITY: Use proper copy of DATA command in error message.
638 Could leak key material. Remotely exploitable. CVE-2016-9963.
644 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
645 and 3.4.4 - once the server is enabled to respond to an OCSP request
646 it does even when not requested, resulting in a stapling non-aware
647 client dropping the TLS connection.
649 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
650 support variable-length bit vectors. No functional change.
652 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
653 The I= interface field on outgoing lines is now after the H= remote
654 host field, same as incoming lines. There is a separate
655 outgoing_interface log selector which allows you to disable the
658 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
659 If not running log_selector +smtp_connection the mainlog would be held
660 open indefinitely after a "too many connections" event, including to a
661 deleted file after a log rotate. Leave the per net connection logging
662 leaving it open for efficiency as that will be quickly detected by the
663 check on the next write.
665 HS/01 Bug 1671: Fix post transport crash.
666 Processing the wait-<transport> messages could crash the delivery
667 process if the message IDs didn't exist for some reason. When
668 using 'split_spool_directory=yes' the construction of the spool
669 file name failed already, exposing the same netto behaviour.
671 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
672 mime_regex ACL conditions.
674 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
675 to DSN fail messages (bounces): remote IP, remote greeting, remote response
676 to HELO, local diagnostic string.
678 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
679 log line to debug. Even when configured with a tls authenticator many
680 client connections are expected to not authenticate in this way, so
681 an authenticate fail is not an error.
683 HS/02 Add the Exim version string to the process info. This way exiwhat
684 gives some more detail about the running daemon.
686 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
687 matter for fast-change records such as DNSBLs.
689 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
690 retry record, even if constant. There may be multiple transports with
691 different interface settings and the retry behaviour needs to be kept
694 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
696 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
698 JH/10 Bug 840: fix log_defer_output option of pipe transport
700 JH/11 Bug 830: use same host for all RCPTS of a message, even under
701 hosts_randomize. This matters a lot when combined with mua_wrapper.
703 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
704 ${quote_pgsql:<string>} operator.
706 JH/13 Bug 1708: avoid misaligned access in cached lookup.
708 JH/14 Change header file name for freeradius-client. Relevant if compiling
709 with Radius support; from the Gentoo tree and checked under Fedora.
711 JH/15 Bug 1712: Introduce $prdr_requested flag variable
713 JH/16 Bug 1714: Permit an empty string as expansion result for transport
714 option transport_filter, meaning no filtering.
716 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
718 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
719 defaults to "*" (all hosts). The variable is now available when not built
720 with TLS, default unset, mainly to enable keeping the testsuite sane.
721 If a server certificate is not supplied (via tls_certificate) an error is
722 logged, and clients will find TLS connections fail on startup. Presumably
723 they will retry in-clear.
724 Packagers of Exim are strongly encouraged to create a server certificate
725 at installation time.
727 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
728 with the $config_file variable.
730 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
731 in transport context, after the attempt, and per-recipient. The latter type
732 is per host attempted. The event data is the error message, and the errno
733 information encodes the lookup type (A vs. MX) used for the (first) host,
734 and the trailing two digits of the smtp 4xx response.
736 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
737 to write to mainlog (or rejectlog, paniclog) in the window between file
738 creation and permissions/ownership being changed. Particularly affects
739 installations where exicyclog is run as root, rather than exim user;
740 result is that the running daemon panics and dies.
742 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
744 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
745 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
746 "pri" and "weight". Note that the previous implicit priority given by the
747 list order is no longer honoured.
749 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
752 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
753 by defining SUPPORT_SOCKS.
755 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
756 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
757 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
758 variables are renamed to proxy_{local,external}_{address,port}.
760 JH/27 Move Internationalisation support from Experimental to mainline, enabled
761 for a build by defining SUPPORT_I18N
763 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
764 of the query string, and make ${quote_redis:} do that quoting.
766 JH/29 Move Events support from Experimental to mainline, enabled by default
767 and removable for a build by defining DISABLE_EVENT.
769 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
771 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
772 cached by the daemon.
774 JH/32 Move Redis support from Experimental to mainline, enabled for a build
775 by defining LOOKUP_REDIS. The libhiredis library is required.
777 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
778 keys are given for lookup.
780 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
781 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
782 only supported when built with TLS support. The PolarSSL SHA routines
783 are still used when the TLS library is too old for convenient support.
785 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
786 openssl_options), for security. OpenSSL forces this from version 1.1.0
787 server-side so match that on older versions.
789 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
790 allocation for $value could be released as the expansion processing
791 concluded, but leaving the global pointer active for it.
793 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
794 and to use the domains and local_parts ACL conditions.
796 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
797 incorrectly not doubled on cutthrough transmission, hence seen as a
798 body-termination at the receiving system - resulting in truncated mails.
799 Commonly the sender saw a TCP-level error, and retransmitted the message
800 via the normal store-and-forward channel. This could result in duplicates
801 received - but deduplicating mailstores were liable to retain only the
802 initial truncated version.
804 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
806 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
808 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
809 we're in there, support oversigning also; bug 1309.
811 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
813 HS/04 Add support for keep_environment and add_environment options.
815 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
816 either intentional arithmetic overflow during PRNG, or testing config-
819 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
820 delivery resulted in actual delivery. Cancel cutthrough before DATA
823 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
824 rejects a recipient: pass the reject to the originator.
826 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
827 Many were false-positives and ignorable, but it's worth fixing the
830 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
831 for the new environment-manipulation done at startup. Move the routines
832 from being local to tls.c to being global via the os.c file.
834 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
835 an extract embedded as result-arg for a map, the first arg for extract
836 is unavailable so we cannot tell if this is a numbered or keyed
837 extraction. Accept either.
843 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
846 JH/02 The smtp transport option "multi_domain" is now expanded.
848 JH/03 The smtp transport now requests PRDR by default, if the server offers
851 JH/04 Certificate name checking on server certificates, when exim is a client,
852 is now done by default. The transport option tls_verify_cert_hostnames
853 can be used to disable this per-host. The build option
854 EXPERIMENTAL_CERTNAMES is withdrawn.
856 JH/05 The value of the tls_verify_certificates smtp transport and main options
857 default to the word "system" to access the system default CA bundle.
858 For GnuTLS, only version 3.0.20 or later.
860 JH/06 Verification of the server certificate for a TLS connection is now tried
861 (but not required) by default. The verification status is now logged by
862 default, for both outbound TLS and client-certificate supplying inbound
865 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
868 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
869 Status Notification (bounce) messages are now MIME format per RFC 3464.
870 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
871 under the control of the dsn_advertise_hosts option, and routers may
872 have a dsn_lasthop option.
874 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
875 default, modifiable by a malware= option. The list separator for
876 the options can now be changed in the usual way. Bug 68.
878 JH/10 The smtp_receive_timeout main option is now expanded before use.
880 JH/11 The incoming_interface log option now also enables logging of the
881 local interface on delivery outgoing connections.
883 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
884 if the interface and destination host and port all match.
886 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
889 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
890 Patch from Andrew Lewis.
892 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
893 now supports optional time-restrictions, weighting, and priority
894 modifiers per server. Patch originally by <rommer@active.by>.
896 JH/16 The spamd_address main option now supports a mixed list of local
897 and remote servers. Remote servers can be IPv6 addresses, and
898 specify a port-range.
900 JH/17 Bug 68: The spamd_address main option now supports an optional
901 timeout value per server.
903 JH/18 Bug 1581: Router and transport options headers_add/remove can
904 now have the list separator specified.
906 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
909 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
912 JH/21 Support for the A6 type of dns record is withdrawn.
914 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
915 rather than the verbs used.
917 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
918 from 255 to 1024 chars.
920 JH/24 Verification callouts now attempt to use TLS by default.
922 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
923 are generic router options now. The defaults didn't change.
925 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
926 Original patch from Alexander Shikoff, worked over by JH.
928 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
929 files if ZCAT_COMMAND is not executable.
931 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
933 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
935 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
936 Normally benign, it bites when the pair was led to by a CNAME;
937 modern usage is to not canonicalize the domain to a CNAME target
938 (and we were inconsistent anyway for A-only vs AAAA+A).
940 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
942 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
943 when evaluating $sender_host_dnssec.
945 JH/31 Check the HELO verification lookup for DNSSEC, adding new
946 $sender_helo_dnssec variable.
948 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
950 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
952 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
954 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
955 documented as working, but never had. Support all but $spam_report.
957 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
958 added for tls authenticator.
960 HS/03 Add perl_taintmode main config option
966 TL/01 When running the test suite, the README says that variables such as
967 no_msglog_check are global and can be placed anywhere in a specific
968 test's script, however it was observed that placement needed to be near
969 the beginning for it to behave that way. Changed the runtest perl
970 script to read through the entire script once to detect and set these
971 variables, reset to the beginning of the script, and then run through
972 the script parsing/test process like normal.
974 TL/02 The BSD's have an arc4random API. One of the functions to induce
975 adding randomness was arc4random_stir(), but it has been removed in
976 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
977 function when detected.
979 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
980 cause callback expansion.
982 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
983 syntax errors in an expansion can be treated as a string instead of
984 logging or causing an error, due to the internal use of bool_lax
985 instead of bool when processing it.
987 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
988 server certificates when making smtp deliveries.
990 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
992 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
994 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
996 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
997 Merged patch from Sebastian Wiedenroth.
999 JH/05 Fix results-pipe from transport process. Several recipients, combined
1000 with certificate use, exposed issues where response data items split
1001 over buffer boundaries were not parsed properly. This eventually
1002 resulted in duplicates being sent. This issue only became common enough
1003 to notice due to the introduction of connection certificate information,
1004 the item size being so much larger. Found and fixed by Wolfgang Breyha.
1006 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
1007 size buffer was used, resulting in syntax errors when an expansion
1010 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1011 version 3.3.6 or later.
1013 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1014 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1015 both become "event_action", the variables become $event_name, $event_data
1016 and $event_defer_errno. There is a new variable $verify_mode, usable in
1017 routers, transports and related events. The tls:cert event is now also
1018 raised for inbound connections, if the main configuration event_action
1021 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1022 early OCSP support, but no stapling (appears to be less than 1.0.0).
1024 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1025 server certificate names available under the smtp transport option
1026 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1029 JH/10 Time-related extraction expansions from certificates now use the main
1030 option "timezone" setting for output formatting, and are consistent
1031 between OpenSSL and GnuTLS compilations. Bug 1541.
1033 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1034 encoded parameter in the incoming message. Bug 1558.
1036 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1037 include certificate info, eximon was claiming there were spoolfile
1040 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1042 JH/14 Log delivery-related information more consistently, using the sequence
1043 "H=<name> [<ip>]" wherever possible.
1045 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1046 are problematic for Debian distribution, omit them from the release
1049 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1051 JH/16 Fix string representation of time values on 64bit time_t architectures.
1054 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1055 output list separator was used.
1060 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1061 checkers that were complaining about end of non-void function with no
1064 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1065 This was a regression introduced in 4.83 by another bugfix.
1067 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1069 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1070 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1076 TF/01 Correctly close the server side of TLS when forking for delivery.
1078 When a message was received over SMTP with TLS, Exim failed to clear up
1079 the incoming connection properly after forking off the child process to
1080 deliver the message. In some situations the subsequent outgoing
1081 delivery connection happened to have the same fd number as the incoming
1082 connection previously had. Exim would try to use TLS and fail, logging
1083 a "Bad file descriptor" error.
1085 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1086 utilities have not been installed.
1088 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1089 temporary space as the ACL may create new global variables.
1091 TL/01 LDAP support uses per connection or global context settings, depending
1092 upon the detected version of the libraries at build time.
1094 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1095 to extract and use the src ip:port in logging and expansions as if it
1096 were a direct connection from the outside internet. PPv2 support was
1097 updated based on HAProxy spec change in May 2014.
1099 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1101 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1102 Properly escape header and check for NULL return.
1104 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1107 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1109 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1110 characters in header names, implemented as a verify condition.
1111 Contributed by Michael Fischer v. Mollard.
1113 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1114 results permerror and temperror. Previous values are deprecated but
1115 still accepted. In a future release, err_perm and err_temp will be
1116 completely removed, which will be a backward incompatibility if the
1117 ACL tests for either of these two old results. Patch contributed by
1118 user bes-internal on the mailing list.
1120 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1122 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1123 selectors, in both main and reject logs.
1125 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1128 JH/07 Add malware type "sock" for talking to simple daemon.
1130 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1132 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1133 routers/transports under cutthrough routing.
1135 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1136 numbers. Touch up "bool" conditional to keep the same definition.
1138 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1140 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1142 JH/12 Expand items in router/transport headers_add or headers_remove lists
1143 individually rather than the list as a whole. Bug 1452.
1145 Required for reasonable handling of multiple headers_ options when
1146 they may be empty; requires that headers_remove items with embedded
1147 colons must have them doubled (or the list-separator changed).
1149 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1150 view the policy declared in the DMARC record. Currently, $dmarc_status
1151 is a combined value of both the record presence and the result of the
1154 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1156 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1157 dnslookup router and the smtp transport (applying to the forward
1160 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1161 of ldap servers used for a specific lookup. Patch provided by Heiko
1164 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1165 New variable $lookup_dnssec_authenticated for observability.
1167 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1168 Patch submitted by Lars Timman.
1170 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1172 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1173 Requires trusted mode and valid format message id, aborts otherwise.
1174 Patch contributed by Heiko Schlichting.
1176 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1177 certextract with support for various fields. Bug 1358.
1179 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1180 is requested by default, modifiable by smtp transport option
1183 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1184 operate on certificate variables to give certificate fingerprints
1185 Also new ${sha256:cert_variable}.
1187 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1189 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1190 Christian Aistleitner.
1192 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1194 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1195 file. Patch from Wolfgang Breyha.
1197 JH/25 Expand the coverage of the delivery $host and $host_address to
1198 client authenticators run in verify callout. Bug 1476.
1200 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1201 align with daemon_smtp_ports. Bug 72.
1203 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1204 support and error reporting did not work properly.
1206 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1207 and is readable. Patch from Andrew Colin Kissa.
1209 TL/14 Enhance documentation of ${run expansion and how it parses the
1210 commandline after expansion, particularly in the case when an
1211 unquoted variable expansion results in an empty value.
1213 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1215 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1216 by renaming away from it. Addresses GH issue 15, reported by
1219 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1221 TL/15 SECURITY: prevent double expansion in math comparison functions
1222 (can expand unsanitized data). Not remotely exploitable.
1229 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1231 PP/02 Make -n do something, by making it not do something.
1232 When combined with -bP, the name of an option is not output.
1234 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1237 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1238 $sender_host_name and config options to manage this, and basic check
1241 PP/05 DSCP support for outbound connections and control modifier for inbound.
1243 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1244 (Only plugin which currently uses this is kerberos4, which nobody should
1245 be using, but we should make it available and other future plugins might
1246 conceivably use it, even though it would break NAT; stuff *should* be
1247 using channel bindings instead).
1249 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1250 name; added for Sendmail compatibility; requires admin caller.
1251 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1252 just ignore it); requires trusted caller.
1253 Also parse but ignore: -Ac -Am -X<logfile>
1256 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1258 TL/02 Add +smtp_confirmation as a default logging option.
1260 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1261 Patch by Magnus Holmgren from 2007-02-20.
1263 TL/04 Bugzilla 1281 - Spec typo.
1264 Bugzilla 1283 - Spec typo.
1265 Bugzilla 1290 - Spec grammar fixes.
1267 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1269 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1271 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1272 mailing list by by Dmitry Isaikin.
1274 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1276 JH/02 Support "G" suffix to numbers in ${if comparisons.
1278 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1280 NM/01 Bugzilla 1197 - Spec typo
1281 Bugzilla 1196 - Spec examples corrections
1283 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1285 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1286 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1289 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1290 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1292 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1293 "acl {{name}{arg}...}", and optional args on acl condition
1296 JH/05 Permit multiple router/transport headers_add/remove lines.
1298 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1300 JH/07 Avoid using a waiting database for a single-message-only transport.
1301 Performance patch from Paul Fisher. Bugzilla 1262.
1303 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1306 JH/09 Add $headers_added variable, with content from use of ACL modifier
1307 add_header (but not yet added to the message). Bugzilla 199.
1309 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1310 Pulled from Bugzilla 817 by Wolfgang Breyha.
1312 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1314 (nb: this is the same fix as in Exim 4.80.1)
1316 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1319 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1322 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1324 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1325 advertise SMTP AUTH mechanism to us, instead of a generic
1326 protocol violation error. Also, make Exim more robust to bad
1327 data from the Dovecot auth socket.
1329 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1331 When a queue runner is handling a message, Exim first routes the
1332 recipient addresses, during which it prunes them based on the retry
1333 hints database. After that it attempts to deliver the message to
1334 any remaining recipients. It then updates the hints database using
1337 So if a recipient address works intermittently, it can get repeatedly
1338 deferred at routing time. The retry hints record remains fresh so the
1339 address never reaches the final cutoff time.
1341 This is a fairly common occurrence when a user is bumping up against
1342 their storage quota. Exim had some logic in its local delivery code
1343 to deal with this. However it did not apply to per-recipient defers
1344 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1346 This change adds a proper retry rule check during routing so that the
1347 final cutoff time is checked against the message's age. We only do
1348 this check if there is an address retry record and there is not a
1349 domain retry record; this implies that previous attempts to handle
1350 the address had the retry_use_local_parts option turned on. We use
1351 this as an approximation for the destination being like a local
1352 delivery, as in LMTP.
1354 I suspect this new check makes the old local delivery cutoff check
1355 redundant, but I have not verified this so I left the code in place.
1357 TF/02 Correct gecos expansion when From: is a prefix of the username.
1359 Test 0254 submits a message to Exim with the header
1363 When I ran the test suite under the user fanf2, Exim expanded
1364 the header to contain my full name, whereas it should have added
1365 a Resent-Sender: header. It erroneously treats any prefix of the
1366 username as equal to the username.
1368 This change corrects that bug.
1370 GF/01 DCC debug and logging tidyup
1371 Error conditions log to paniclog rather than rejectlog.
1372 Debug lines prefixed by "DCC: " to remove any ambiguity.
1374 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1376 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1377 Bug spotted by Jeremy Harris; was flawed since initial commit.
1378 Would have resulted in OCSP responses post-SNI triggering an Exim
1379 NULL dereference and crash.
1381 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1383 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1384 Bug detection, analysis and fix by Samuel Thibault.
1385 Bugzilla 1331, Debian bug #698092.
1387 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1389 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1390 Server implementation by Todd Lyons, client by JH.
1391 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1392 config variable "prdr_enable" controls whether the server
1393 advertises the facility. If the client requests PRDR a new
1394 acl_data_smtp_prdr ACL is called once for each recipient, after
1395 the body content is received and before the acl_smtp_data ACL.
1396 The client is controlled by both of: a hosts_try_prdr option
1397 on the smtp transport, and the server advertisement.
1398 Default client logging of deliveries and rejections involving
1399 PRDR are flagged with the string "PRDR".
1401 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1402 fclose(). Diagnosis by Todd Lyons.
1404 PP/17 Update configure.default to handle IPv6 localhost better.
1405 Patch by Alain Williams (plus minor tweaks).
1408 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1409 This is now consistent with GnuTLS, and is now documented: the
1410 previous undocumented portable approach to treating the option as
1411 unset was to force an expansion failure. That still works, and
1412 an empty string is now equivalent.
1414 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1415 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1416 not performing validation itself.
1418 PP/20 Added force_command boolean option to pipe transport.
1419 Patch from Nick Koston, of cPanel Inc.
1421 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1424 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1426 PP/21 Fix eximon continuous updating with timestamped log-files.
1427 Broken in a format-string cleanup in 4.80, missed when I repaired the
1428 other false fix of the same issue.
1429 Report and fix from Heiko Schlichting.
1432 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1433 Report from Prashanth Katuri.
1435 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1436 It's SecureTransport, so affects any MacOS clients which use the
1437 system-integrated TLS libraries, including email clients.
1439 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1440 using a MIME ACL for non-SMTP local injection.
1441 Report and assistance in diagnosis by Warren Baker.
1443 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1445 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1447 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1448 last id that failed so it may be referenced in subsequent ACL's.
1450 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1453 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1454 ldap library initialization, allowing self-signed CA's to be
1455 used. Also properly sets require_cert option later in code by
1456 using NULL (global ldap config) instead of ldap handle (per
1457 session). Bug diagnosis and testing by alxgomz.
1459 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1460 the src/util/ subdirectory.
1462 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1463 renamed to Transport Post Delivery Action by Jeremy Harris, as
1466 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1467 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1468 redis_servers = needs to be configured which will be used by the redis
1469 lookup. Patch from Warren Baker, of The Packet Hub.
1471 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1473 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1474 hostname or reverse DNS when processing a host list. Used suggestions
1475 from multiple comments on this bug.
1477 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1479 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1480 Missed a few lines, added it to make the runtest require no keyboard
1483 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1484 contains upper case chars. Make router use caseful_local_part.
1486 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1487 support when GnuTLS has been built with p11-kit.
1493 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1495 This, or similar/improved, will also be change PP/11 of 4.82.
1501 PP/01 Handle short writes when writing local log-files.
1502 In practice, only affects FreeBSD (8 onwards).
1503 Bugzilla 1053, with thanks to Dmitry Isaikin.
1505 NM/01 Bugzilla 949 - Documentation tweak
1507 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1510 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1512 PP/02 Implemented gsasl authenticator.
1514 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1516 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1517 `pkg-config foo` for cflags/libs.
1519 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1520 with rest of GSASL and with heimdal_gssapi.
1522 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1523 `pkg-config foo` for cflags/libs for the TLS implementation.
1525 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1526 properties get this fed in as external SSF. A number of robustness
1527 and debugging improvements to the cyrus_sasl authenticator.
1529 PP/08 cyrus_sasl server now expands the server_realm option.
1531 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1532 Patch by Jeremy Harris.
1534 PP/10 Added dbmjz lookup type.
1536 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1538 PP/12 MAIL args handles TAB as well as SP, for better interop with
1539 non-compliant senders.
1540 Analysis and variant patch by Todd Lyons.
1542 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1543 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1544 Patch from Dirk Mueller <dmueller@suse.com>
1546 PP/13 tls_peerdn now print-escaped for spool files.
1547 Observed some $tls_peerdn in wild which contained \n, which resulted
1548 in spool file corruption.
1550 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1551 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1552 or write after TLS renegotiation, which otherwise led to messages
1555 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1556 as a tracking header (ie: a signed header comes before the signature).
1557 Patch from Wolfgang Breyha.
1559 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1560 comma-sep list; embedded commas doubled.
1562 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1564 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1566 Report and patch from Dmitry Banschikov.
1568 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1569 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1570 failures. We appear to now support TLS1.1+ with Exim.
1572 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1573 lets Exim select keys and certificates based upon TLS SNI from client.
1574 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1575 before an outbound SMTP session. New log_selector, +tls_sni.
1577 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1578 NULL dereference. Report and patch from Alun Jones.
1580 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1581 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1582 Not seeing resolver debug output on NetBSD, but suspect this is a
1583 resolver implementation change.
1585 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1586 Left warnings. Added "eximon gdb" invocation mode.
1588 PP/21 Defaulting "accept_8bitmime" to true, not false.
1590 PP/22 Added -bw for inetd wait mode support.
1592 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1593 locate the relevant includes and libraries. Made this the default.
1595 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1596 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1598 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1599 This may cause build issues on older platforms.
1601 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1602 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1603 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1604 Added SNI support via GnuTLS too.
1605 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1607 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1609 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1610 Applied second patch from Janne, implementing suggestion to default
1611 multiple-strings-in-record handling to match SPF spec.
1613 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1615 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1616 read-only, out of scope).
1617 Patch from Wolfgang Breyha, report from Stuart Northfield.
1619 PP/29 Fix three issues highlighted by clang analyser static analysis.
1620 Only crash-plausible issue would require the Cambridge-specific
1621 iplookup router and a misconfiguration.
1622 Report from Marcin Mirosław.
1624 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1626 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1627 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1628 As part of this, removing so much warning spew let me fix some minor
1629 real issues in debug logging.
1631 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1632 assignment on my part. Fixed.
1634 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1635 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1636 Janne Snabb (who went above and beyond: thank you).
1638 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1639 string otherwise requires a connection and a bunch more work and it's
1640 relatively easy to get wrong. Should also expose TLS library linkage
1643 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1644 64-bit ${eval} (JH/03).
1646 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1647 GNU libc to support some of the 64-bit stuff, should not lead to
1648 conflicts. Defined before os.h is pulled in, so if a given platform
1649 needs to override this, it can.
1651 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1652 protection layer was required, which is not implemented.
1653 Bugzilla 1254, patch from Wolfgang Breyha.
1655 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1656 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1657 tls_dhparam take prime identifiers. Also unbreak combination of
1658 OpenSSL+DH_params+TLSSNI.
1660 PP/39 Disable SSLv2 by default in OpenSSL support.
1666 PP/01 Solaris build fix for Oracle's LDAP libraries.
1667 Bugzilla 1109, patch from Stephen Usher.
1669 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1671 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1674 TF/02 Fix a couple more cases where we did not log the error message
1675 when unlink() failed. See also change 4.74-TF/03.
1677 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1678 lock up or crash if it happened to be inside a call to libc when it
1679 got a SIGUSR1 from exiwhat.
1681 The SIGUSR1 handler appends the current process status to the process
1682 log which is later printed by exiwhat. It used to use the general
1683 purpose logging code to do this, but several functions it calls are
1684 not safe for signals.
1686 The new output code in the SIGUSR1 handler is specific to the process
1687 log, and simple enough that it's easy to inspect for signal safety.
1688 Removing some special cases also simplifies the general logging code.
1689 Removing the spurious timestamps from the process log simplifies
1692 TF/04 Improved ratelimit ACL condition.
1694 The /noupdate option has been deprecated in favour of /readonly which
1695 has clearer semantics. The /leaky, /strict, and /readonly update modes
1696 are mutually exclusive. The update mode is no longer included in the
1697 database key; it just determines when the database is updated. (This
1698 means that when you upgrade Exim will forget old rate measurements.)
1700 Exim now checks that the per_* options are used with an update mode that
1701 makes sense for the current ACL. For example, when Exim is processing a
1702 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1703 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1704 must specify per_mail/readonly. If you omit the update mode it defaults to
1705 /leaky where that makes sense (as before) or /readonly where required.
1707 The /noupdate option is now undocumented but still supported for
1708 backwards compatibility. It is equivalent to /readonly except that in
1709 ACLs where /readonly is required you may specify /leaky/noupdate or
1710 /strict/noupdate which are treated the same as /readonly.
1712 A useful new feature is the /count= option. This is a generalization
1713 of the per_byte option, so that you can measure the throughput of other
1714 aggregate values. For example, the per_byte option is now equivalent
1715 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1717 The per_rcpt option has been generalized using the /count= mechanism
1718 (though it's more complicated than the per_byte equivalence). When it is
1719 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1720 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1721 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1722 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1723 Note that using per_rcpt with a non-readonly update mode in more than
1724 one ACL will cause the recipients to be double-counted. (The per_mail
1725 and per_byte options don't have this problem.)
1727 The handling of very low rates has changed slightly. If the computed rate
1728 is less than the event's count (usually one) then this event is the first
1729 after a long gap. In this case the rate is set to the same as this event's
1730 count, so that the first message of a spam run is counted properly.
1732 The major new feature is a mechanism for counting the rate of unique
1733 events. The new per_addr option counts the number of different
1734 recipients that someone has sent messages to in the last time period. It
1735 behaves like per_rcpt if all the recipient addresses are different, but
1736 duplicate recipient addresses do not increase the measured rate. Like
1737 the /count= option this is a general mechanism, so the per_addr option
1738 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1739 example, measure the rate that a client uses different sender addresses
1740 with the options per_mail/unique=$sender_address. There are further
1741 details in the main documentation.
1743 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1745 TF/06 Removed a few PCRE remnants.
1747 TF/07 Automatically extract Exim's version number from tags in the git
1748 repository when doing development or release builds.
1750 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1751 Bugzilla 879. Patch from Paul Fisher.
1753 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1754 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1757 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1759 PP/05 Variable $av_failed, true if the AV scanner deferred.
1760 Bugzilla 1078. Patch from John Horne.
1762 PP/06 Stop make process more reliably on build failure.
1763 Bugzilla 1087. Patch from Heiko Schlittermann.
1765 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1766 Bugzilla 1089. Patch from Heiko Schlittermann.
1768 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1769 Bugzilla 1131. Patch from Holger Weiß.
1771 PP/09 Handle IPv6 addresses with SPF.
1772 Bugzilla 860. Patch from Wolfgang Breyha.
1774 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1776 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1779 PP/11 match_* no longer expand right-hand-side by default.
1780 New compile-time build option, EXPAND_LISTMATCH_RHS.
1781 New expansion conditions, "inlist", "inlisti".
1783 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1785 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1787 PP/14 fix log_write() format string regression from TF/03.
1788 Bugzilla 1152. Patch from Dmitry Isaikin.
1794 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1796 PP/02 Harmonised TLS library version reporting; only show if debugging.
1797 Layout now matches that introduced for other libraries in 4.74 PP/03.
1799 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1801 PP/04 New "dns_use_edns0" global option.
1803 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1806 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1807 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1809 TK/01 Updated PolarSSL code to 0.14.2.
1810 Bugzilla 1097. Patch from Andreas Metzler.
1812 PP/07 Catch divide-by-zero in ${eval:...}.
1813 Fixes bugzilla 1102.
1815 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1818 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1819 format-string attack -- SECURITY: remote arbitrary code execution.
1821 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1822 time unintentionally subject to list matching rules, letting the header
1823 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1824 arbitrary string expansion). This allowed for information disclosure.
1826 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1827 INT_MIN/-1 -- value coerced to INT_MAX.
1833 NM/01 Workaround for PCRE version dependency in version reporting
1836 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1837 This fixes portability to compilers other than gcc, notably
1838 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1840 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1841 makefiles for portability to HP-UX and POSIX correctness.
1843 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1844 Also via indented variable definition in the Makefile.
1845 (Debugging by Oliver Heesakkers).
1847 PP/02 Restore caching of spamd results with expanded spamd_address.
1848 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1850 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1851 Improves build reliability. Fix from: Frank Elsner
1853 NM/02 Fix wide character breakage in the rfc2047 coding
1854 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1856 NM/03 Allow underscore in dnslist lookups
1857 Fixes bug 1026. Patch from Graeme Fowler
1859 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1860 Code patches from Adam Ciarcinski of NetBSD.
1862 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1865 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1866 is logged, to avoid truncation. Patch from John Horne.
1868 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1869 Patch from Jakob Hirsch.
1871 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1872 SQL string expansion failure details.
1873 Patch from Andrey Oktyabrski.
1875 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1876 Patch from Simon Arlott.
1878 PP/09 New lookups functionality failed to compile on old gcc which rejects
1879 extern declarations in function scope.
1880 Patch from Oliver Fleischmann
1882 PP/10 Use sig_atomic_t for flags set from signal handlers.
1883 Check getgroups() return and improve debugging.
1884 Fixed developed for diagnosis in bug 927 (which turned out to be
1887 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1888 Patch from Mark Zealey.
1890 PP/12 Bugzilla 1056: Improved spamd server selection.
1891 Patch from Mark Zealey.
1893 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1894 Based on patch from Heiko Schlittermann.
1896 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1897 Patch from Uwe Doering, sign-off by Michael Haardt.
1899 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1900 variable declaration deep within a block. Bug and patch from
1903 PP/15 lookups-Makefile IRIX compatibility coercion.
1905 PP/16 Make DISABLE_DKIM build knob functional.
1907 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1908 Patch by Simon Arlott
1910 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1911 variable argument macros. Our copy now differs from upstream.
1917 TF/01 Failure to get a lock on a hints database can have serious
1918 consequences so log it to the panic log.
1920 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1921 controlled using the smtp_confirmation log selector.
1923 TF/03 Include the error message when we fail to unlink a spool file.
1925 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1926 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1927 for maintaining out-of-tree patches for some time.
1929 PP/01 Bugzilla 139: Documentation and portability issues.
1930 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1931 Handle per-OS dynamic-module compilation flags.
1933 PP/02 Let /dev/null have normal permissions.
1934 The 4.73 fixes were a little too stringent and complained about the
1935 permissions on /dev/null. Exempt it from some checks.
1936 Reported by Andreas M. Kirchwitz.
1938 PP/03 Report version information for many libraries, including
1939 Exim version information for dynamically loaded libraries. Created
1940 version.h, now support a version extension string for distributors
1941 who patch heavily. Dynamic module ABI change.
1943 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1944 privilege escalation vulnerability whereby the Exim run-time user
1945 can cause root to append content of the attacker's choosing to
1948 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1951 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1952 If dropping privileges for untrusted macros, we disabled normal logging
1953 on the basis that it would fail; for the Exim run-time user, this is not
1954 the case, and it resulted in successful deliveries going unlogged.
1955 Fixed. Reported by Andreas Metzler.
1961 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1962 only prepend for the Resent-* case. Fixes regression introduced in
1963 Exim 4.70 by NM/22 for Bugzilla 607.
1965 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1966 increasing numbers of administrators be bitten by this.
1968 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1970 PP/03 Bugzilla 994: added openssl_options main configuration option.
1972 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1974 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1976 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1978 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1979 without a peer certificate, leading to a segfault because of an
1980 assumption that peers always have certificates. Be a little more
1981 paranoid. Problem reported by Martin Tscholak.
1983 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1984 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1985 NB: ClamAV planning to remove STREAM in "middle of 2010".
1986 CL also introduces -bmalware, various -d+acl logging additions and
1987 more caution in buffer sizes.
1989 PP/09 Implemented reverse_ip expansion operator.
1991 PP/10 Bugzilla 937: provide a "debug" ACL control.
1993 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1995 PP/12 Bugzilla 973: Implement --version.
1997 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1999 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
2001 PP/15 Bugzilla 816: support multiple condition rules on Routers.
2003 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
2004 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
2005 ignore trailing whitespace.
2007 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2009 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2012 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2013 Notification from Dr Andrew Aitchison.
2015 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2016 ExtendedDetectionInfo response format.
2017 Notification from John Horne.
2019 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2022 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2023 XSL and documented dependency on system catalogs, with examples of how
2026 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2029 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2030 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2031 configuration file which is writeable by the Exim user or group.
2033 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2034 of configuration files to cover files specified with the -C option if
2035 they are going to be used with root privileges, not just the default
2038 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2039 option (effectively making it always true).
2041 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2042 files to be used while preserving root privileges.
2044 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2045 that rogue child processes cannot use them.
2047 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2048 run-time user, instead of root.
2050 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2051 Exim run-time user without dropping privileges.
2053 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2054 result string, instead of calling string_vformat() twice with the same
2057 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2058 for other users. Others should always drop root privileges if they use
2059 -C on the command line, even for a whitelisted configure file.
2061 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2063 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2069 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2070 $data_path, and $header_path variables; fixed documentation bugs and
2073 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2074 exipick to access non-standard spools, including the "frozen" queue
2077 NM/01 Bugzilla 965: Support mysql stored procedures.
2078 Patch from Alain Williams
2080 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2082 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2083 Patch from Andreas Metzler
2085 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2086 Patch from Kirill Miazine
2088 NM/05 Bugzilla 671: Added umask to procmail example.
2090 JJ/03 installed exipick 20100323.0, fixing doc bug
2092 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2093 directory. Notification and patch from Dan Rosenberg.
2095 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2097 TK/02 Improve log output when DKIM signing operation fails.
2099 MH/01 Treat the transport option dkim_domain as a colon separated
2100 list, not as a single string, and sign the message with each element,
2101 omitting multiple occurences of the same signer.
2103 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2104 Bugzilla 985, 986. Patch by Simon Arlott
2106 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2107 Patch by Simon Arlott
2109 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2110 MBX locking. Notification from Dan Rosenberg.
2116 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2118 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2120 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2122 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2124 NM/04 Bugzilla 915: Flag broken perl installation during build.
2130 TK/01 Added patch by Johannes Berg that expands the main option
2131 "spamd_address" if it starts with a dollar sign.
2133 TK/02 Write list of recipients to X-Envelope-Sender header when building
2134 the mbox-format spool file for content scanning (suggested by Jakob
2137 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2138 (http://www.dcc-servers.net/) support via dccifd. Activated by
2139 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2141 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2142 by Mark Daniel Reidel <mr@df.eu>.
2144 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2145 When building exim an external PCRE library is now needed -
2146 PCRE is a system library on the majority of modern systems.
2147 See entry on PCRE_LIBS in EDITME file.
2149 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2150 conversation. Added nologin parameter to request.
2151 Patch contributed by Kirill Miazine.
2153 TF/01 Do not log submission mode rewrites if they do not change the address.
2155 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2157 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2158 log files in place. Contributed by Roberto Lima.
2160 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2162 TF/03 Bugzilla 615: When checking the local_parts router precondition
2163 after a local_part_suffix or local_part_prefix option, Exim now
2164 does not use the address's named list lookup cache, since this
2165 contains cached lookups for the whole local part.
2167 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2168 Robert Millan. Documentation is in experimental-spec.txt.
2170 TF/04 Bugzilla 668: Fix parallel build (make -j).
2172 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2174 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2175 Patch provided by Jan Srzednicki.
2177 TF/05 Leading white space used to be stripped from $spam_report which
2178 wrecked the formatting. Now it is preserved.
2180 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2181 that they are available at delivery time.
2183 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2185 TF/08 TLS error reporting now respects the incoming_interface and
2186 incoming_port log selectors.
2188 TF/09 Produce a more useful error message if an SMTP transport's hosts
2189 setting expands to an empty string.
2191 NM/06 Bugzilla 744: EXPN did not work under TLS.
2192 Patch provided by Phil Pennock.
2194 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2195 Patch provided by Richard Godbee.
2197 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2198 acl_smtp_notquit, added index entry.
2200 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2201 Patch provided by Eugene Bujak.
2203 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2204 accept(). Patch provided by Maxim Dounin.
2206 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2207 Patch provided by Phil Pennock.
2209 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2211 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2212 Patch provided by Brad "anomie" Jorsch.
2214 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2215 Patch provided by Dean Brooks.
2217 TK/05 Add native DKIM support (does not depend on external libraries).
2219 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2220 Patch provided by Graeme Fowler.
2222 NM/16 Bugzilla 851: Documentation example syntax fix.
2224 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2226 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2229 NM/19 Bugzilla 745: TLS version reporting.
2230 Patch provided by Phil Pennock.
2232 NM/20 Bugzilla 167: bool: condition support.
2233 Patch provided by Phil Pennock.
2235 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2236 clients. Patch provided by Phil Pennock.
2238 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2239 Patch provided by Brad "anomie" Jorsch.
2241 NM/23 Bugzilla 687: Fix misparses in eximstats.
2242 Patch provided by Heiko Schlittermann.
2244 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2245 Patch provided by Heiko Schlittermann.
2247 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2248 plus update to original patch.
2250 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2252 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2253 Patch provided by David Brownlee.
2255 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2257 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2259 NM/30 Bugzilla 888: TLS documentation bugfixes.
2261 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2263 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2264 Unlike the original bugzilla I have changed all shell scripts in src tree.
2266 NM/33 Bugzilla 898: Transport filter timeout fix.
2267 Patch by Todd Rinaldo.
2269 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2270 Patch by Serge Demonchaux.
2272 NM/35 Bugzilla 39: Base64 decode bug fixes.
2273 Patch by Jakob Hirsch.
2275 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2277 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2279 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2281 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2287 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2288 ALT-N's libdkim that I have put here:
2289 http://duncanthrax.net/exim-experimental/
2291 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2292 were called 'true' and it seems that C99 defines that as a reserved
2293 keyword to be used with 'bool' variable types. That means you could
2294 not include C99-style headers which use bools without triggering
2295 build errors in sieve.c.
2297 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2298 as mailq or other aliases. Changed the --help handling significantly
2299 to do whats expected. exim_usage() emits usage/help information.
2301 SC/01 Added the -bylocaldomain option to eximstats.
2303 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2305 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2307 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2313 PH/01 Another patch from the Sieve maintainer.
2315 PH/02 When an IPv6 address is converted to a string for single-key lookup
2316 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2317 dots are used instead of colons so that keys in lsearch files need not
2318 contain colons. This was done some time before quoting was made available
2319 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2320 (notated using the quote facility) so as to distinguish them from IPv4
2321 keys. This meant that lookups for IP addresses in host lists did not work
2322 for iplsearch lookups.
2324 This has been fixed by arranging for IPv6 addresses to be expressed with
2325 colons if the lookup type is iplsearch. This is not incompatible, because
2326 previously such lookups could never work.
2328 The situation is now rather anomalous, since one *can* have colons in
2329 ordinary lsearch keys. However, making the change in all cases is
2330 incompatible and would probably break a number of configurations.
2332 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2335 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2336 conversion specification without a maximum field width, thereby enabling
2337 a rogue spamd server to cause a buffer overflow. While nobody in their
2338 right mind would setup Exim to query an untrusted spamd server, an
2339 attacker that gains access to a server running spamd could potentially
2340 exploit this vulnerability to run arbitrary code as the Exim user.
2342 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2343 $primary_hostname instead of what libspf2 thinks the hosts name is.
2345 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2346 a directory entry by the name of the lookup key. Previously, if a
2347 symlink pointed to a non-existing file or a file in a directory that
2348 Exim lacked permissions to read, a lookup for a key matching that
2349 symlink would fail. Now it is enough that a matching directory entry
2350 exists, symlink or not. (Bugzilla 503.)
2352 PH/03 The body_linecount and body_zerocount variables are now exported in the
2355 PH/04 Added the $dnslist_matched variable.
2357 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2358 This means they are set thereafter only if the connection becomes
2361 PH/06 Added the client_condition to authenticators so that some can be skipped
2362 by clients under certain conditions.
2364 PH/07 The error message for a badly-placed control=no_multiline_responses left
2365 "_responses" off the end of the name.
2367 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2369 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2370 (without spaces) instead of just copying the configuration text.
2372 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2374 PH/11 Added $max_received_linelength.
2376 PH/12 Added +ignore_defer and +include_defer to host lists.
2378 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2379 way in which PCRE > 7.0 is built.
2381 PH/14 Implemented queue_only_load_latch.
2383 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2384 MAIL command. The effect was to mangle the value on 64-bit systems.
2386 PH/16 Another patch from the Sieve maintainer.
2388 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2390 PH/18 If a system quota error occurred while trying to create the file for
2391 a maildir delivery, the message "Mailbox is full" was not appended to the
2392 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2393 only to a quota excession during the actual writing of the file.
2395 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2396 characters?) which causes problems in log lines. The DN values are now
2397 passed through string_printing() before being added to log lines.
2399 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2400 and InterBase are left for another time.)
2402 PH/21 Added message_body_newlines option.
2404 PH/22 Guard against possible overflow in moan_check_errorcopy().
2406 PH/23 POSIX allows open() to be a macro; guard against that.
2408 PH/24 If the recipient of an error message contained an @ in the local part
2409 (suitably quoted, of course), incorrect values were put in $domain and
2410 $local_part during the evaluation of errors_copy.
2416 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2417 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2420 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2421 issue a MAIL command.
2423 PH/02 In an ACL statement such as
2425 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2427 if a client was not listed at all, or was listed with a value other than
2428 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2429 the condition was not true (as it should be), so access was not denied.
2430 The bug was that the ! inversion was incorrectly passed on to the second
2431 item. This has been fixed.
2433 PH/03 Added additional dnslists conditions == and =& which are different from
2434 = and & when the dns lookup returns more than one IP address.
2436 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2437 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2439 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2440 FSYNC, which compiles an option called disable_fsync that allows for
2441 bypassing fsync(). The documentation is heavily laced with warnings.
2443 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2445 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2446 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2447 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2448 including adding "make clean"; (3) Added -fPIC when compiling the test
2449 dynamically loaded module, to get rid of a warning.
2451 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2452 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2453 = 0s. The bug is otherwise harmless.
2455 PH/07 There was a bug in the dovecot authenticator such that the value of
2456 $auth1 could be overwritten, and so not correctly preserved, after a
2457 successful authentication. This usually meant that the value preserved by
2458 the server_setid option was incorrect.
2460 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2462 PH/09 Installed PCRE release 7.0.
2464 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2465 run for batched SMTP input. It is now run at the start of every message
2466 in the batch. While fixing this I discovered that the process information
2467 (output by running exiwhat) was not always getting set for -bs and -bS
2468 input. This is fixed, and it now also says "batched" for BSMTP.
2470 PH/11 Added control=no_pipelining.
2472 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2473 patch, slightly modified), and move the expansion of helo_data till after
2474 the connection is made in the smtp transport (so it can use these
2477 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2479 PH/14 Added log_selector = +pid.
2481 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2483 PH/16 Add ${if forany and ${if forall.
2485 PH/17 Added dsn_from option to vary the From: line in DSNs.
2487 PH/18 Flush SMTP output before performing a callout, unless control =
2488 no_callout_flush is set.
2490 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2491 was true (the default) a successful delivery failed to delete the retry
2492 item, thus causing premature timeout of the address. The bug is now
2495 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2497 PH/21 Long custom messages for fakedefer and fakereject are now split up
2498 into multiline responses in the same way that messages for "deny" and
2499 other ACL rejections are.
2501 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2502 with slight modification.
2504 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2505 draft, changing the syntax and factoring some duplicate code".
2507 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2508 for deliveries of the second and subsequent messages over the same SMTP
2511 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2512 ${reduce, with only minor "tidies".
2514 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2516 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2517 expansion side effects.
2519 PH/27 When a message times out after an over-quota error from an Exim-imposed
2520 quota, the bounce message says "mailbox is full". This message was not
2521 being given when it was a system quota that was exceeded. It now should
2524 MH/03 Made $recipients available in local_scan(). local_scan() already has
2525 better access to the recipient list through recipients_list[], but
2526 $recipients can be useful in postmaster-provided expansion strings.
2528 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2529 in the case of a MAIL command with additional options following the
2530 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2531 were accidentally chopped off.
2533 PH/29 SMTP synchronization checks are implemented when a command is read -
2534 there is a check that no more input is waiting when there shouldn't be
2535 any. However, for some commands, a delay in an ACL can mean that it is
2536 some time before the response is written. In this time, more input might
2537 arrive, invalidly. So now there are extra checks after an ACL has run for
2538 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2539 pipelining has not been advertised.
2541 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2543 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2544 returned in the SMTP response when smtp_return_error_details was false.
2545 This has been fixed.
2547 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2548 instead of the C I/O that was originally supplied, because problems were
2549 reported on Solaris.
2551 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2552 Exim which did not show up earlier: it was assuming that a call to
2553 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2554 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2555 was a macro that became an assignment, so it seemed to work. This has
2556 changed to a proper function call with a void return, hence the compile
2557 error. Exim's code has been fixed.
2559 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2562 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2564 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2566 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2567 because it turns out to be too big or there is a timeout) there is no
2568 "Completed" line in the log. When some messages of this type were
2569 selected by exigrep, they were listed as "not completed". Others were
2570 picked up by some special patterns. I have improved the selection
2571 criteria to be more general.
2573 PH/38 The host_find_failed option in the manualroute router can now be set
2574 to "ignore", to completely ignore a host whose IP address cannot be
2575 found. If all hosts are ignored, the behaviour is controlled by the new
2576 host_all_ignored option.
2578 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2579 homing or because of multiple MX records with /mx) generated more than
2580 one IP address, and the following item turned out to be the local host,
2581 all the secondary addresses of the first item were incorrectly removed
2582 from the list, along with the local host and any following hosts (which
2583 is what is supposed to happen).
2585 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2586 whoever called Exim into the -H file. In the case of the daemon it was
2587 behaving confusingly. When first started, it used values for whoever
2588 started the daemon, but after a SIGHUP it used the Exim user (because it
2589 calls itself on a restart). I have changed the code so that it now always
2592 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2593 message are rejected with the same error (e.g. no authentication or bad
2594 sender address), and a DATA command is nevertheless sent (as can happen
2595 with PIPELINING or a stupid MUA), the error message that was given to the
2596 RCPT commands is included in the rejection of the DATA command. This is
2597 intended to be helpful for MUAs that show only the final error to their
2600 PH/42 Another patch from the Sieve maintainer.
2602 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2603 Eximstats - Fixed some broken HTML links and added missing column headers
2605 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2606 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2608 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2610 PH/43 Yet another patch from the Sieve maintainer.
2612 PH/44 I found a way to check for a TCP/IP connection going away before sending
2613 the response to the final '.' that terminates a message, but only in the
2614 case where the client has not sent further data following the '.'
2615 (unfortunately, this is allowed). However, in many cases there won't be
2616 any further data because there won't be any more messages to send. A call
2617 to select() can be used: if it shows that the input is "ready", there is
2618 either input waiting, or the socket has been closed. An attempt to read
2619 the next input character can distinguish the two cases. Previously, Exim
2620 would have sent an OK response which the client would never have see.
2621 This could lead to message repetition. This fix should cure that, at
2622 least in a lot of common cases.
2624 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2625 advertised in response to EHLO.
2631 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2632 fixed by 4.65/MH/01 (is this a record?) are fixed:
2634 (i) An empty string was always treated as zero by the numeric comparison
2635 operators. This behaviour has been restored.
2637 (ii) It is documented that the numeric comparison operators always treat
2638 their arguments as decimal numbers. This was broken in that numbers
2639 starting with 0 were being interpreted as octal.
2641 While fixing these problems I realized that there was another issue that
2642 hadn't been noticed. Values of message_size_limit (both the global option
2643 and the transport option) were treated as octal if they started with 0.
2644 The documentation was vague. These values are now always treated as
2645 decimal, and I will make that clear in the documentation.
2651 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2652 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2655 MH/01 Don't check that the operands of numeric comparison operators are
2656 integers when their expansion is in "skipping" mode (fixes bug
2657 introduced by 4.64-PH/07).
2659 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2660 child addresses, Exim now panics and dies. Previously, because the count
2661 is held in a short int, deliveries were likely to be lost. As such a
2662 large number of recipients for a single message is ridiculous
2663 (performance will be very, very poor), I have chosen to impose a limit
2664 rather than extend the field.
2670 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2671 leftover -K file (the existence of which was triggered by #402).
2672 While we were at it, introduced process PID as part of the -K
2673 filename. This should rule out race conditions when creating
2676 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2677 processing considerably. Previous code took too long for large mails,
2678 triggering a timeout which in turn triggers #401.
2680 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2681 in the DK code in transports.c. sendfile() is not really portable,
2682 hence the _LINUX specificness.
2684 TF/01 In the add_headers option to the mail command in an Exim filter,
2685 there was a bug that Exim would claim a syntax error in any
2686 header after the first one which had an odd number of characters
2689 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2690 callout verification, Exim cached a "reject" for the entire domain. This
2691 is correct for most verifications, but it is not correct for a recipient
2692 verification with use_sender or use_postmaster set, because in that case
2693 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2694 case of MAIL FROM:<> rejection from other early rejections (e.g.
2695 rejection of HELO). When verifying a recipient using a non-null MAIL
2696 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2697 Whatever the result of the callout, the value of the domain cache is
2698 left unchanged (for any other kind of callout, getting as far as trying
2699 RCPT means that the domain itself is ok).
2701 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2704 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2705 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2706 session, the daemon ignores ECONNECT errors and logs others; it now
2707 ignores EPIPE as well.
2709 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2710 (quoted-printable decoding).
2712 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2713 later the small subsequent patch to fix an introduced bug.
2715 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2717 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2719 PH/08 An error is now given if message_size_limit is specified negative.
2721 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2722 to be given (somewhat) arbitrary names.
2724 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2727 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2728 miscellaneous code fixes
2730 PH/10 Added the log_reject_target ACL modifier to specify where to log
2733 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2734 hostname. This is wrong, because it relates to the incoming message (and
2735 probably the interface on which it is arriving) and not to the outgoing
2736 callout (which could be using a different interface). This has been
2737 changed to use the value of the helo_data option from the smtp transport
2738 instead - this is what is used when a message is actually being sent. If
2739 there is no remote transport (possible with a router that sets up host
2740 addresses), $smtp_active_hostname is used.
2742 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2743 tweaks were necessary in order to get it to work (see also 21 below):
2744 (a) The code assumed that strncpy() returns a negative number on buffer
2745 overflow, which isn't the case. Replaced with Exim's string_format()
2747 (b) There were several signed/unsigned issues. I just did the minimum
2748 hacking in of casts. There is scope for a larger refactoring.
2749 (c) The code used strcasecmp() which is not a standard C function.
2750 Replaced with Exim's strcmpic() function.
2751 (d) The code set only $1; it now sets $auth1 as well.
2752 (e) A simple test gave the error "authentication client didn't specify
2753 service in request". It would seem that Dovecot has changed its
2754 interface. Fortunately there's a specification; I followed it and
2755 changed what the client sends and it appears to be working now.
2757 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2760 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2761 address A is aliased to B and C, where B exists and C does not. Without
2762 -v the output is "A verified" because verification stops after a
2763 successful redirection if more than one address is generated. However,
2764 with -v the child addresses are also verified. Exim was outputting "A
2765 failed to verify" and then showing the successful verification for C,
2766 with its parentage. It now outputs "B failed to verify", showing B's
2767 parentage before showing the successful verification of C.
2769 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2770 look up a TXT record in a specific list after matching in a combined
2773 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2774 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2775 they consult the DNS. I had assumed they would set it the way they
2776 wanted; and indeed my experiments on Linux seem to show that in some
2777 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2778 To be on the safe side, however, I have now made the interface to
2779 host_find_byname() similar to host_find_bydns(), with an argument
2780 containing the DNS resolver options. The host_find_byname() function now
2781 sets these options at its start, just as host_find_bydns() does. The smtp
2782 transport options dns_qualify_single and dns_search_parents are passed to
2783 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2784 of host_find_byname() use the default settings of RES_DEFNAMES
2785 (qualify_single) but not RES_DNSRCH (search_parents).
2787 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2788 spool_read_header() do less string testing, by means of a preliminary
2789 switch on the second character of optional "-foo" lines. (This is
2790 overdue, caused by the large number of possibilities that now exist.
2791 Originally there were few.) While I was there, I also converted the
2792 str(n)cmp tests so they don't re-test the leading "-" and the first
2793 character, in the hope this might squeeze out yet more improvement.
2795 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2796 flag allowing group syntax was set by the header_syntax check but not
2797 turned off, possible causing trouble later; (2) The flag was not being
2798 set at all for the header_verify test, causing "group"-style headers to
2799 be rejected. I have now set it in this case, and also caused header_
2800 verify to ignore an empty address taken from a group. While doing this, I
2801 came across some other cases where the code for allowing group syntax
2802 while scanning a header line wasn't quite right (mostly, not resetting
2803 the flag correctly in the right place). These bugs could have caused
2804 trouble for malformed header lines. I hope it is now all correct.
2806 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2807 with the "reply" argument non-NULL. The code, however (which originally
2808 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2809 but it didn't always do it. This confused somebody who was copying the
2810 code for some other use. I have removed all the tests.
2812 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2813 feature that was used to support insecure browsers during the U.S. crypto
2814 embargo. It requires special client support, and Exim is probably the
2815 only MTA that supported it -- and would never use it because real RSA is
2816 always available. This code has been removed, because it had the bad
2817 effect of slowing Exim down by computing (never used) parameters for the
2818 RSA_EXPORT functionality.
2820 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2821 authenticator to fail if there's a tab character in the incoming data
2822 (there should never be unless someone is messing about, as it's supposed
2823 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2824 if the connection is using TLS or if the remote IP is the same as the
2825 local IP, and the "valid-client-cert option" if a client certificate has
2828 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2829 authenticators. This can be used for authorization after authentication
2830 succeeds. (In the case of plaintext, it servers for both authentication
2833 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2834 if any retry times were supplied.
2836 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2837 connection, where there is no client IP address to check. In this
2838 situation, the verify now always succeeds.
2840 PH/25 Applied John Jetmore's -Mset patch.
2842 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2844 PH/27 In a string expansion for a processed (not raw) header when multiple
2845 headers of the same name were present, leading whitespace was being
2846 removed from all of them, but trailing whitespace was being removed only
2847 from the last one. Now trailing whitespace is removed from each header
2848 before concatenation. Completely empty headers in a concatenation (as
2849 before) are ignored.
2851 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2852 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2854 PH/29 [Removed. This was a change that I later backed out, and forgot to
2855 correct the ChangeLog entry (that I had efficiently created) before
2856 committing the later change.]
2858 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2859 address errors (4xx response to RCPT) over the same connection as other
2860 messages routed to the same hosts. Such deliveries are always "forced",
2861 so retry times are not inspected. This resulted in far too many retries
2862 for the affected addresses. The effect occurred only when there were more
2863 hosts than the hosts_max_try setting in the smtp transport when it had
2864 the 4xx errors. Those hosts that it had tried were not added to the list
2865 of hosts for which the message was waiting, so if all were tried, there
2866 was no problem. Two fixes have been applied:
2868 (i) If there are any address or message errors in an SMTP delivery, none
2869 of the hosts (tried or untried) are now added to the list of hosts
2870 for which the message is waiting, so the message should not be a
2871 candidate for sending over the same connection that was used for a
2872 successful delivery of some other message. This seems entirely
2873 reasonable: after all the message is NOT "waiting for some host".
2874 This is so "obvious" that I'm not sure why it wasn't done
2875 previously. Hope I haven't missed anything, but it can't do any
2876 harm, as the worst effect is to miss an optimization.
2878 (ii) If, despite (i), such a delivery is accidentally attempted, the
2879 routing retry time is respected, so at least it doesn't keep
2880 hammering the server.
2882 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2883 in ${readsocket because some servers need this prod.
2885 PH/32 Added some extra debug output when updating a wait-xxx database.
2887 PH/33 The hint "could be header name not terminated by colon", which has been
2888 given for certain expansion errors for a long time, was not being given
2889 for the ${if def:h_colon_omitted{... case.
2891 PH/34 The spec says: "With one important exception, whenever a domain list is
2892 being scanned, $domain contains the subject domain." There was at least
2893 one case where this was not true.
2895 PH/35 The error "getsockname() failed: connection reset by peer" was being
2896 written to the panic log as well as the main log, but it isn't really
2897 panic-worthy as it just means the connection died rather early on. I have
2898 removed the panic log writing for the ECONNRESET error when getsockname()
2901 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2902 runs only) independently of the message's sender address. This meant
2903 that, if the 4xx error was in fact related to the sender, a different
2904 message to the same recipient with a different sender could confuse
2905 things. In particular, this can happen when sending to a greylisting
2906 server, but other circumstances could also provoke similar problems.
2907 I have changed the default so that the retry time for these errors is now
2908 based a combination of the sender and recipient addresses. This change
2909 can be overridden by setting address_retry_include_sender=false in the
2912 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2913 remote server are returned as part of bounce messages. This was not
2914 happening for LMTP over a pipe (the lmtp transport), but now it is the
2915 same for both kinds of LMTP.
2917 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2918 in header lines that were in fact CNAMEs. This is no longer the case.
2920 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2921 and queue runs started by the daemon processed all messages. This has
2922 been fixed so that -R and -S can now usefully be given with -q<time>.
2924 PH/40 Import PCRE release 6.7 (fixes some bugs).
2926 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2928 PH/42 Give an error if -q is specified more than once.
2930 PH/43 Renamed the variables $interface_address and $interface_port as
2931 $received_ip_address and $received_port, to make it clear that these
2932 values apply to message reception, and not to the outgoing interface when
2933 a message is delivered. (The old names remain recognized, of course.)
2935 PH/44 There was no timeout on the connect() call when using a Unix domain
2936 socket in the ${readsocket expansion. There now is.
2938 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2939 be meaningful with "accept".
2941 SC/01 Eximstats V1.43
2942 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2944 SC/02 Eximstats V1.44
2945 Use a glob alias rather than an array ref in the generated
2946 parser. This improves both readability and performance.
2948 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2949 Collect SpamAssassin and rejection statistics.
2950 Don't display local sender or destination tables unless
2951 there is data to show.
2952 Added average volumes into the top table text output.
2954 SC/04 Eximstats V1.46
2955 Collect data on the number of addresses (recipients)
2956 as well as the number of messages.
2958 SC/05 Eximstats V1.47
2959 Added 'Message too big' to the list of mail rejection
2960 reasons (thanks to Marco Gaiarin).
2962 SC/06 Eximstats V1.48
2963 Mainlog lines which have GMT offsets and are too short to
2964 have a flag are now skipped.
2966 SC/07 Eximstats V1.49 (Alain Williams)
2967 Added the -emptyok flag.
2969 SC/08 Eximstats V1.50
2970 Fixes for obtaining the IP address from reject messages.
2972 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2973 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2974 whitespace changes from 4.64-PH/27
2976 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2979 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2980 are found, allow negative numbers in numeric criteria)
2982 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2984 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2985 to match changes made in 4.64-PH/43
2987 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2989 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2990 "domain" is actually an IP address, and force a failure. This locks out
2991 those revolvers/nameservers that support "A-for-A" lookups, in
2992 contravention of the specifications.
2994 PH/48 When a host name was looked up from an IP address, and the subsequent
2995 forward lookup of the name timed out, the host name was left in
2996 $sender_host_name, contrary to the specification.
2998 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2999 restricted to single-key lookups, Exim was not diagnosing an error if
3000 * or *@ was used with a query-style lookup.
3002 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
3004 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
3005 long ago, but noone interested enough thought of it. Let's just say that
3006 the "1.1" means that there are some new functions that weren't there at
3007 some point in the past.
3009 PH/51 Error processing for expansion failure of helo_data from an smtp
3010 transport during callout processing was broken.
3012 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3013 tested/used via the -bh/-bhc/-bs options.
3015 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3016 bug, fixed in subsequent PCRE releases).
3018 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3019 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3021 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3027 SC/01 Use a glob alias rather than an array ref in eximstats generated
3028 parser. This improves both readability and performance.
3030 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3031 Don't display local sender or destination tables in eximstats unless
3032 there is data to show.
3033 Added average volumes into the eximstats top table text output.
3035 SC/03 Collect data on the number of addresses (recipients) as well
3036 as the number of messages in eximstats.
3038 TF/01 Correct an error in the documentation for the redirect router. Exim
3039 does (usually) call initgroups() when daemonizing.
3041 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3042 with consistent privilege compared to when running as a daemon.
3044 TF/03 Note in the spec that $authenticated_id is not set for local
3045 submissions from trusted users.
3047 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3048 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3050 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3051 by adding some example configuration directives to the default
3052 configuration file. A little bit of work is required to uncomment the
3053 directives and define how usernames and passwords are checked, but
3054 there is now a framework to start from.
3056 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3057 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3058 without this. I don't know how relevant this is to other LDAP libraries.
3060 PH/02 Add the verb name to the "unknown ACL verb" error.
3062 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3064 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3066 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3067 directory not expanded when it should be if an expanded home directory
3068 was set for the address (which is overridden by the transport).
3070 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3073 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3074 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3075 because it is too late at that time, and has no effect.
3077 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3078 security issue with \' (bugzilla #107). I could not use the
3079 PQescapeStringConn() function, because it needs a PGconn value as one of
3082 PH/08 When testing addresses using -bt, indicate those final addresses that
3083 are duplicates that would not cause an additional delivery. At least one
3084 person was confused, thinking that -bt output corresponded to deliveries.
3085 (Suppressing duplicates isn't a good idea as you lose the information
3086 about possibly different redirections that led to the duplicates.)
3088 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3089 systems where poll() doesn't work, in particular OS X.
3091 PH/10 Added more information to debugging output for retry time not reached.
3093 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3094 operations in malware.c.
3096 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3099 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3100 syslog_duplication set false, log lines that would normally be written
3101 both the the main log and to the reject log were not written to syslog at
3104 PH/14 In the default configuration, change the use of "message" in ACL warn
3105 statements to "add_header".
3107 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3108 not followed by a command (e.g. "seen endif").
3110 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3111 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3114 PH/17 Added extra conditions to the default value of delay_warning_condition
3118 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3119 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3120 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3123 The Auto-Submitted: and various List- headers are standardised, whereas I
3124 don't think Precedence: ever was.
3126 PH/18 Refactored debugging code in route_finduser() to show more information,
3127 in particular, the error code if getpwnam() issues one.
3129 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3130 This is apparently needed in addition to the PH/07 change above to avoid
3131 any possible encoding problems.
3133 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3134 but not after initializing Perl.
3136 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3137 output them only if debugging. By default they are written stderr,
3138 apparently, which is not desirable.
3140 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3143 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3146 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3148 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3149 authenticated or an ident call has been made. Suppress the default
3150 values for $authenticated_id and $authenticated_sender (but permit -oMai
3151 and -oMas) when testing with -bh.
3153 PH/24 Re-jigged the order of the tests in the default configuration so that the
3154 tests for valid domains and recipients precede the DNS black list and CSA
3155 tests, on the grounds that those ones are more expensive.
3157 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3158 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3159 This bug exists in every version of Exim that I still have, right back to
3162 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3163 However, an attempt to turn on case-sensitivity in a regex key by
3164 including (?-i) didn't work because the subject string was already
3165 lowercased, and the effects were non-intuitive. It turns out that a
3166 one-line patch can be used to allow (?-i) to work as expected.
3172 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3173 other effects) broke the use of negated acl sub-conditions.
3175 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3178 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3179 "Deny" causes Exim to reject the incoming connection with a 554 error.
3180 Unfortunately, if there is a major crisis, such as a disk failure,
3181 tcp-wrappers gives "deny", whereas what one would like would be some
3182 kind of temporary error. A kludge has been added to help with this.
3183 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3184 554 error is used if errno is still zero or contains ENOENT (which occurs
3185 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3188 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3190 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3191 errors. Otherwise a message that provokes a temporary error (when other
3192 messages do not) can cause a whole host to time out.
3194 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3195 addresses were routed directly to files or pipes from a redirect router.
3196 File deliveries just didn't batch; pipe deliveries might have suffered
3199 PH/06 A failure to get a lock for a hints database would erroneously always say
3200 "Failed to get write lock", even when it was really a read lock.
3202 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3203 of 0600. This has been changed to use the value of the lockfile_mode
3204 option (which defaults to 0600).
3206 PH/08 Applied small patch from the Sieve maintainer.
3208 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3209 folder from quota calculations, a direct delivery into this folder messed
3210 up the contents of the maildirsize file. This was because the regex was
3211 used only to exclude .Trash (or whatever) when the size of the mailbox
3212 was calculated. There was no check that a delivery was happening into an
3213 excluded directory. This bug has been fixed by ignoring all quota
3214 processing for deliveries into excluded directories.
3216 PH/10 Added the maildirfolder_create_regex option to appendfile.
3222 PH/01 The code for finding all the local interface addresses on a FreeBSD
3223 system running IPv6 was broken. This may well have applied to all BSD
3224 systems, as well as to others that have similar system calls. The broken
3225 code found IPv4 interfaces correctly, but gave incorrect values for the
3226 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3227 that it would not match correctly against @[] and not recognize the IPv6
3230 PH/02 The ipliteral router was not recognizing addresses of the form user@
3231 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3233 PH/03 Added disable_ipv6.
3235 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3236 decoded form, because it is most often used to construct To: headers
3237 lines in autoreplies, and the decoded form may well be syntactically
3238 invalid. However, $reply_address has leading white space removed, and all
3239 newlines turned into spaces so that the autoreply transport does not
3242 PH/05 If group was specified without a user on a router, and no group or user
3243 was specified on a transport, the group from the router was ignored.
3245 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3246 for visible compile-time settings that can be used to change these
3247 numbers, for those that want even more. Backwards compatibility with old
3248 spool files has been maintained. However, going back to a previous Exim
3249 release will lost any variables that are in spool files.
3251 PH/07 Two small changes when running in the test harness: increase delay when
3252 passing a TCP/IP connection to a new process, in case the original
3253 process has to generate a bounce, and remove special handling of
3254 127.0.0.2 (sic), which is no longer necessary.
3256 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3257 be the same on different OS.
3259 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3262 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3263 whether --show-vars was specified or not
3265 JJ/02 exipick: Added support for new ACL variable spool format introduced
3268 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3269 syntactically invalid From: or Reply-to: line, and a filter used this to
3270 generate an autoreply, and therefore failed to obtain an address for the
3271 autoreply, Exim could try to deliver to a non-existent relative file
3272 name, causing unrelated and misleading errors. What now happens is that
3273 it logs this as a hard delivery error, but does not attempt to create a
3276 PH/11 The exinext utility has a -C option for testing purposes, but although
3277 the given file was scanned by exinext itself; it wasn't being passed on
3278 when Exim was called.
3280 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3281 an end-of-file indication when reading a command response.
3283 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3284 compiled. In many other places in Exim, IPv6 addresses are always
3285 recognized, so I have changed this. It also means that IPv4 domain
3286 literals of the form [IPV4:n.n.n.n] are now always recognized.
3288 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3289 used if the router is not running as root, for example, when verifying at
3290 ACL time, or when using -bh. The debugging output from this situation was
3291 non-existent - all you got was a failure to exec. I have made two
3294 (a) Failures to set uid/gid, the current directory, or a process leader
3295 in a subprocess such as that created by queryprogram now generate
3296 suitable debugging output when -d is set.
3298 (b) The queryprogram router detects when it is not running as root,
3299 outputs suitable debugging information if -d is set, and then runs
3300 the subprocess without attempting to change uid/gid.
3302 PH/15 Minor change to Makefile for building test_host (undocumented testing
3305 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3306 additional section of a DNS packet that returns MX or SRV records.
3307 Instead, it always explicitly searches for A/AAAA records. This avoids
3308 major problems that occur when a DNS server includes only records of one
3309 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3310 fixed another bug: if SRV records were looked up and the corresponding
3311 address records were *not* found in the additional section, the port
3312 values from the SRV records were lost.
3314 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3315 using the correct key (the original address) when searching the retry
3316 rules in order to find which one to use for generating the retry hint.
3318 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3319 adding the default one. Similarly, if it contains a Reply-To: header, the
3320 errors_reply_to option, if set, is not used.
3322 PH/19 When calculating a retry time, Exim used to measure the "time since
3323 failure" by looking at the "first failed" field in the retry record. Now
3324 it does not use this if it is later than than the arrival time of the
3325 message. Instead it uses the arrival time. This makes for better
3326 behaviour in cases where some deliveries succeed, thus re-setting the
3327 "first failed" field. An example is a quota failure for a huge message
3328 when small messages continue to be delivered. Without this change, the
3329 "time since failure" will always be short, possible causing more frequent
3330 delivery attempts for the huge message than are intended.
3331 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3333 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3334 $1, $2, $3) because the numerical variables can be reset during some
3335 expansion items (e.g. "match"), thereby losing the authentication data.
3337 PH/21 Make -bV show the size of off_t variables so that the test suite can
3338 decide whether to run tests for quotas > 2G.
3340 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3341 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3342 filecount value is greater than 2G or if a quota value is greater than 2G
3343 on a system where the size of off_t is not greater than 4, a panic error
3346 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3347 never match. The debug and -bh output now contains an explicit error
3348 message indicating a malformed IPv4 address or mask.
3350 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3351 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3352 PH/23 above applies.
3354 PH/25 Do not write to syslog when running in the test harness. The only
3355 occasion when this arises is a failure to open the main or panic logs
3356 (for which there is an explicit test).
3358 PH/26 Added the /no_tell option to "control=freeze".
3360 PH/27 If a host name lookup failed very early in a connection, for example, if
3361 the IP address matched host_lookup and the reverse lookup yielded a name
3362 that did not have a forward lookup, an error message of the form "no IP
3363 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3364 could be logged. Now it outputs the IP address instead of "NULL".
3366 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3367 allows the sender and the authenticated sender to be set when
3368 submitting a message from within Exim. Since child_open_exim() is
3369 documented for local_scan(), the new function should be too.
3371 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3372 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3373 results in an empty string is now treated as unset.
3375 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3377 PH/31 Added sender_verify_fail logging option.
3379 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3380 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3381 tidied the source and removed it altogether.
3383 PH/33 When a queue run was abandoned because the load average was too high, a
3384 log line was always written; now it is written only if the queue_run log
3385 selector is set. In addition, the log line for abandonment now contains
3386 information about the queue run such as the pid. This is always present
3387 in "start" and "stop" lines but was omitted from the "abandon" line.
3389 PH/34 Omit spaces between a header name and the colon in the error message that
3390 is given when verify = headers_syntax fails (if there are lots of them,
3391 the message gets confusing).
3393 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3394 names, as there are now some PTR records that contain slashes. This check
3395 is only to protect against broken name servers that fall over on strange
3396 characters, so the fact that it applies to all lookups doesn't matter.
3398 PH/36 Now that the new test suite is complete, we can remove some of the
3399 special code in Exim that was needed for the old test suite. For example,
3400 sorting DNS records because real resolvers return them in an arbitrary
3401 order. The new test suite's fake resolver always returns records in the
3404 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3405 bounces) except when queue_only is set, to avoid logging races between
3406 the different processes.
3408 PH/38 Panic-die if .include specifies a non-absolute path.
3410 PH/39 A tweak to the "H" retry rule from its user.
3412 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3413 a label. They prevented compilation on older perls.
3415 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3416 a warning to be raised on newish perls.
3418 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3419 on queue. Changes to match documented behaviour of showing count of
3420 messages matching specified criteria.
3422 PH/40 Changed the default ident timeout from 30s to 5s.
3424 PH/41 Added support for the use of login_cap features, on those BSD systems
3425 that have them, for controlling the resources used by pipe deliveries.
3427 PH/42 The content-scanning code uses fopen() to create files in which to put
3428 message data. Previously it was not paying any attention to the mode of
3429 the files. Exim runs with umask(0) because the rest of the code creates
3430 files with open(), and sets the required mode explicitly. Thus, these
3431 files were ending up world-writeable. This was not a big issue, because,
3432 being within the spool directory, they were not world-accessible. I have
3433 created a function called modefopen, which takes an additional mode
3434 argument. It sets umask(777), creates the file, chmods it to the required
3435 mode, then resets the umask. All the relevant calls to fopen() in the
3436 content scanning code have been changed to use this function.
3438 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3439 to 24 hours. This avoids potential overflow problems when processing G
3440 and H retry rules. I suspect nobody ever tinkers with this value.
3442 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3444 PH/45 When the plaintext authenticator is running as a client, the server's
3445 challenges are checked to ensure they are valid base64 strings. By
3446 default, the authentication attempt is cancelled if an invalid string is
3447 received. Setting client_ignore_invalid_base64 true ignores these errors.
3448 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3449 they are received. Thus, the responses can be made to depend on the
3450 challenges. If an invalid string is ignored, an empty string is placed in
3453 PH/46 Messages that are created by the autoreply transport now contains a
3454 References: header, in accordance with RFCs 2822 and 3834.
3456 PH/47 Added authenticated_sender_force to the smtp transport.
3458 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3460 PH/49 Installed latest patch from the Sieve maintainer.
3462 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3463 was also set, the appendfile transport was unnecessarily scanning a
3464 directory of message files (e.g. for maildir delivery) to find the count
3465 of files (along with the size), even though it did not need this
3466 information. It now does the scan only if it needs to find either the
3467 size of the count of files.
3469 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3471 PH/52 Two bugs concerned with error handling when the smtp transport is
3474 (i) Exim was not creating retry information for temporary errors given
3475 for individual recipients after the DATA command when the smtp transport
3476 was used in LMTP mode. This meant that they could be retried too
3477 frequently, and not timed out correctly.
3479 (ii) Exim was setting the flag that allows error details to be returned
3480 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3481 recipients that were returned after the DATA command.
3483 PH/53 This is related to PH/52, but is more general: for any failing address,
3484 when detailed error information was permitted to be returned to the
3485 sender, but the error was temporary, then after the final timeout, only
3486 "retry timeout exceeded" was returned. Now it returns the full error as
3487 well as "retry timeout exceeded".
3489 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3490 do this, and (what is worse) MTAs that accept it.
3492 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3493 will now be deprecated.
3495 PH/56 New os.c-cygwin from the Cygwin maintainer.
3497 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3498 formats (previously only available in exim formats via -bpr, -bpru,
3499 and -bpra. Now also available in native and exiqgrep formats)
3501 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3502 with very large, slow to parse queues
3504 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3506 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3508 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3509 responses to authentication challenges, though it was showing the
3510 challenges; (ii) I've removed the CR characters from the debug output for
3513 PH/58 Allow for the insertion of a newline as well as a space when a string
3514 is turned into more than one encoded-word during RFC 2047 encoding. The
3515 Sieve code now uses this.
3517 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3518 data_4xx, lost_connection, tls_required.
3520 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3521 message was being sent as an SMTP response.
3523 PH/61 Add -l and -k options to exicyclog.
3525 PH/62 When verifying, if an address was redirected to one new address, so that
3526 verification continued, and the new address failed or deferred after
3527 having set something in $address_data, the value of $address_data was not
3528 passed back to the ACL. This was different to the case when no
3529 redirection occurred. The value is now passed back in both cases.
3531 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3532 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3533 use login_cap.h, so on its own it isn't the distinguishing feature. The
3534 new name refers directly to the setclassresources() function.
3536 PH/65 Added configuration files for NetBSD3.
3538 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3540 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3541 is preferred over IPv4.
3543 PH/68 The bounce_return_message and bounce_return_body options were not being
3544 honoured for bounces generated during the reception of non-SMTP messages.
3545 In particular, this applied to messages rejected by the ACL. This bug has
3546 been fixed. However, if bounce_return_message is true and bounce_return_
3547 body is false, the headers that are returned for a non-SMTP message
3548 include only those that have been read before the error was detected.
3549 (In the case of an ACL rejection, they have all been read.)
3551 PH/69 The HTML version of the specification is now built in a directory called
3552 spec_html instead of spec.html, because the latter looks like a path with
3553 a MIME-type, and this confuses some software.
3555 PH/70 Catch two compiler warnings in sieve.c.
3557 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3558 function verify_get_ident() calls ip_connect() to connect a socket, but
3559 if the "connect()" function timed out, ip_connect() used to close the
3560 socket. However, verify_get_ident() also closes the socket later, and in
3561 between Exim writes to the log, which may get opened at this point. When
3562 the socket was closed in ip_connect(), the log could get the same file
3563 descriptor number as the socket. This naturally causes chaos. The fix is
3564 not to close the socket in ip_connect(); the socket should be closed by
3565 the function that creates it. There was only one place in the code where
3566 this was missing, in the iplookup router, which I don't think anybody now
3567 uses, but I've fixed it anyway.
3569 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3570 well as to direct DNS lookups. Otherwise the handling of names in host
3571 lists is inconsistent and therefore confusing.
3577 PH/01 Two changes to the default runtime configuration:
3579 (1) Move the checks for relay_from_hosts and authenticated clients from
3580 after to before the (commented out) DNS black list checks.
3582 (2) Add control=submission to the relay_from_hosts and authenticated
3583 clients checks, on the grounds that messages accepted by these
3584 statements are most likely to be submissions.
3586 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3588 (1) Generate an error if the third argument for the ${prvs expansion is
3591 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3594 (3) Reset the variables that are obtained from the first argument of
3595 ${prvscheck and used in the second argument before leaving the code,
3596 because their memory is reclaimed, so using them afterwards may do
3599 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3600 one (it's much easier than Tom thought :-).
3602 (5) Because of (4), we can now allow for the use of $prvscheck_result
3603 inside the third argument.
3605 PH/03 For some reason, the default setting of PATH when running a command from
3606 a pipe transport was just "/usr/bin". I have changed it to
3609 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3610 anything to be listed in the output from -bV.
3612 PH/05 When a filter generated an autoreply, the entire To: header line was
3613 quoted in the delivery log line, like this:
3615 => >A.N.Other <ano@some.domain> <original@ddress> ...
3617 This has been changed so that it extracts the operative address. There
3618 may be more than one such address. If so, they are comma-separated, like
3621 => >ano@some.domain,ona@other.domain <original@ddress> ...
3623 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3624 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3625 not being looked up in the rDNS to get a host name, Exim was showing the
3626 IP address twice in Received: lines, even though the IP addresses were
3627 identical. For example:
3629 Received: from [1.2.3.4] (helo=[1.2.3.4])
3631 However, if the real host name was known, it was omitting the HELO data
3632 if it matched the actual IP address. This has been tidied up so that it
3633 doesn't show the same IP address twice.
3635 PH/07 When both +timestamp and +memory debugging was on, the value given by
3636 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3637 called by the debug printing, thereby overwriting the timestamp buffer.
3638 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3640 PH/08 When the original message was included in an autoreply transport, it
3641 always said "this is a copy of the message, including all the headers",
3642 even if body_only or headers_only was set. It now gives an appropriate
3645 PH/09 Applied a patch from the Sieve maintainer which:
3647 o fixes some comments
3648 o adds the (disabled) notify extension core
3649 o adds some debug output for the result of if/elsif tests
3650 o points to the current vacation draft in the documentation
3651 and documents the missing references header update
3655 o fixes a bug in processing the envelope test (when testing
3656 multiple envelope elements, the last element determined the
3659 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3660 Electronic Mail") by including:
3662 Auto-submitted: auto-generated
3664 in the messages that it generates (bounce messages and others, such as
3665 warnings). In the case of bounce messages for non-SMTP messages, there was
3666 also a typo: it was using "Auto_submitted" (underscore instead of
3667 hyphen). Since every message generated by Exim is necessarily in response
3668 to another message, thes have all been changed to:
3670 Auto-Submitted: auto-replied
3672 in accordance with these statements in the RFC:
3674 The auto-replied keyword:
3676 - SHOULD be used on messages sent in direct response to another
3677 message by an automatic process,
3679 - MUST NOT be used on manually-generated messages,
3681 - MAY be used on Delivery Status Notifications (DSNs) and Message
3682 Disposition Notifications (MDNs),
3684 - MUST NOT be used on messages generated by automatic or periodic
3685 processes, except for messages which are automatic responses to
3688 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3689 to the default Received: header definition.
3691 PH/12 Added log selector acl_warn_skipped (default on).
3693 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3694 variables because (a) they are in the wrong storage pool and (b) even if
3695 they were copied, it wouldn't work properly because of the caching.
3697 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3698 checking when decoding. Apparently there are clients that generate
3699 overlong encoded strings. Why am I not surprised?
3701 PH/15 If the first argument of "${if match_address" was not empty, but did not
3702 contain an "@" character, Exim crashed. Now it writes a panic log message
3703 and treats the condition as false.
3705 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3707 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3708 extension "envelope-auth". The code is finished and in agreement with
3709 other implementations, but there is no documentation so far and in fact,
3710 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3711 not changing the active code.
3713 Print executed "if" and "elsif" statements when debugging is used. This
3714 helps a great deal to understand what a filter does.
3716 Document more things not specified clearly in RFC3028. I had all this
3717 sorted out, when out of a sudden new issues came to my mind. Oops."
3719 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3722 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3723 canonical form (as documented). However, after a host name lookup from
3724 the IP address, check_host() was doing a simple string comparison with
3725 addresses acquired from the DNS when checking that the found name did
3726 have the original IP as one of its addresses. Since any found IPv6
3727 addresses are likely to be in abbreviated form, the comparison could
3728 fail. Luckily, there already exists a function for doing the comparison
3729 by converting both addresses to binary, so now that is used instead of
3730 the text comparison.
3732 PH/20 There was another similar case to PH/19, when a complete host name was
3733 given in a host list; looking up its IP address could give an abbreviated
3734 form, whereas the current host's name might or might not be abbreviated.
3735 The same fix has been applied.
3741 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3742 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3745 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3746 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3748 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3750 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3751 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3752 Also, for "Auto-Submitted", treat anything other than "no" as
3753 non-personal, in accordance with RFC 3834. (Previously it treated
3754 anything starting "auto-" as non-personal.)
3756 TF/01 The control=submission/name=... option had a problem with syntax
3757 errors if the name included a slash character. The /name= option
3758 now slurps the rest of the string, so it can include any characters
3759 but it must come last in the list of options (after /sender_retain
3762 PH/05 Some modifications to the interface to the fake nameserver for the new
3770 TK/01 Added the "success_on_redirect" address verification option. See
3771 NewStuff for rationale and an example.
3773 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3775 PH/02 Patch to exigrep to allow it to work on syslog lines.
3777 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3778 fread() to skip over the body file's header line, because in Cygwin the
3779 header line is locked and is inaccessible.
3781 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3782 co-exist for some time) to make it clear that it is the Exim ID that is
3783 referenced, not the Message-ID: header line.
3785 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3786 string_format() function, because snprintf() does not exist on all
3789 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3790 ${stat: expansion item.
3792 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3793 protocol synchronization error", to keep the pedants happy.
3795 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3796 well as for IRIX systems, when gcc is being used. See the host.c source
3799 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3801 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3804 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3805 transport and to the smtp transport in LMTP mode.
3807 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3809 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3810 run a filter in a subprocess. This could lead to confusion in subsequent
3811 lookups in the parent process. There should also be a search_tidyup() at
3812 the end of the subprocess.
3814 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3815 only if the host matched helo_try_verify_hosts, which caused the
3816 verification to occur when the EHLO/HELO command was issued. The ACL just
3817 tested the remembered result. Now, if a previous verification attempt has
3818 not happened, "verify = helo" does it there and then.
3820 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3822 TK/03 Fix log output including CR from clamd.
3824 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3825 exist provoked a memory error which could cause a segfault.
3827 PH/15 Installed PCRE 6.2
3829 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3831 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3832 of the problem. Specifically, suggested +O2 rather than +O1 for the
3835 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3837 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3838 record, the port was not correctly logged when the outgoing_port log
3839 selector was set (it logged the transort's default port).
3841 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3842 fallback_hosts, and "hosts" in the smtp transport.
3844 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3845 host errors such as "Connection refused".
3847 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3848 authentication with radiusclient 0.4.9:
3850 - Error returned from rc_read_config was caught wrongly
3851 - Username/password not passed on to radius server due to wrong length.
3853 The presumption is that some radiusclient API changes for 4.51/PH/17
3854 were not taken care of correctly. The code is still untested by me (my
3855 Linux distribution still has 0.3.2 of radiusclient), but it was
3856 contributed by a Radius user.
3858 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3859 expanding the "port" option of the smtp transport.
3861 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3862 while reading a MIME header. Thanks to Tom Hughes for a patch.
3864 PH/24 Include config.h inside local_scan.h so that configuration settings are
3867 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3868 that in an ACL for RCPT (for example), you can examine exactly what was
3871 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3872 commands, but it was not correctly comparing the address with the actual
3873 client host address. Thus, it would show the EHLO address in Received:
3874 header lines when this was not necessary.
3876 PH/27 Added the % operator to ${eval:}.
3878 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3879 it should be ignoring failures (because with -C, for example, it has lost
3880 privilege). It wasn't ignoring creation failures other than "already
3883 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3886 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3887 because an input line was too long, either on its own, or by virtue of
3888 too many continuations, the temporary file was not being removed, and the
3889 return code was incorrect.
3891 PH/31 Missing "BOOL" in function definition in filtertest.c.
3893 PH/32 Applied Sieve patches from the maintainer.
3895 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3897 PH/33 Added "verify = not_blind".
3899 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3900 Local/Makefile (with some defaults set). These are used in built scripts
3901 such as exicyclog, but they have never been used in the exim_install
3902 script (though there are many overriding facilities there). I have
3903 arranged that the exim_install script now takes note of these two
3906 PH/35 Installed configuration files for Dragonfly.
3908 PH/36 When a locally submitted message by a trusted user did not contain a
3909 From: header, and the sender address was obtained from -f or from an SMTP
3910 MAIL command, and the trusted user did not use -F to supply a sender
3911 name, $originator_name was incorrectly used when constructing a From:
3912 header. Furthermore, $originator_name was used for submission mode
3913 messages from external hosts without From: headers in a similar way,
3914 which is clearly wrong.
3916 PH/37 Added control=suppress_local_fixups.
3918 PH/38 When log_selector = +received_sender was set, and the addition of the
3919 sender made the log line's construction buffer exactly full, or one byte
3920 less than full, an overflow happened when the terminating "\n" was
3923 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3924 when the result of a list match is failure because a DNS lookup failed.
3926 PH/40 RM_COMMAND is now used in the building process.
3928 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3929 the "build-* directories that it finds.
3931 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3932 address in a domain literal was a prefix of an interface address.
3934 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3935 when verifying a sender address, unless rewrite_headers is false.
3937 PH/44 Wrote a long comment about why errors_to addresses are verified as
3938 recipients, not senders.
3940 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3941 the ratelimit ACL was added.
3943 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3945 PH/46 Added extra information about PostgreSQL errors to the error string.
3947 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3948 avoiding the need to install special zones in a real server. This is
3949 backwards compatible; if it can't find the fake resolver, it drops back.
3950 Thus, both old and new test suites can be run.
3952 TF/02 Added util/ratelimit.pl
3954 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3955 clock is set back in time.
3957 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3958 Candler <B.Candler@pobox.com>.
3960 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3961 OK for addresses that are the result of redirections.
3963 PH/48 A number of further additions for the benefit of the new test suite,
3964 including a fake gethostbyname() that interfaces to the fake DNS resolver
3967 TF/06 The fix for widen_domains has also been applied to qualify_single and
3968 search_parents which are the other dnslookup options that can cause
3971 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3974 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3976 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3977 service" if it encountered an MX record with an empty target hostname.
3978 The message is now "an MX or SRV record indicated no SMTP service".
3980 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3981 if the DNS of the sending site is misconfigured. This is quite a
3982 common situation. This change restores the behaviour of treating a
3983 helo verification defer as a failure.
3985 PH/51 If self=fail was set on a router, the bounce message did not include the
3986 actual error message.
3992 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3994 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3995 command itself ended in error, the underlying message about the transport
3996 filter timeout was being overwritten with the pipe command error. Now the
3997 underlying error message should be appended to the second error message.
3999 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
4001 PH/02 Reduce the amount of output that "make" produces by default. Full output
4002 can still be requested.
4004 PH/03 The warning log line about a condition test deferring for a "warn" verb
4005 was being output only once per connection, rather than after each
4006 occurrence (because it was using the same function as for successful
4007 "warn" verbs). This seems wrong, so I have changed it.
4009 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4010 it should not have, which might have caused a crash in the right
4011 circumstances, but probably never did.
4013 PH/04 Installed a modified version of Tony Finch's patch to make submission
4014 mode fix the return path as well as the Sender: header line, and to
4015 add a /name= option so that you can make the user's friendly name appear
4018 TF/03 Added the control = fakedefer ACL modifier.
4020 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4021 Mark Lowes for thorough testing.
4023 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4025 TK/03 Merged latest SRS patch from Miles Wilton.
4027 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4028 with the definition in sysexits.h (which is #included earlier).
4029 Fortunately, Exim does not actually use EX_OK. The code used to try to
4030 preserve the sysexits.h value, by assuming that macro definitions were
4031 scanned for macro replacements. I have been disabused of this notion,
4032 so now the code just undefines EX_OK before #including unistd.h.
4034 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4035 in the smtp transport. When a block could not be written in a single
4036 write() function, the timeout was being re-applied to each part-write.
4037 This seems wrong - if the receiver was accepting one byte at a time it
4038 would take for ever. The timeout is now adjusted when this happens. It
4039 doesn't have to be particularly precise.
4041 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4042 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4044 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4045 without a domain if the check to <postmaster@domain> fails.
4047 SC/01 Eximstats: added -xls and the ability to specify output files
4048 (patch written by Frank Heydlauf).
4050 SC/02 Eximstats: use FileHandles for outputting results.
4052 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4054 SC/04 Eximstats: fixed display of large numbers with -nvr option
4056 SC/05 Eximstats: fixed merging of reports with empty tables.
4058 SC/06 Eximstats: added the -include_original_destination flag
4060 SC/07 Eximstats: removed tabs and trailing whitespace.
4062 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4064 TK/06 MBOX spool code: Add real "From " MBOX separator line
4065 so the .eml file is really in mbox format (even though
4066 most programs do not really care). Patch from Alex Miller.
4068 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4069 The latter is generated from $received_to and is only set if the
4070 message has one envelope recipient. SA can use these headers,
4071 obviously out-of-the-box. Patch from Alex Miller.
4073 PH/08 The ${def test on a variable was returning false if the variable's
4074 value was "0", contrary to what the specification has always said!
4075 The result should be true unless the variable is empty.
4077 PH/09 The syntax error of a character other than { following "${if
4078 def:variable_name" (after optional whitespace) was not being diagnosed.
4079 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4080 accidental colon was present, for example, could give incorrect results.
4082 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4083 result is used (not including appendfile, where other changes are about
4086 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4087 This involved changing a lot of size variables from int to off_t. It
4088 should work with maildirs and everything.
4090 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4091 spamd dying while we are connected to it.
4093 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4096 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4097 function for BDB 4.3.
4099 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4101 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4102 the sources for PCRE have been re-arranged and more files are now
4105 PH/15 The code I had for printing potentially long long variables in PH/11
4106 above was not the best (it lost precision). The length of off_t variables
4107 is now inspected at build time, and an appropriate printing format (%ld
4108 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4109 to be "long long int" or "long int". This is needed for the internal
4110 formatting function string_vformat().
4112 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4113 the configuration file to be ":syslog", then the script "guesses" where
4114 the logs files are, rather than using the compiled in default. In our
4115 case the guess is not the same as the compiled default, so the script
4116 suddenly stopped working when I started to use syslog. The patch checks
4117 to see if log_file_path is "". If so, it attempts to read it from exim
4118 with no configuration file to get the compiled in version, before it
4119 falls back to the previous guessing code."
4121 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4122 implementing BATV in an Exim configuration. See NewStuff for the gory
4125 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4126 Makefile that are specific to HP-UX.
4128 PH/18 If the "use_postmaster" option was set for a recipient callout together
4129 with the "random" option, the postmaster address was used as the MAIL
4130 FROM address for the random test, but not for the subsequent recipient
4131 test. It is now used for both.
4133 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4134 patch removes a few documentation additions to RFC 3028, because the
4135 latest draft now contains them. It adds the new en;ascii-case comparator
4136 and a new error check for 8bit text in MIME parts. Comparator and
4137 require names are now matched exactly. I enabled the subaddress
4138 extension, but it is not well tested yet (read: it works for me)."
4140 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4141 rework some of the code of TK/09 above to avoid the hardwired use of
4142 "%lld" and "long long". Replaced the call to snprintf() with a call to
4145 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4146 records point to non-existent hosts", "retry timeout exceeded", and
4147 "retry time not reached for any host after a long failure period".
4149 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4150 experimental DomainKeys support:
4152 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4153 (2) On an error such as an illegally used "control", the wrong name for
4154 the control was given.
4156 These problems did NOT occur unless DomainKeys support was compiled.
4158 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4160 PH/24 Added ${if match_ip condition.
4162 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4163 databases so that it will be absolutely obvious if a crash occurs in the
4164 DB library. This is a regular occurrence (often caused by mis-matched
4167 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4168 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4169 picked up on a user's system that detects such things. There doesn't seem
4170 to be a gcc warning option for this - only an attribute that has to be
4171 put on the function's prototype. It seems that in Fedora Core 4 they have
4172 set this on a number of new functions. No doubt there will be more in due
4175 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4176 specify a transport. However, if an address that was verified by such a
4177 router was the subject of a callout, Exim crashed because it tried to
4178 read the rcpt_include_affixes from the non-existent transport. Now it
4179 just assumes that the setting of that option is false. This bug was
4180 introduced by 4.51/PH/31.
4182 PH/28 Changed -d+all to exclude +memory, because that information is very
4183 rarely of interest, but it makes the output a lot bigger. People tend to
4184 do -d+all out of habit.
4186 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4187 code in os-type was giving problems when libc.so lives in lib64, like on
4190 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4191 aren't the modern standard, and it seems that some systems' include files
4192 don't always have them. Exim was already checking for some of the newer
4193 ones like T_AAAA, and defining it itself. I've added checks for all the
4194 record types that Exim uses.
4196 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4197 not automatically generating a new one, as it is supposed to. This
4198 prevented TLS from working. If the file did exist, but contained invalid
4199 data, a new version was generated, as expected. It was only the case of a
4200 non-existent file that was broken.
4202 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4203 with a change in libdomainkeys > 0.64.
4205 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4206 from DNS. If the selector record carries the flag, it now has
4207 precedence over the domain-wide flag.
4209 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4211 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4212 the use of a port name that isn't defined in /etc/services) occurred, the
4213 message was deferred as in a normal delivery, and thus remained on the
4214 spool, instead of being failed because of the mua_wrapper setting. This
4215 is now fixed, and I tidied up some of the mua_wrapper messages at the
4218 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4219 the messages in a hash of arrays rather than using individual hashes.
4220 This is a bit cleaner and results in dramatic memory savings, albeit
4221 at a slight CPU cost.
4223 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4224 as requested by Marc Sherman.
4226 SC/10 Eximstats: added histograms for user specified patterns as requested
4229 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4231 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4232 fopen() in the content-scanning modules that did not already have it.
4238 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4239 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4241 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4243 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4245 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4246 to test Sieve filters that use "vacation".
4248 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4249 that changes the way the GnuTLS parameters are stored in the cache file.
4250 The new format can be generated externally. For backward compatibility,
4251 if the data in the cache doesn't make sense, Exim assumes it has read an
4252 old-format file, and it generates new data and writes a new file. This
4253 means that you can't go back to an older release without removing the
4256 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4257 work if there are any delivery delays because "one_time" forces the
4258 parent to be marked "delivered", so its unseen clone is never tried
4259 again. For this reason, Exim now forbids the simultaneous setting of
4262 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4263 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4264 entry if you want to know the details. The fix, however, neglected to
4265 consider the case when local delivery batching is involved. The test for
4266 "previously delivered" was not happening when checking to see if an
4267 address could be batched with a previous (undelivered) one; under
4268 certain circumstances this could lead to multiple deliveries to the same
4271 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4272 in its include files, and this causes problems building Exim.
4274 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4275 header_syntax) but Exim was just ignoring anything given after a slash.
4276 In particular, this caused confusion with an attempt to use "verify =
4277 reverse_host_lookup/defer_ok". An error is now given when options are
4278 supplied for verify items that do not have them. (Maybe reverse_host_
4279 lookup should have a defer_ok option, but that's a different point.)
4281 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4282 defined by RFC 821) to 2048, because there were problems with some AUTH
4283 commands, and RFC 1869 says the size should be increased for extended
4284 SMTP commands that take arguments.
4286 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4289 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4290 "unknown" error; now it says that the functionality isn't in the binary.
4292 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4293 an address' error message when a string expansion fails (syntax or
4294 whatever). Otherwise the password may appear in the log. Following change
4295 PH/42 below, there is no longer a chance of it appearing in a bounce
4298 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4300 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4301 of its addresses was ever tried. (Bugzilla bug #2.)
4303 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4304 the result incorrectly in the debug output. (It correctly added a newline
4305 to what was transported.)
4307 TF/01 Added $received_time.
4309 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4310 commented out examples of how to interface to a virus scanner and to
4311 SpamAssassin. Also added commented examples of av_scanner and
4312 spamd_address settings.
4314 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4315 and controls are allowed in which ACLs. There were a couple of minor
4316 errors. Some of the entries in the conditions table (which is a table of
4317 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4318 negation of where the condition IS allowed.
4320 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4322 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4323 header file does not have a version number, so I've had to invent a new
4324 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4325 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4326 radiusclient), but it was contributed by a Radius user.
4328 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4329 files or directories, for OpenSSL.
4331 PH/19 When an Exim process that is running as root has to create an Exim log
4332 file, it does so in a subprocess that runs as exim:exim so as to get the
4333 ownership right at creation (otherwise, other Exim processes might see
4334 the file with the wrong ownership). There was no test for failure of this
4335 fork() call, which would lead to the process getting stuck as it waited
4336 for a non-existent subprocess. Forks do occasionally fail when resources
4337 run out. I reviewed all the other calls to fork(); they all seem to check
4340 PH/20 When checking for unexpected SMTP input at connect time (before writing
4341 the banner), Exim was not dealing correctly with a non-positive return
4342 from the read() function. If the client had disconnected by this time,
4343 the result was a log entry for a synchronization error with an empty
4344 string after "input=" when read() returned zero. If read() returned -1
4345 (an event I could not check), uninitialized data bytes were printed.
4346 There were reports of junk text (parts of files, etc) appearing after
4349 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4351 PH/22 Added support for macro redefinition, and (re)definition in between
4352 driver and ACL definitions.
4354 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4355 forgetting to use the resulting value; it was using the unexpanded value.
4357 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4358 hadn't been configured. The fix is from Juergen Kreileder, who
4359 understands it better than I do:
4361 "Here's what I see happening with three configured cyrus_sasl
4362 authenticators configured (plain, login, cram-md5):
4364 On startup auth_cyrus_sasl_init() gets called for each of these.
4365 This means three calls to sasl_listmech() without a specified mech_list.
4366 => SASL tests which mechs of all available mechs actually work
4367 => three warnings about OTP not working
4368 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4370 With the patch, sasl_listmech() also gets called three times. But now
4371 SASL's mech_list option is set to the server_mech specified in the the
4372 authenticator. Or in other words, the answer from sasl_listmech()
4373 gets limited to just the mech you're testing for (which is different
4375 => the return list contains just 'plain' or 'login', 'cram-md5' or
4376 nothing depending on the value of ob->server_mech.
4378 I've just tested the patch: Authentication still works fine,
4379 unavailable mechs specified in the exim configuration are still
4380 caught, and the auth.log warnings about OTP are gone."
4382 PH/25 When debugging is enabled, the contents of the command line are added
4383 to the debugging output, even when log_selector=+arguments is not
4386 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4387 answer is "GNU", and only if the return is "GNU/something" is the answer
4390 PH/27 $acl_verify_message is now set immediately after the failure of a
4391 verification in an ACL, and so is available in subsequent modifiers. In
4392 particular, the message can be preserved by coding like this:
4394 warn !verify = sender
4395 set acl_m0 = $acl_verify_message
4397 Previously, $acl_verify_message was set only while expanding "message"
4398 and "log_message" when a very denied access.
4400 PH/28 Modified OS/os.c-Linux with
4402 -#ifndef OS_LOAD_AVERAGE
4403 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4405 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4406 nomenclature these days.)
4408 PH/29 Installed patch from the Sieve maintainer that adds the options
4409 sieve_useraddress and sieve_subaddress to the redirect router.
4411 PH/30 In these circumstances:
4412 . Two addresses routed to the same list of hosts;
4413 . First host does not offer TLS;
4414 . First host accepts first address;
4415 . First host gives temporary error to second address;
4416 . Second host offers TLS and a TLS session is established;
4417 . Second host accepts second address.
4418 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4419 and peerdn, if requested) that were in fact used only for the second
4422 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4423 attention to any local part prefix or suffix that was matched by the
4424 router that accepted the address. It now behaves in the same way as it
4425 does for delivery: the affixes are removed from the local part unless
4426 rcpt_include_affixes is set on the transport.
4428 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4429 timeout during the DATA phase of an incoming message.
4431 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4432 applied a patch sanctioned by the Sieve maintainer.
4434 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4435 the uid or gid is negative. A case of a negative gid caused this to be
4436 noticed. The fix allows for either to be negative.
4438 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4439 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4442 PH/36 The change PH/12 above was broken. Fixed it.
4444 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4445 the grounds that routing the same address twice would always produce the
4446 same answer. This might have been true once, but it is certainly no
4447 longer true now. Routing a child address may depend on the previous
4448 routing that produced that child. Some complicated redirection strategies
4449 went wrong when messages had multiple recipients, and made Exim's
4450 behaviour dependent on the order in which the addresses were given.
4452 I have moved the duplicate checking until after the routing is complete.
4453 Exim scans the addresses that are assigned to local and remote
4454 transports, and removes any duplicates. This means that more work will be
4455 done, as duplicates will always all be routed, but duplicates are
4456 presumably rare, so I don't expect this is of any significance.
4458 For deliveries to pipes, files, and autoreplies, the duplicate checking
4459 still happens during the routing process, since they are not going to be
4462 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4463 It corrects a timeout issue with spamd. This is Ian's comment: "The
4464 background is that sometimes spamd either never reads data from a
4465 connection it has accepted, or it never writes response data. The exiscan
4466 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4467 blindly assumes that writes won't block so it may never time out."
4469 PH/39 Allow G after quota size as well as K and M.
4471 PH/40 The value set for $authenticated_id in an authenticator may not contain
4472 binary zeroes or newlines because the value is written to log lines and
4473 to spool files. There was no check on this. Now the value is run through
4474 the string_printing() function so that such characters are converted to
4475 printable escape sequences.
4477 PH/41 $message_linecount is a new variable that contains the total number of
4478 lines in the message. Compare $body_linecount, which is the count for the
4481 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4482 bounce and delay warning messages, except in certain special cases, which
4485 (a) An SMTP error message from a remote host;
4486 (b) A message specified in a :fail: redirection;
4487 (c) A message specified in a "fail" command in a system filter;
4488 (d) A message specified in a FAIL return from the queryprogram router;
4489 (e) A message specified by the cannot_route_message router option.
4491 In these cases only, Exim does include the error details in bounce and
4492 warning messages. There are also a few cases where bland messages such
4493 as "unrouteable address" or "local delivery error" are given.
4495 PH/43 $value is now also set for the "else" part of a ${run expansion.
4497 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4498 being worked on, but at least Exim now implements the latest version to
4501 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4502 process to complete was treated as a delivery failure, a timeout while
4503 writing the message to the pipe was logged, but erroneously treated as a
4504 successful delivery. Such timeouts include transport filter timeouts. For
4505 consistency with the overall process timeout, these timeouts are now
4506 treated as errors, giving rise to delivery failures by default. However,
4507 there is now a new Boolean option for the pipe transport called
4508 timeout_defer, which, if set TRUE, converts the failures into defers for
4509 both kinds of timeout. A transport filter timeout is now identified in
4512 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4513 systems where "make" and "gmake" are different, calling "gmake" at top
4514 level broke things. I've arranged for the value of $(MAKE) to be passed
4515 from the Makefile to this script so that it can call the same version of
4519 A note about Exim versions 4.44 and 4.50
4520 ----------------------------------------
4522 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4523 changes of various kinds. As a consequence, a big documentation update was
4524 needed. This delayed the release for rather longer than seemed good, especially
4525 in the light of a couple of (minor) security issues. Therefore, the changes
4526 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4527 release. So 4.44 and 4.50 are in effect two different branches that both start
4530 I have left the 4.50 change log unchanged; it contains all the changes since
4531 4.43. The change log for 4.44 is below; many of its items are identical to
4532 those for 4.50. This seems to be the most sensible way to preserve the
4533 historical information.
4539 1. Minor wording change to the doc/README.SIEVE file.
4541 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4542 computation of the current number of files was incorrect.
4544 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4545 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4548 4. Give more explanation in the error message when the command for a transport
4549 filter fails to execute.
4551 5. There are several places where Exim runs a non-Exim command in a
4552 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4553 was being done only for the command run by the queryprogram router. It is
4554 now done for all such subprocesses. The other cases are: ${run, transport
4555 filters, and the commands run by the lmtp and pipe transports.
4557 6. Added CONFIGURE_GROUP build-time option.
4559 7. Some older OS have a limit of 256 on the maximum number of file
4560 descriptors. Exim was using setrlimit() to set 1000 as a large value
4561 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4562 systems. I've change it so that if it can't get 1000, it tries for 256.
4564 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4565 was an oversight, and furthermore, ever since the addition of extra
4566 controls (e.g. 4.43/32), the checks on when to allow different forms of
4567 "control" were broken. There should now be diagnostics for all cases when a
4568 control that does not make sense is encountered.
4570 9. Added the /retain_sender option to "control=submission".
4572 10. $recipients is now available in the predata ACL (oversight).
4574 11. Tidy the search cache before the fork to do a delivery from a message
4575 received from the command line. Otherwise the child will trigger a lookup
4576 failure and thereby defer the delivery if it tries to use (for example) a
4577 cached ldap connection that the parent has called unbind on.
4579 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4580 of $address_data from the recipient verification was clobbered by the
4581 sender verification.
4583 13. The value of address_data from a sender verification is now available in
4584 $sender_address_data in subsequent conditions in the ACL statement.
4586 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4588 15. Added a new option "connect=<time>" to callout options, to set a different
4591 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4592 was its contents. (It was OK if the option was not defined at all.)
4594 17. A "Completed" log line is now written for messages that are removed from
4595 the spool by the -Mrm option.
4597 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4598 information about exactly what failed.
4600 19. Added -dd to debug only the daemon process.
4602 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4603 handles timeouts, both on the server side and network timeouts. Renamed the
4604 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4606 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4607 It is now set to "smtps".
4609 22. $host_address is now set to the target address during the checking of
4610 ignore_target_hosts.
4612 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4613 being passed; this would have caused $sender_host_name to have been used if
4614 matching the list had actually called for a host name (not very likely,
4615 since this list is usually IP addresses). A host name is now passed as
4618 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4619 code that specifies a non-restarting handler (typically sigaction() in
4620 modern systems) in an attempt to fix a rare and obscure crash bug.
4622 25. Narrowed the window for a race in the daemon that could cause it to ignore
4623 SIGCHLD signals. This is not a major problem, because they are used only to
4624 wake it up if nothing else does.
4626 26. A malformed maildirsize file could cause Exim to calculate negative values
4627 for the mailbox size or file count. Odd effects could occur as a result.
4628 The maildirsize information is now recalculated if the size or filecount
4631 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4632 support for a long time. Removed HAVE_SYS_VFS_H.
4634 28. Installed the latest version of exipick from John Jetmore.
4636 29. In an address list, if the pattern was not a regular expression, an empty
4637 subject address (from a bounce message) matched only if the pattern was an
4638 empty string. Non-empty patterns were not even tested. This was the wrong
4639 because it is perfectly reasonable to use an empty address as part of a
4640 database query. An empty address is now tested by patterns that are
4641 lookups. However, all the other forms of pattern expect the subject to
4642 contain a local part and a domain, and therefore, for them, an empty
4643 address still always fails if the pattern is not itself empty.
4645 30. Exim went into a mad DNS loop when attempting to do a callout where the
4646 host was specified on an smtp transport, and looking it up yielded more
4647 than one IP address.
4649 31. Re-factored the code for checking spool and log partition space into a
4650 function that finds that data and another that does the check. The former
4651 is then used to implement four new variables: $spool_space, $log_space,
4652 $spool_inodes, and $log_inodes.
4654 32. The RFC2047 encoding function was originally intended for short strings
4655 such as real names; it was not keeping to the 75-character limit for
4656 encoded words that the RFC imposes. It now respects the limit, and
4657 generates multiple encoded words if necessary. To be on the safe side, I
4658 have increased the buffer size for the ${rfc2047: expansion operator from
4661 33. It is now permitted to omit both strings after an "if" condition; if the
4662 condition is true, the result is "true". As before, when the second string
4663 is omitted, a false condition yields an empty string. This makes it less
4664 cumbersome to write custom ACL and router conditions.
4666 34. Failure to deliver a bounce message always caused it to be frozen, even if
4667 there was an errors_to setting on the router. The errors_to setting is now
4670 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4671 canonical form (fully expanded) before being placed in
4672 $sender_host_address.
4674 36. The table in the code that translates DNS record types into text (T_A to
4675 "A" for instance) was missing entries for NS and CNAME. It is just possible
4676 that this could have caused confusion if both these types were looked up
4677 for the same domain, because the text type is used as part of Exim's
4678 per-process caching. But the chance of anyone hitting this buglet seems
4681 37. The dnsdb lookup has been extended in a number of ways.
4683 (1) There is a new type, "zns", which walks up the domain tree until it
4684 finds some nameserver records. It should be used with care.
4686 (2) There is a new type, "mxh", which is like "mx" except that it returns
4687 just the host names, not the priorities.
4689 (3) It is now possible to give a list of domains (or IP addresses) to be
4690 looked up. The behaviour when one of the lookups defers can be
4691 controlled by a keyword.
4693 (4) It is now possible to specify the separator character for use when
4694 multiple records are returned.
4696 38. The dnslists ACL condition has been extended: it is now possible to supply
4697 a list of IP addresses and/or domains to be looked up in a particular DNS
4700 39. Added log_selector=+queue_time_overall.
4702 40. When running the queue in the test harness, wait just a tad after forking a
4703 delivery process, to get repeatability of debugging output.
4705 41. Include certificate and key file names in error message when GnuTLS fails
4706 to set them up, because the GnuTLS error message doesn't include the name
4707 of the failing file when there is a problem reading it.
4709 42. Allow both -bf and -bF in the same test run.
4711 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4713 44. The "Exiscan patch" is now merged into the mainline Exim source.
4715 45. Sometimes the final signoff response after QUIT could fail to get
4716 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4717 < 0 before doing a fflush(). This bug looks as though it goes back to the
4718 introduction of TLS in release 3.20, but "sometimes" must have been rare
4719 because the tests only now provoked it.
4721 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4722 (this can affect the format of dates).
4724 47. exim_tidydb, when checking for the continued existence of a message for
4725 which it has found a message-specific retry record, was not finding
4726 messages that were in split spool directories. Consequently, it was
4727 deleting retry records that should have stayed in existence.
4729 48. Steve fixed some bugs in eximstats.
4731 49. The SPA authentication driver was not abandoning authentication and moving
4732 on to the next authenticator when an expansion was forced to fail,
4733 contradicting the general specification for all authenticators. Instead it
4734 was generating a temporary error. It now behaves as specified.
4736 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4737 (the order specifies the preference for clients). The order is now AES256,
4738 AES128, 3DES, ARCFOUR128.
4740 51. Small patch to Sieve code - explicitly set From: when generating an
4743 52. Exim crashed if a remote delivery caused a very long error message to be
4744 recorded - for instance if somebody sent an entire SpamAssassin report back
4745 as a large number of 550 error lines. This bug was coincidentally fixed by
4746 increasing the size of one of Exim's internal buffers (big_buffer) that
4747 happened as part of the Exiscan merge. However, to be on the safe side, I
4748 have made the code more robust (and fixed the comments that describe what
4751 53. Now that there can be additional text after "Completed" in log lines (if
4752 the queue_time_overall log selector is set), a one-byte patch to exigrep
4753 was needed to allow it to recognize "Completed" as not the last thing in
4756 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4757 patch that reportedly fixes this has been added. I am not expert enough to
4758 create a test for it. This is what the patch creator wrote:
4760 "I found a little strange behaviour of ldap code when working with
4761 Windows 2003 AD Domain, where users was placed in more than one
4762 Organization Units. When I tried to give exim partial DN, the exit code
4763 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4764 But simultaneously result of request was absolutely normal ldap result,
4765 so I produce this patch..."
4767 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4768 REFERENCE, so I have modified the code to exclude the patch when that macro
4771 55. Some experimental protocols are using DNS PTR records for new purposes. The
4772 keys for these records are domain names, not reversed IP addresses. The
4773 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4774 leaves it alone. Component reversal etc. now happens only for IP addresses.
4777 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4779 57. Double the size of the debug message buffer (to 2048) so that more of very
4780 long debug lines gets shown.
4782 58. The exicyclog utility now does better if the number of log files to keep
4783 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4785 59. Two changes related to the smtp_active_hostname option:
4787 (1) $smtp_active_hostname is now available as a variable.
4788 (2) The default for smtp_banner uses $smtp_active_hostname instead
4789 of $primary_hostname.
4791 60. The host_aton() function is supposed to be passed a string that is known
4792 to be a valid IP address. However, in the case of IPv6 addresses, it was
4793 not checking this. This is a hostage to fortune. Exim now panics and dies
4794 if the condition is not met. A case was found where this could be provoked
4795 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4796 components; fortuitously, this particular loophole had already been fixed
4797 by change 4.50/55 above.
4799 If there are any other similar loopholes, the new check in host_aton()
4800 itself should stop them being exploited. The report I received stated that
4801 data on the command line could provoke the exploit when Exim was running as
4802 exim, but did not say which command line option was involved. All I could
4803 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4804 running as the user.
4807 61. There was a buffer overflow vulnerability in the SPA authentication code
4808 (which came originally from the Samba project). I have added a test to the
4809 spa_base64_to_bits() function which I hope fixes it.
4812 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4813 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4815 63. The daemon start-up calls getloadavg() while still root for those OS that
4816 need the first call to be done as root, but it missed one case: when
4817 deliver_queue_load_max is set with deliver_drop_privilege. This is
4818 necessary for the benefit of the queue runner, because there is no re-exec
4819 when deliver_drop_privilege is set.
4821 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4822 This has been fixed.
4824 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4825 was in use, was not putting the data itself into the right store pool;
4826 consequently, it could be overwritten for a subsequent message in the same
4827 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4830 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4832 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4833 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4834 the function were treating the return as a boolean value, which happened to
4835 work because 0=false and not-0=true, but is not correct code.
4837 68. The host_aton() function was not handling scoped IPv6 addresses (those
4838 with, for example, "%eth0" on the end) correctly.
4840 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4841 negated items (that is, ~something) in unsigned ints. Some compilers
4842 apparently mutter when there is no cast.
4844 70. If an address verification called from an ACL failed, and did not produce a
4845 user-specific message (i.e. there was only a "system" message), nothing was
4846 put in $acl_verify_message. In this situation, it now puts the system
4849 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4850 session; change 4.31/43 added the unwanted input to the log line - except
4851 that it did not do this in the start of session case. It now does.
4853 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4854 This is wrong and can cause the other end to generate a synchronization
4855 error if it is another Exim or anything else that does the synchronization
4856 check. A QUIT command is no longer sent after a timeout.
4858 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4859 during host lookups.
4861 74. The defer_ok option of callout verification was not working if it was used
4862 when verifying addresses in header lines, that is, for this case:
4864 verify = header_sender/callout=defer_ok
4866 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4867 those file descriptors could be used for SMTP connections. If anything
4868 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4869 could be sent to the SMTP client, causing chaos. The daemon now opens
4870 stdin, stdout, and stderr to /dev/null when it puts itself into the
4873 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4874 log by default. The user can override this with suitable Perl magic.
4876 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4877 the log message when discard triggers, was not working for the DATA ACL or
4878 for the non-SMTP ACL.
4880 78. Error message wording change in sieve.c.
4882 79. If smtp_accept_max_per_host was set, the number of connections could be
4883 restricted to fewer than expected, because the daemon was trying to set up
4884 a new connection before checking whether the processes handling previous
4885 connections had finished. The check for completed processes is now done
4886 earlier. On busy systems, this bug wouldn't be noticed because something
4887 else would have woken the daemon, and it would have reaped the completed
4890 80. If a message was submitted locally by a user whose login name contained one
4891 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4892 It caused a spool format error. I have fixed the spool reading code. A
4893 related problem was that the "from" clause in the Received: line became
4894 illegal because of the space(s). It is now covered by ${quote_local_part.
4896 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4899 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4901 83. Patch from Sieve maintainer to fix unterminated string problem in
4902 "vacation" handling.
4904 84. Some minor changes to the Linux configuration files to help with other
4905 OS variants using glibc.
4907 85. One more patch for Sieve to update vacation handling to latest spec.
4910 ----------------------------------------------------
4911 See the note above about the 4.44 and 4.50 releases.
4912 ----------------------------------------------------
4918 1. Change 4.43/35 introduced a bug that caused file counts to be
4919 incorrectly computed when quota_filecount was set in an appendfile
4922 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4923 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4926 3. Give more explanation in the error message when the command for a transport
4927 filter fails to execute.
4929 4. There are several places where Exim runs a non-Exim command in a
4930 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4931 was being done only for the command run by the queryprogram router. It is
4932 now done for all such subprocesses. The other cases are: ${run, transport
4933 filters, and the commands run by the lmtp and pipe transports.
4935 5. Some older OS have a limit of 256 on the maximum number of file
4936 descriptors. Exim was using setrlimit() to set 1000 as a large value
4937 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4938 systems. I've change it so that if it can't get 1000, it tries for 256.
4940 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4941 was an oversight, and furthermore, ever since the addition of extra
4942 controls (e.g. 4.43/32), the checks on when to allow different forms of
4943 "control" were broken. There should now be diagnostics for all cases when a
4944 control that does not make sense is encountered.
4946 7. $recipients is now available in the predata ACL (oversight).
4948 8. Tidy the search cache before the fork to do a delivery from a message
4949 received from the command line. Otherwise the child will trigger a lookup
4950 failure and thereby defer the delivery if it tries to use (for example) a
4951 cached ldap connection that the parent has called unbind on.
4953 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4954 of $address_data from the recipient verification was clobbered by the
4955 sender verification.
4957 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4958 was its contents. (It was OK if the option was not defined at all.)
4960 11. A "Completed" log line is now written for messages that are removed from
4961 the spool by the -Mrm option.
4963 12. $host_address is now set to the target address during the checking of
4964 ignore_target_hosts.
4966 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4967 being passed; this would have caused $sender_host_name to have been used if
4968 matching the list had actually called for a host name (not very likely,
4969 since this list is usually IP addresses). A host name is now passed as
4972 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4973 code that specifies a non-restarting handler (typically sigaction() in
4974 modern systems) in an attempt to fix a rare and obscure crash bug.
4976 15. Narrowed the window for a race in the daemon that could cause it to ignore
4977 SIGCHLD signals. This is not a major problem, because they are used only to
4978 wake it up if nothing else does.
4980 16. A malformed maildirsize file could cause Exim to calculate negative values
4981 for the mailbox size or file count. Odd effects could occur as a result.
4982 The maildirsize information is now recalculated if the size or filecount
4985 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4986 support for a long time. Removed HAVE_SYS_VFS_H.
4988 18. Updated exipick to current release from John Jetmore.
4990 19. Allow an empty sender to be matched against a lookup in an address list.
4991 Previously the only cases considered were a regular expression, or an
4994 20. Exim went into a mad DNS lookup loop when doing a callout where the
4995 host was specified on the transport, if the DNS lookup yielded more than
4998 21. The RFC2047 encoding function was originally intended for short strings
4999 such as real names; it was not keeping to the 75-character limit for
5000 encoded words that the RFC imposes. It now respects the limit, and
5001 generates multiple encoded words if necessary. To be on the safe side, I
5002 have increased the buffer size for the ${rfc2047: expansion operator from
5005 22. Failure to deliver a bounce message always caused it to be frozen, even if
5006 there was an errors_to setting on the router. The errors_to setting is now
5009 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5010 canonical form (fully expanded) before being placed in
5011 $sender_host_address.
5013 24. Updated eximstats to version 1.33
5015 25. Include certificate and key file names in error message when GnuTLS fails
5016 to set them up, because the GnuTLS error message doesn't include the name
5017 of the failing file when there is a problem reading it.
5019 26. Expand error message when OpenSSL has problems setting up cert/key files.
5022 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5023 (this can affect the format of dates).
5025 28. exim_tidydb, when checking for the continued existence of a message for
5026 which it has found a message-specific retry record, was not finding
5027 messages that were in split spool directories. Consequently, it was
5028 deleting retry records that should have stayed in existence.
5030 29. eximstats updated to version 1.35
5031 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5032 1.35 - bugfix such that pie charts by volume are generated correctly
5034 30. The SPA authentication driver was not abandoning authentication and moving
5035 on to the next authenticator when an expansion was forced to fail,
5036 contradicting the general specification for all authenticators. Instead it
5037 was generating a temporary error. It now behaves as specified.
5039 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5040 (the order specifies the preference for clients). The order is now AES256,
5041 AES128, 3DES, ARCFOUR128.
5043 31. Small patch to Sieve code - explicitly set From: when generating an
5046 32. Exim crashed if a remote delivery caused a very long error message to be
5047 recorded - for instance if somebody sent an entire SpamAssassin report back
5048 as a large number of 550 error lines. This bug was coincidentally fixed by
5049 increasing the size of one of Exim's internal buffers (big_buffer) that
5050 happened as part of the Exiscan merge. However, to be on the safe side, I
5051 have made the code more robust (and fixed the comments that describe what
5054 33. Some experimental protocols are using DNS PTR records for new purposes. The
5055 keys for these records are domain names, not reversed IP addresses. The
5056 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5057 leaves it alone. Component reversal etc. now happens only for IP addresses.
5060 34. The host_aton() function is supposed to be passed a string that is known
5061 to be a valid IP address. However, in the case of IPv6 addresses, it was
5062 not checking this. This is a hostage to fortune. Exim now panics and dies
5063 if the condition is not met. A case was found where this could be provoked
5064 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5065 components; fortuitously, this particular loophole had already been fixed
5066 by change 4.50/55 or 4.44/33 above.
5068 If there are any other similar loopholes, the new check in host_aton()
5069 itself should stop them being exploited. The report I received stated that
5070 data on the command line could provoke the exploit when Exim was running as
5071 exim, but did not say which command line option was involved. All I could
5072 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5073 running as the user.
5076 35. There was a buffer overflow vulnerability in the SPA authentication code
5077 (which came originally from the Samba project). I have added a test to the
5078 spa_base64_to_bits() function which I hope fixes it.
5081 36. The daemon start-up calls getloadavg() while still root for those OS that
5082 need the first call to be done as root, but it missed one case: when
5083 deliver_queue_load_max is set with deliver_drop_privilege. This is
5084 necessary for the benefit of the queue runner, because there is no re-exec
5085 when deliver_drop_privilege is set.
5087 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5088 was in use, was not putting the data itself into the right store pool;
5089 consequently, it could be overwritten for a subsequent message in the same
5090 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5093 38. Sometimes the final signoff response after QUIT could fail to get
5094 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5095 < 0 before doing a fflush(). This bug looks as though it goes back to the
5096 introduction of TLS in release 3.20, but "sometimes" must have been rare
5097 because the tests only now provoked it.
5103 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5104 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5105 Now it returns an integer. A number of calls were still expecting a T/F
5106 return. Fortuitously, in all cases, the tests worked in OK situations,
5107 which is the norm. However, things would have gone wrong on any write
5108 failures on the smtp file descriptor. This function is used when sending
5109 messages over SMTP and also when doing verify callouts.
5111 2. When Exim is called to do synchronous delivery of a locally submitted
5112 message (the -odf or -odi options), it no longer closes stderr before doing
5115 3. Implemented the mua_wrapper option.
5117 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5119 5. Implemented the functions header_remove(), header_testname(),
5120 header_add_at_position(), and receive_remove_recipient(), and exported them
5123 6. If an ACL "warn" statement specified the addition of headers, Exim already
5124 inserted X-ACL-Warn: at the start if there was no header name. However, it
5125 was not making this test for the second and subsequent header lines if
5126 there were newlines in the string. This meant that an invalid header could
5127 be inserted if Exim was badly configured.
5129 7. Allow an ACL "warn" statement to add header lines at the start or after all
5130 the Received: headers, as well as at the end.
5132 8. Added the rcpt_4xx retry error code.
5134 9. Added postmaster_mailfrom=xxx to callout verification option.
5136 10. Added mailfrom=xxxx to the callout verification option, for verify=
5139 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5140 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5142 12. Inserted some casts to stop certain compilers warning when using pointer
5143 differences as field lengths or precisions in printf-type calls (mostly
5144 affecting debugging statements).
5146 13. Added optional readline() support for -be (dynamically loaded).
5148 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5149 same clock tick as a message's arrival, so that its received time was the
5150 same as the "first fail" time on the retry record, and that message
5151 remained on the queue past the ultimate address timeout, every queue runner
5152 would try a delivery (because it was past the ultimate address timeout) but
5153 after another failure, the ultimate address timeout, which should have then
5154 bounced the address, did not kick in. This was a "< instead of <=" error;
5155 in most cases the first failure would have been in the next clock tick
5156 after the received time, and all would be well.
5158 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5159 being recognized when the domain list was tested by the match_domain
5160 condition in an expansion string.
5162 16. Added the ${str2b64: operator.
5164 17. Exim was always calling setrlimit() to set a large limit for the number of
5165 processes, without checking whether the existing limit was already
5166 adequate. (It did check for the limit on file descriptors.) Furthermore,
5167 errors from getrlimit() and setrlimit() were being ignored. Now they are
5168 logged to the main and panic logs, but Exim does carry on, to try to do its
5169 job under whatever limits there are.
5171 18. Imported PCRE 5.0.
5173 19. Trivial typo in log message " temporarily refused connection" (the leading
5176 20. If the log selector return_path_on_delivery was set and an address was
5177 redirected to /dev/null, the delivery process crashed because it assumed
5178 that a return path would always be set for a "successful" delivery. In this
5179 case, the whole delivery is bypassed as an optimization, and therefore no
5182 21. Internal re-arrangement: the function for sending a challenge and reading
5183 a response while authentication was assuming a zero-terminated challenge
5184 string. It's now changed to take a pointer and a length, to allow for
5185 binary data in such strings.
5187 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5189 23. Exim was not respecting finduser_retries when seeking the login of the
5190 uid under which it was called; it was always trying 10 times. (The default
5191 setting of finduser_retries is zero.) Also, it was sleeping after the final
5192 failure, which is pointless.
5194 24. Implemented tls_on_connect_ports.
5196 25. Implemented acl_smtp_predata.
5198 26. If the domain in control=submission is set empty, Exim assumes that the
5199 authenticated id is a complete email address when it generates From: or
5200 Sender: header lines.
5202 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5203 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5204 chown and chgrp in /bin and hostname in /usr/bin.
5206 28. Exim was keeping the "process log" file open after each use, just as it
5207 does for the main log. This opens the possibility of it remaining open for
5208 long periods when the USR1 signal hits a daemon. Occasional processlog
5209 errors were reported, that could have been caused by this. Anyway, it seems
5210 much more sensible not to leave this file open at all, so that is what now
5213 29. The long-running daemon process does not normally write to the log once it
5214 has entered its main loop, and it closes the log before doing so. This is
5215 so that log files can straightforwardly be renamed and moved. However,
5216 there are a couple of unusual error situations where the daemon does write
5217 log entries, and I had neglected to close the log afterwards.
5219 30. The text of an SMTP error response that was received during a remote
5220 delivery was being truncated at 512 bytes. This is too short for some of
5221 the long messages that one sometimes sees. I've increased the limit to
5224 31. It is now possible to make retry rules that apply only when a message has a
5225 specific sender, in particular, an empty sender.
5227 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5228 it possible to be selective about when SMTP synchronization is enforced.
5230 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5232 32. Implemented hosts_connection_nolog.
5234 33. Added an ACL for QUIT.
5236 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5239 35. Added mailbox_size and mailbox_filecount to appendfile.
5241 36. Added control = no_multiline_responses to ACLs.
5243 37. There was a bug in the logic of the code that waits for the clock to tick
5244 in the case where the clock went backwards by a substantial amount such
5245 that the microsecond fraction of "now" was more than the microsecond
5246 fraction of "then" (but the whole seconds number was less).
5248 38. Added support for the libradius Radius client library this is found on
5249 FreeBSD (previously only the radiusclient library was supported).
5255 1. When certain lookups returned multiple values in the form name=value, the
5256 quoting of the values was not always being done properly. Specifically:
5257 (a) If the value started with a double quote, but contained no whitespace,
5259 (b) If the value contained whitespace other than a space character (i.e.
5260 tabs or newlines or carriage returns) it was not quoted.
5261 This fix has been applied to the mysql and pgsql lookups by writing a
5262 separate quoting function and calling it from the lookup code. The fix
5263 should probably also be applied to nisplus, ibase and oracle lookups, but
5264 since I cannot test any of those, I have not disturbed their existing code.
5266 2. A hit in the callout cache for a specific address caused a log line with no
5267 reason for rejecting RCPT. Now it says "Previous (cached) callout
5268 verification failure".
5270 3. There was an off-by-one bug in the queryprogram router. An over-long
5271 return line was truncated at 256 instead of 255 characters, thereby
5272 overflowing its buffer with the terminating zero. As well as fixing this, I
5273 have increased the buffer size to 1024 (and made a note to document this).
5275 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5276 when Exim is waiting for an SMTP response from a remote server, Exim
5277 restarts its select() call on the socket, thereby resetting its timeout.
5278 This is not a problem when such interrupts are rare. Somebody set up a cron
5279 job to run exiwhat every 2 minutes, which is less than the normal select()
5280 timeout (5 or 10 minutes). This meant that the select() timeout never
5281 kicked in because it was always reset. I have fixed this by comparing the
5282 time when an interrupt arrives with the time at the start of the first call
5283 to select(). If more time than the timeout has elapsed, the interrupt is
5284 treated as a timeout.
5286 5. Some internal re-factoring in preparation for the addition of Sieve
5287 extensions (by MH). In particular, the "personal" test is moved to a
5288 separate function, and given an option for scanning Cc: and Bcc: (which is
5289 not set for Exim filters).
5291 6. When Exim created an email address using the login of the caller as the
5292 local part (e.g. when creating a From: or Sender: header line), it was not
5293 quoting the local part when it contained special characters such as @.
5295 7. Installed new OpenBSD configuration files.
5297 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5298 try to make them clearer.
5300 9. Callout options, other than the timeout value, were being ignored when
5301 verifying sender addresses in header lines. For example, when using
5303 verify = header_sender/callout=no_cache
5305 the cache was (incorrectly) being used.
5307 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5308 only the Cygwin environment.
5310 11. When return_path_on_delivery was set as a log selector, if different remote
5311 addresses in the same message used different return paths and parallel
5312 remote delivery occurred, the wrong values would sometimes be logged.
5313 (Whenever a remote delivery process finished, the return path value from
5314 the most recently started remote delivery process was logged.)
5316 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5317 lines when AUTH and/or TLS are in use. This is the "received protocol"
5318 field. Exim used to use "asmtp" for authenticated SMTP, without any
5319 indication (in the protocol name) for TLS use. Now it follows the RFC and
5320 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5321 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5322 names appear in log lines as well as in Received: header lines.
5324 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5325 extensions, and comparison tests, and to fix some bugs.
5327 14. Changes to the "personal" filter test:
5329 (1) The test was buggy in that it was just doing the equivalent of
5330 "contains" tests on header lines. For example, if a user's address was
5331 anne@some.where, the "personal" test would incorrectly be true for
5333 To: susanne@some.where
5335 This test is now done by extracting each address from the header in turn,
5336 and checking the entire address. Other tests that are part of "personal"
5337 are now done using regular expressions (for example, to check local parts
5338 of addresses in From: header lines).
5340 (2) The list of non-personal local parts in From: addresses has been
5341 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5342 taken from the Sieve specification recommendations.
5344 (3) If the message contains any header line starting with "List-" it is
5345 treated as non-personal.
5347 (4) The test for "circular" in the Subject: header line has been removed
5348 because it now seems ill-conceived.
5350 15. Minor typos in src/EDITME comments corrected.
5352 16. Installed latest exipick from John Jetmore.
5354 17. If headers_add on a router specified a text string that was too long for
5355 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5356 of string_sprintf() is now avoided.
5358 18. $message_body_size was not set (it was always zero) when running the DATA
5359 ACL and the local_scan() function.
5361 19. For the "mail" command in an Exim filter, no default was being set for
5362 the once_repeat time, causing a random time value to be used if "once" was
5363 specified. (If the value happened to be <= 0, no repeat happened.) The
5364 default is now 0s, meaning "never repeat". The "vacation" command was OK
5365 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5366 (I found it when inspecting the code).
5368 20. There is now an overall timeout for performing a callout verification. It
5369 defaults to 4 times the callout timeout, which applies to individual SMTP
5370 commands during the callout. The overall timeout applies when there is more
5371 than one host that can be tried. The timeout is checked before trying the
5372 next host. This prevents very long delays if there are a large number of
5373 hosts and all are timing out (e.g. when the network connections are timing
5374 out). The value of the overall timeout can be changed by specifying an
5375 additional sub-option for "callout", called "maxwait". For example:
5377 verify = sender/callout=5s,maxwait=20s
5379 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5380 to the end before writing, but this should make it even safer).
5382 22. Exim was forgetting that it had advertised PIPELINING for the second and
5383 subsequent messages on an SMTP connection. It was also not resetting its
5384 memory on STARTTLS and an internal HELO.
5386 23. When Exim logs an SMTP synchronization error within a session, it now
5387 records whether PIPELINING has been advertised or not.
5389 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5390 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5391 rather than long int.
5393 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5395 26. Added the never_mail option to autoreply.
5401 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5402 crash if the getsockname() call failed; this can happen if a connection is
5403 closed very soon after it is established. The problem was simply in the
5404 order in which certain operations were done, causing Exim to try to write
5405 to the SMTP stream before it had set up the file descriptor. The bug has
5406 been fixed by making things happen in the correct order.
5412 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5413 before the connection was closed, thus losing the rejection response.
5415 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5416 some early Solaris releases, but causes trouble in current releases where
5417 socklen_t is defined.
5419 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5422 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5425 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5426 that releasing the top of it at the end releases what was used for sub-
5427 expansions (unless the block got too big). However, discard this block if
5428 the first thing is a variable or header, so that we can use its block when
5429 it is dynamic (useful for very large $message_headers, for example).
5431 6. Lookups now cache *every* query, not just the most recent. A new, separate
5432 store pool is used for this. It can be recovered when all lookup caches are
5433 flushed. Lookups now release memory at the end of their result strings.
5434 This has involved some general refactoring of the lookup sources.
5436 7. Some code has been added to the store_xxx() functions to reduce the amount
5437 of flapping under certain conditions.
5439 8. log_incoming_interface used to affect only the <= reception log lines. Now
5440 it causes the local interface and port to be added to several more SMTP log
5441 lines, for example "SMTP connection from", and rejection lines.
5443 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5445 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5447 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5448 could be overwritten at the end of the current message (or the start of a
5449 new message if it was set in a HELO ACL). The value is now preserved for
5450 the duration of the SMTP connection.
5452 12. If a transport had a headers_rewrite setting, and a matching header line
5453 contained an unqualified address, that address was qualified, even if it
5454 did not match any rewriting rules. The underlying bug was that the values
5455 of the flags that permit the existence of unqualified sender and recipient
5456 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5457 non-local messages, and by -bnq for local messages) were not being
5458 preserved with the message after it was received.
5460 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5461 "next input=" as part of the text comprising the host identity instead of
5462 the correct text. The code was using the same buffer for two different
5463 strings. However, depending on which order the printing function evaluated
5464 its arguments, the bug did not always show up. Under Linux, for example, my
5465 test suite worked just fine.
5467 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5468 doesn't work with some older versions of Perl. It has been changed to "my",
5469 which in any case is probably the better facility to use.
5471 15. A really picky compiler found some instances of statements for creating
5472 error messages that either had too many or two few arguments for the format
5475 16. The size of the buffer for calls to the DNS resolver has been increased
5476 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5477 for addresses that have a lot of PTR records. This alleviates a problem; it
5478 does not fully solve it.
5480 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5481 buffer now truncates the list and logs the incident, which is the same
5482 action as happens when Exim is looking up a host name and its aliases.
5483 Previously in this situation something unpredictable would happen;
5484 sometimes it was "internal error: store_reset failed".
5486 18. If a server dropped the connection unexpectedly when an Exim client was
5487 using GnuTLS and trying to read a response, the client delivery process
5488 crashed while trying to generate an error log message.
5490 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5491 string, for example:
5493 warn message = H1: something\nH2: something
5495 the text was added as a single header line from Exim's point of view
5496 though it ended up OK in the delivered message. However, searching for the
5497 second and subsequent header lines using $h_h2: did not work. This has been
5498 fixed. Similarly, if a system filter added multiple headers in this way,
5499 the routers could not see them.
5501 20. Expanded the error message when iplsearch is called with an invalid key to
5502 suggest using net-iplsearch in a host list.
5504 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5505 ACLs are no longer actually imposed (and a message to that effect is
5508 22. If a "gecos" field in a passwd entry contained escaped characters, in
5509 particular, if it contained a \" sequence, Exim got it wrong when building
5510 a From: or a Sender: header from that name. A second bug also caused
5511 incorrect handling when an unquoted " was present following a character
5512 that needed quoting.
5514 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5515 was not being matched caselessly.
5517 24. Arranged for all hyphens in the exim.8 source to be escaped with
5520 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5521 properly. Recipient callout cache records were still being keyed to include
5522 the sender, even when use_sender was set false. This led to far more
5523 callouts that were necessary. The sender is no longer included in the key
5524 when use_sender is false.
5526 26. Added "control = submission" modifier to ACLs.
5528 27. Added the ${base62d: operator to decode base 62 numbers.
5530 28. dnsdb lookups can now access SRV records.
5532 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5533 the configuration file.
5535 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5536 (-v) mode. This makes the output for a verbose queue run more intelligible.
5538 31. Added a use_postmaster feature to recipient callouts.
5540 32. Added the $body_zerocount variable, containing the number of binary zero
5541 bytes in the message body.
5543 33. The time of last modification of the "new" subdirectory is now used as the
5544 "mailbox time last read" when there is a quota error for a maildir
5547 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5549 35. Added +ignore_unknown as a special item in host lists.
5551 36. Code for decoding IPv6 addresses in host lists is now included, even if
5552 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5553 address was recognized as an IP address, but was then not correctly decoded
5554 into binary, causing unexpected and incorrect effects when compared with
5561 1. Very minor rewording of debugging text in manualroute to say "list of
5562 hosts" instead of "hostlist".
5564 2. If verify=header_syntax was set, and a header line with an unqualified
5565 address (no domain) and a large number of spaces between the end of the
5566 name and the colon was received, the reception process suffered a buffer
5567 overflow, and (when I tested it) crashed. This was caused by some obsolete
5568 code that should have been removed. The fix is to remove it!
5570 3. When running in the test harness, delay a bit after writing a bounce
5571 message to get a bit more predictability in the log output.
5573 4. Added a call to search_tidyup() just before forking a reception process. In
5574 theory, someone could use a lookup in the expansion of smtp_accept_max_
5575 per_host which, without the tidyup, could leave open a database connection.
5577 5. Added the variables $recipient_data and $sender_data which get set from a
5578 lookup success in an ACL "recipients" or "senders" condition, or a router
5579 "senders" option, similar to $domain_data and $local_part_data.
5581 6. Moved the writing of debug_print from before to after the "senders" test
5584 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5585 problems for message scanning, either using a data ACL, or using
5586 local_scan() because the Received: header was not generated till after they
5587 were called (in order to set the time as the time of reception completion).
5588 I have revised the way this works. The header is now generated after the
5589 body is received, but before the ACL or local_scan() are called. After they
5590 are run, the timestamp in the header is updated.
5596 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5597 before starting a queue runner without re-exec. This happened only when
5598 deliver_drop_privilege was set or when the Exim user was set to root. The
5599 effect of the bug was that timeouts during subsequent deliveries caused
5600 crashes instead of being properly handled. The handler is now left at its
5601 default (and expected) setting.
5603 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5604 message, again when deliver_drop_privilege is set or Exim is run as root.
5605 The bug described in (1) was not present in this case, but the tidying up
5606 of the other signals was missing. I have made the two cases consistent.
5608 3. The ignore_target_hosts setting on a manualroute router was being ignored
5609 for hosts that were looked up using the /MX notation.
5611 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5614 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5615 operated on the sender address. After changing the $sender_address to <>
5616 for the sender address verify, Exim was re-instated it as the original
5617 (before rewriting) address, but remembering that it had rewritten it, so it
5618 wasn't rewriting it again. This bug also had the effect of breaking the
5619 sender address verification caching when the sender address was rewritten.
5621 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5622 This has been changed so that if the ip literal address matches
5623 ignore_target_hosts, the router declines.
5625 7. Added expansion conditions match_domain, match_address, and match_local_
5626 part (NOT match_host).
5628 8. The placeholder for the Received: header didn't have a length field set.
5630 9. Added code to Exim itself and to exim_lock to test for a specific race
5631 condition that could lead to file corruption when using MBX delivery. The
5632 issue is with the lockfile that is created in /tmp. If this file is removed
5633 after a process has opened it but before that process has acquired a lock,
5634 there is the potential for a second process to recreate the file and also
5635 acquire a lock. This could lead to two Exim processes writing to the file
5636 at the same time. The added code performs the same test as UW imapd; it
5637 checks after acquiring the lock that its file descriptor still refers to
5638 the same named file.
5640 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5641 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5644 11. Added the smtp_active_hostname option. If used, this will typically be made
5645 to depend on the incoming interface address. Because $interface_address is
5646 not set up until the daemon has forked a reception process, error responses
5647 that can happen earlier (such as "too many connections") no longer contain
5650 12. If an expansion in a condition on a "warn" statement fails because a lookup
5651 defers, the "warn" statement is abandoned, and the next ACL statement is
5652 processed. Previously this caused the whole ACL to be aborted.
5654 13. Added the iplsearch lookup type.
5656 14. Added ident_timeout as a log selector.
5658 15. Added tls_certificate_verified as a log selector.
5660 16. Added a global option tls_require_ciphers (compare the smtp transport
5661 option of the same name). This controls incoming TLS connections.
5663 17. I finally figured out how to make tls_require_ciphers do a similar thing
5664 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5665 before starting the TLS session.
5667 18. Tabs are now shown as \t in -bP output.
5669 19. If the log selector return_path_on_delivery was set, Exim crashed when
5670 bouncing a message because it had too many Received: header lines.
5672 20. If two routers both had headers_remove settings, and the first one included
5673 a superfluous trailing colon, the final name in the first list and the
5674 first name in the second list were incorrectly joined into one item (with a
5675 colon in the middle).
5681 1. Added -C and -D options to the exinext utility, mainly to make it easier
5682 to include in the automated testing, but these could be helpful when
5683 multiple configurations are in use.
5685 2. The exinext utility was not formatting the output nicely when there was
5686 an alternate port involved in the retry record key, nor when there was a
5687 message id as well (for retries that were specific to a specific message
5688 and a specific host). It was also confused by IPv6 addresses, because of
5689 the additional colons they contain. I have fixed the IPv4 problem, and
5690 patched it up to do a reasonable job for IPv6.
5692 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5693 delivery, the log line now contains "pipelined" if PIPELINING was used.
5695 4. An SMTP transport process used to panic and die if the bind() call to set
5696 an explicit outgoing interface failed. This has been changed; it is now
5697 treated in the same way as a connect() failure.
5699 5. A reference to $sender_host_name in the part of a conditional expansion
5700 that was being skipped was still causing a DNS lookup. This no longer
5703 6. The def: expansion condition was not recognizing references to header lines
5704 that used bh_ and bheader_.
5706 7. Added the _cache feature to named lists.
5708 8. The code for checking quota_filecount in the appendfile transport was
5709 allowing one more file than it should have been.
5711 9. For compatibility with Sendmail, the command line option
5719 and sets the incoming protocol and host name (for trusted callers). The
5720 host name and its colon can be omitted when only the protocol is to be set.
5721 Note the Exim already has two private options, -pd and -ps, that refer to
5722 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5723 "s", but I don't think that's a major issue.
5725 10. A number of refactoring changes to the code, none of which should affect
5728 (a) The number of logging options was getting close to filling up the
5729 32-bit word that was used as a bit map. I have split them into two classes:
5730 those that are passed in the argument to log_write(), and those that are
5731 only ever tested independently outside of that function. These are now in
5732 separate 32-bit words, so there is plenty of room for expansion again.
5733 There is no change in the user interface or the logging behaviour.
5735 (b) When building, for example, log lines, the code previously used a
5736 macro that called string_cat() twice, in order to add two strings. This is
5737 not really sufficiently general. Furthermore, there was one instance where
5738 it was actually wrong because one of the argument was used twice, and in
5739 one call a function was used. (As it happened, calling the function twice
5740 did not affect the overall behaviour.) The macro has been replaced by a
5741 function that can join an arbitrary number of extra strings onto a growing
5744 (c) The code for expansion conditions now uses a table and a binary chop
5745 instead of a serial search (which was left over from when there were very
5746 few conditions). Also, it now recognizes conditions like "pam" even when
5747 the relevant support is not compiled in: a suitably worded error message is
5748 given if an attempt is made to use such a condition.
5750 11. Added ${time_interval:xxxxx}.
5752 12. A bug was causing one of the ddress fields not to be passed back correctly
5753 from remote delivery subprocesses. The field in question was not being
5754 subsequently used, so this caused to problems in practice.
5756 13. Added new log selectors queue_time and deliver_time.
5758 14. Might have fixed a bug in maildirsizefile handling that threw up
5759 "unexpected character" debug warnings, and recalculated the data
5760 unnecessarily. In any case, I expanded the warning message to give more
5763 15. Added the message "Restricted characters in address" to the statements in
5764 the default ACL that block characters like @ and % in local parts.
5766 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5767 Three changes have been made:
5769 (a) There was a serious bug; a negative response to MAIL caused the whole
5770 recipient domain to be cached as invalid, thereby blocking all messages
5771 to all local parts at the same domain, from all senders. This bug has
5772 been fixed. The domain is no longer cached after a negative response to
5773 MAIL if the sender used is not empty.
5775 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5778 (c) A new callout option, "use_sender" has been added for people who want
5779 the modified behaviour.
5785 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5788 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5789 indeed breaks things for older releases.
5791 3. Added additional logging to the case where there is a problem reading data
5792 from a filter that is running in a subprocess using a pipe, in order to
5793 try to track down a specific problem.
5795 4. Testing facility fudge: when running in the test harness and attempting
5796 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5797 getting "No route to host". Convert this to a timeout.
5799 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5802 6. Some OS don't have socklen_t but use size_t instead. This affects the
5803 fifth argument of getsockopt() amongst other things. This is now
5804 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5805 can be set for individual OS. I have set it for SunOS5, OSF1, and
5806 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5807 some earlier ones do not.
5809 7. Change 4.30/15 was not doing the test caselessly.
5811 8. The standard form for an IPv6 address literal was being rejected by address
5812 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5813 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5814 this, as well as the form without the "IPv6" on the front (but only when
5815 address literals are enabled, of course).
5817 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5819 10. Exim crashed if a message with an empty sender address specified by -f
5820 encountered a router with an errors_to setting. This could be provoked only
5821 by a command such as
5825 where an empty string was supplied; "<>" did not hit this bug.
5827 11. Installed PCRE release 4.5.
5829 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5830 remained set. It is now erased.
5832 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5833 times from message ids (which are base 36 rather than the normal 62).
5835 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5836 were being counted as actual protocol errors, and logged if the log
5837 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5838 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5839 and DATA following a set of rejected RCPTs do not count as protocol errors.
5840 In other words, Exim assumes they were pipelined, though this may not
5841 actually be the case. Of course, in all cases the client gets an
5842 appropriate error code.
5844 15. If a lookup fails in an ACL condition, a message about the failure may
5845 be available; it is used if testing the ACL cannot continue, because most
5846 such messages specify what the cause of the deferral is. However, some
5847 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5848 that caused an old message to be retained and used if a later statement
5849 caused a defer, replacing the real cause of the deferral.
5851 16. If an IP address had so many PTR records that the DNS lookup buffer
5852 was not large enough to hold them, Exim could crash while trying to process
5853 the truncated data. It now detects and logs this case.
5855 17. Further to 4.21/58, another change has been made: if (and only if) the
5856 first line of a message (the first header line) ends with CRLF, a bare LF
5857 in a subsequent header line has a space inserted after it, so as not to
5858 terminate the header.
5860 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5861 unnecessarily, used atoi() instead of strtol(), and didn't check the
5862 termination when getting file sizes from file names by regex.
5864 19. Completely re-implemented the support for maildirsize files, in the light
5865 of a number of problems with the previous contributed implementation
5866 (4.30/29). In particular:
5868 . If the quota is zero, the maildirsize file is maintained, but no quota is
5871 . If the maildir directory does not exist, it is created before any attempt
5872 to write a maildirsize file.
5874 . The quota value in the file is just a cache; if the quota is changed in
5875 the transport, the new value overrides.
5877 . A regular expression is available for excluding directories from the
5880 20. The autoreply transport checks the characters in options that define the
5881 message's headers; it allows continued headers, but it was checking with
5882 isspace() after an embedded newline instead of explicitly looking for a
5885 21. If all the "regular" hosts to which an address was routed had passed their
5886 expiry times, and had not reached their retry times, the address was
5887 bounced, even if fallback hosts were defined. Now Exim should go on to try
5890 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5891 equivalent code in the SMTP transport. Some hosts send humungous responses
5892 to HELO/EHLO, more than 1024 it seems.
5894 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5895 gives compiler warnings in some environments. I've now done it "properly",
5898 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5899 (because of problems with the built-in one) was declared to return uschar *
5900 instead of char *, causing compiler failure.
5902 25. Fixed a file descriptor leak when processing alias/forward files.
5904 26. Fixed a minor format string issue in dbfn.c.
5906 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5908 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5909 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5912 29. When a sender address is verified, it is cached, to save repeating the test
5913 when there is more than one recipient in a message. However, when the
5914 verification involves a callout, it is possible for different callout
5915 options to be set for different recipients. It is too complicated to keep
5916 track of this in the cache, so now Exim always runs a verification when a
5917 callout is required, relying on the callout cache for the optimization.
5918 The overhead is duplication of the address routing, but this should not be
5921 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5922 to be verified with callout=postmaster, and the main callout worked but the
5923 postmaster check failed, the verification correctly failed. However, if a
5924 subsequent RCPT command asked for sender verification *without* the
5925 postmaster check, incorrect caching caused this verification also to fail,
5928 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5929 it was not caching the DNS options (qualify_single, search_parents) that
5930 were used when the lookup failed. A subsequent lookup with different
5931 options therefore always gave the same answer, though there were cases
5932 where it should not have. (Example: a "domains = !$mx_any" option on a
5933 dnslookup router: the "domains" option is always processed without any
5934 widening, but the router might have qualify_single set.) Now Exim uses the
5935 cached value only when the same options are set.
5937 32. Added John Jetmore's "exipick" utility to the distribution.
5939 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5940 than a timeout (e.g. a certificate is required, and is not provided), an
5941 Exim server now closes the connection immediately. Previously it waited for
5942 the client to close - but if the client is SSL, it seems that they each
5943 wait for each other, leading to a delay before one of them times out.
5945 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5946 maintained 0.8.x compatibility because I don't think many are using it, and
5947 it is clearly obsolete.
5949 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5952 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5953 client certificate was expired. A simple patch fixes this, though I don't
5954 understand the full logic of why the verify callback is called multiple
5957 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5958 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5959 which causes problems with some clients (such as the Certicom SSL Plus
5960 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5961 disables the coutermeasure allowing Eudora to connect."
5963 38. Exim was not checking that a write() to a log file succeeded. This could
5964 lead to Bad Things if a log got too big, in particular if it hit a file
5965 size limit. Exim now panics and dies if it cannot write to a log file, just
5966 as it does if it cannot open a log file.
5968 39. Modified OS/Makefile-Linux so that it now contains
5970 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5972 The two -D definitions ensure that Exim is compiled with large file
5973 support, which makes it possible to handle log files that are bigger than
5976 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5977 instance) a domain was checked against a named list that involved a lookup,
5978 causing $domain_data to be set, then another domain was checked against the
5979 same list, then the first domain was re-checked, the value of $domain_data
5980 after the final check could be wrong. In particular, if the second check
5981 failed, it could be set empty. This bug probably also applied to
5984 41. The strip_trailing_dot option was not being applied to the address given
5985 with the -f command-line option.
5987 42. The code for reading a message's header from the spool was incrementing
5988 $received_count, but never initializing it. This meant that the value was
5989 incorrect (doubled) while delivering a message in the same process in which
5990 it was received. In the most common configuration of Exim, this never
5991 happens - a fresh exec is done - but it can happen when
5992 deliver_drop_privilege is set.
5994 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5995 it now includes up to 150 characters of the unexpected data in the log
5998 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5999 and building data strings. The size of both of these buffers was 10 000
6000 bytes - far larger than anybody would *ever* want, thought I. Needless to
6001 say, somebody hit the limit. I have increased the maximum line length to
6002 20 000 and the maximum data length of concatenated lines to 100 000. I have
6003 also fixed two bugs, because there was no checking on these buffers. Tsk,
6004 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
6005 buffer is too small.
6007 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6008 lsearch lookups. Now it does.
6010 46. When parsing a route_list item in a manualroute router, a fixed-length
6011 buffer was used for the list of hosts. I made this 1024 bytes long,
6012 thinking that nobody would ever have a list of hosts that long. Wrong.
6013 Somebody had a whole pile of complicated expansion conditions, and the
6014 string was silently truncated, leading to an expansion error. It turns out
6015 that it is easier to change to an unlimited length (owing to other changes
6016 that have happened since this code was originally written) than to build
6017 structure for giving a limitation error. The length of the item that
6018 expands into the list of hosts is now unlimited.
6020 47. The lsearch lookup could not handle data where the length of text line was
6021 more than 4095 characters. Such lines were truncated, leading to shortened
6022 data being returned. It should now handle lines of any length.
6024 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6025 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6028 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6029 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6030 no longer happens in comment lines. A list of replacements is now placed
6031 at the head of all of the source files, except those whose only change is
6032 to replace PERL_COMMAND in the very first #! line.
6034 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6035 messages on the queue, with a bottom-up merge sort, using code contributed
6036 by Michael Haardt. This should make operations like -bp somewhat faster on
6037 large queues. It won't affect queue runners, except when queue_run_in_order
6040 51. Installed eximstats 1.31 in the distribution.
6042 52. Added support for SRV lookups to the dnslookup router.
6044 53. If an ACL referred to $message_body or $message_body_end, the value was not
6045 reset for any messages that followed in the same SMTP session.
6047 54. The store-handling optimization for building very long strings was not
6048 differentiating between the different store pools. I don't think this
6049 actually made any difference in practice, but I've tidied it.
6051 55. While running the routers to verify a sender address, $sender_address
6052 was still set to the sender address. This is wrong, because when routing to
6053 send a bounce to the sender, it would be empty. Therefore, I have changed
6054 it so that, while verifying a sender address, $sender_address is set to <>.
6055 (There is no change to what happens when verifying a recipient address.)
6057 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6058 target A or AAAA records (if not already returned) without resetting the
6059 qualify_single or search_parents options of the DNS resolver. These are
6060 inappropriate in this case because the targets of MX and SRV records must
6061 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6062 target that, when qualified, matched something in the local domain. These
6063 two options are now turned off when doing these lookups.
6065 57. It seems that at least some releases of Reiserfs (which does not have the
6066 concept of a fixed number of inodes) returns zero and not -1 for the
6067 number of available inodes. This interacted badly with check_spool_inodes,
6068 which assumed that -1 was the "no such thing" setting. What I have done is
6069 to check that the total number of inodes is greater than zero before doing
6070 the test of how many are available.
6072 58. When a "warn" ACL statement has a log_message modifier, the message is
6073 remembered, and not repeated. This is to avoid a lot of repetition when a
6074 message has many recipients that cause the same warning to be written.
6075 However, Exim was preserving the list of already written lines for an
6076 entire SMTP session, which doesn't seem right. The memory is now reset if a
6077 new message is started.
6079 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6080 debugging output unless log_rewrite was also set.
6082 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6083 of (int)(handle) when we know that handle contains (void *)(-1).
6085 61. The Exim daemon panic-logs an error return when it closes the incoming
6086 connection. However "connection reset by peer" seems to be common, and
6087 isn't really an error worthy of noting specially, so that particular error
6090 62. When Exim is trying to find all the local interfaces, it used to panic and
6091 die if the ioctl to get the interface flags failed. However, it seems that
6092 on at least one OS (Solaris 9) it is possible to have an interface that is
6093 included in the list of interfaces, but for which you get a failure error
6094 for this call. This happens when the interface is not "plumbed" into a
6095 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6096 failure of the "get flags" call assumes that the interface is down.
6098 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6099 makes life easier for people who are doing arithmetic on fields extracted
6100 from dates, where you often get leading zeros that should not be
6101 interpreted as octal.
6103 64. Added qualify_domain to the redirect router, to override the global
6106 65. If a pathologically long header line contained very many addresses (the
6107 report of this problem mentioned 10 000) and each of them was rewritten,
6108 Exim could use up a very large amount of memory. (It kept on making new
6109 copies of the header line as it rewrote, and never released the old ones.)
6110 At the expense of a bit more processing, the header rewriting function has
6111 been changed so that it no longer eats memory in this way.
6113 66. The generation of the Received: header has been moved from the time that a
6114 message starts to be received, to the time that it finishes. The timestamp
6115 in the Received: header should now be very close to that of the <= log
6116 line. There are two side-effects of this change:
6118 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6119 logged header lines no longer include the local Received: line, because
6120 it has not yet been created. The same applies to a copy of the message
6121 that is returned to a non-SMTP sender when a message is rejected.
6123 (b) When a filter file is tested using -bf, no additional Received: header
6124 is added to the test message. After some thought, I decided that this
6127 This change does not affect the value of $received_for. It is still set
6128 after address rewriting, but before local_scan() is called.
6130 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6132 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6133 gave an unhelpful panic error message, and a defer error. I have managed to
6134 change this behaviour so that it now rejects any supplied certificate,
6135 which seems right, as the list of acceptable certificates is empty.
6137 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6138 gave an unhelpful defer error. I have not managed to make this reject any
6139 supplied certificates, but the error message it gives is "no certificate
6140 supplied", which is not helpful.
6142 70. exigrep's output now also includes lines that are not associated with any
6143 message, but which match the given pattern. Implemented by a patch from
6144 Martin Sluka, which also tidied up the Perl a bit.
6146 71. Recipient callout verification, like sender verification, was using <> in
6147 the MAIL FROM command. This isn't really the right thing, since the actual
6148 sender may affect whether the remote host accepts the recipient or not. I
6149 have changed it to use the actual sender in the callout; this means that
6150 the cache record is now keyed on a recipient/sender pair, not just the
6151 recipient address. There doesn't seem to be a real danger of callout loops,
6152 since a callout by the remote host to check the sender would use <>.
6153 [SEE ABOVE: changed after hitting problems.]
6155 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6156 temporary errors. However, in the case of such a code being given after
6157 the end of a data transmission (i.e. after ".") Exim was failing to write
6158 a retry record for the message. (Yes, there was some broken host that was
6159 actually sending 8xx at this point.)
6161 73. An unknown lookup type in a host list could cause Exim to panic-die when
6162 the list was checked. (An example that provoked this was putting <; in the
6163 middle of a list instead of at the start.) If this happened during a DATA
6164 ACL check, a -D file could be left lying around. This kind of configuration
6165 error no longer causes Exim to die; instead it causes a defer error. The
6166 incident is still logged to the main and panic logs.
6168 74. Buglet left over from Exim 3 conversion. The message "too many messages
6169 in one connection" was written to the rejectlog but not the mainlog, except
6170 when address rewriting (yes!) was being logged.
6172 75. Added write_rejectlog option.
6174 76. When a system filter was run not as root (that is, when system_filter_user
6175 was set), the values of the $n variables were not being returned to the
6176 main process; thus, they were not subsequently available in the $sn
6179 77. Added +return_path_on_delivery log selector.
6181 78. A connection timeout was being treated differently from recipients deferred
6182 when testing hosts_max_try with a message that was older than the host's
6183 retry timeout. (The host should not be counted, thus allowing all hosts to
6184 be tried at least once before bouncing.) This may have been the cause of an
6185 occasionally reported bug whereby a message would remain on the queue
6186 longer than the retry timeout, but would be bounced if a delivery was
6187 forced. I say "may" because I never totally pinned down the problem;
6188 setting up timeout/retry tests is difficult. See also the next item.
6190 79. The ultimate address timeout was not being applied to errors that involved
6191 a combination of host plus message (for example, a timeout on a MAIL
6192 command). When an address resolved to a number of possible hosts, and they
6193 were not all tried for each delivery (e.g. because of hosts_max_try), a
6194 message could remain on the queue longer than the retry timeout.
6196 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6199 81. Fixed an obscure SMTP outgoing bug which required at least the following
6200 conditions: (a) there was another message waiting for the same server;
6201 (b) the server returned 5xx to all RCPT commands in the first message so
6202 that the message was not completed; (c) the server dropped the connection
6203 or gave a negative response to the RSET that Exim sends to abort the
6204 transaction. The observed case was a dropped connection after DATA that had
6205 been sent in pipelining mode. That is, the server had advertised PIPELINING
6206 but was not implementing it correctly. The effect of the bug was incorrect
6207 behaviour, such as trying another host, and this could lead to a crash.
6213 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6214 and daemon.c were passed as pointers to ints; they should have been
6215 pointers to socklen_t variables (which are typically unsigned ints).
6217 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6220 3. Fixed a really odd bug that affected only the testing scheme; patching a
6221 certain fixed string in the binary changed the value of another string that
6222 happened to be identical to the end of the original first string.
6224 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6225 name", it returns that address as the IP address. On some operating
6226 systems (e.g. Solaris), it also passes back the IP address string as the
6227 "host name". However, on others (e.g. Linux), it passes back an empty
6228 string. Exim wasn't checking for this, and was changing the host name to an
6229 empty string, assuming it had been canonicalized.
6231 5. Although rare, it is permitted to have more than one PTR record for a given
6232 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6233 all the names associated with an address, because they do in Solaris.
6234 However, it seems that they do not in Linux for data that comes from the
6235 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6236 I found this out when I moved to a new Linux workstation and tried to run
6237 the Exim test suite.
6239 To get round this problem I have changed the code so that it now does its
6240 own call to the DNS to look up PTR records when searching for a host name.
6241 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6242 addresses that are only in /etc/hosts are still found.
6244 This behaviour is, however, controlled by an option called host_lookup_
6245 order, which defaults to "bydns:byaddr". If people want to use the other
6246 order, or indeed, just use one or the other means of lookup, they can
6247 specify it in this variable.
6249 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6250 some operating systems, this comes back from gethostbyaddr() as an empty
6251 string, and this is what Exim used to test for. However, it seems that in
6252 other systems, "." is yielded. Exim now tests for this case too.
6254 7. The values of check_spool_space and check_log_space are now held internally
6255 as a number of kilobytes instead of an absolute number of bytes. If a
6256 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6257 kilobyte. This means that much larger values can be stored.
6259 8. Exim monitor: an attempt to get the action menu when not actually pointing
6260 at a message produces an empty menu entitled "No message selected". This
6261 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6262 no entries in it ("Shell widget menu has zero width and/or height"). So I
6263 have added a single, blank menu entry in this case.
6265 9. Added ${quote_local_part.
6267 10. MIME decoding is now applied to the contents of Subject: header lines when
6270 11. Now that a reference to $sender_host_address automatically causes a reverse
6271 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6272 host lookup before query-style lookups in lists that might use this
6273 variable. This has therefore been abolished, and the "net-" prefix is no
6274 longer necessary for query-style lookups.
6276 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6277 have been a typo for LFLAGS, so it has been changed.
6279 13. The install script calls Exim with "-C /dev/null" in order to find the
6280 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6281 to be output. However, since Exim outputs its version number before the
6282 error, it didn't break the script. It just looked ugly. I fixed this by
6283 always allowing "-C /dev/null" if the caller is root.
6285 14. Ignore overlarge ACL variable number when reading spool file - insurance
6286 against a later release with more variables having written the file.
6288 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6289 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6290 this, as well as the form without the "IPv6" on the front.
6292 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6293 OS/Makefile-Darwin file.
6295 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6296 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6298 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6299 variable. However, some OS use ints for inodes. I've added cast to long int
6300 to get rid of the compiler warning.
6302 19. I had forgotten to lock out "/../" in configuration file names when
6303 ALT_CONFIG_PREFIX was set.
6305 20. Routers used for verification do not need to specify transports. However,
6306 if such a router generated a host list, and callout was configured, Exim
6307 crashed, because it could not find a port number from the (non-existent)
6308 transport. It now assumes port 25 in this circumstance.
6310 21. Added the -t option to exigrep.
6312 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6313 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6314 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6315 them is set, LOOKUP_LSEARCH is forced.
6317 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6319 24. Added sender and host information to the "rejected by local_scan()" log
6320 line; previously there was no indication of these.
6322 25. Added .include_if_exists.
6324 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6325 Linux. It turns out that not all file systems support this. Apparently some
6326 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6327 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6328 not supported on the file descriptor, is now ignored when Exim is trying to
6329 sync a directory. This applies only to Linux.
6331 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6333 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6334 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6337 29. Added support for maildirsize files from supplied patch (modified a bit).
6339 30. The use of :fail: followed by an empty string could lead Exim to respond to
6340 sender verification failures with (e.g.):
6342 550 Verification failed for <xxx>
6343 550 Sender verify failed
6345 where the first response line was missing the '-' that indicates it is not
6346 the final line of the response.
6348 31. The loop for finding the name of the user that called Exim had a hardwired
6349 limit of 10; it now uses the value of finduser_retries, which is used for
6350 all other user lookups.
6352 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6355 33. Exim was neglecting to zero errno before one call of strtol() when
6356 expanding a string and expecting an integer value. On some systems this
6357 resulted in spurious "integer overflow" errors. Also, it was casting the
6358 result into an int without checking.
6360 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6361 did not work. The code looks as if it has *never* worked, though it appears
6362 to have been documented since at least release 1.62. I have made it work.
6364 35. The "timeout_DNS" error in retry rules, also documented since at least
6365 1.62, also never worked. As it isn't clear exactly what this means, and
6366 clearly it isn't a major issue, I have abolished the feature by treating it
6367 as "timeout", and writing a warning to the main and panic logs.
6369 36. The display of retry rules for -brt wasn't always showing the error code
6372 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6373 timeout_connect_A, timeout_connect_MX.
6375 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6376 to the empty sender.
6378 39. The daemon was not analysing the content of -oX till after it had closed
6379 stderr and disconnected from the controlling terminal. This meant that any
6380 syntax errors were only noted on the panic log, and the return code from
6381 the command was 0. By re-arranging the code a little, I've made the
6382 decoding happen first, so such errors now appear on stderr, and the return
6383 code is 1. However, the actual setting up of the sockets still happens in
6384 the disconnected process, so errors there are still only recorded on the
6387 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6388 connections (as happens on some IP stacks) was logged at start up time as
6389 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6390 it from "IPv6 and IPv4", which means that two separate sockets are being
6393 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6394 says whether AF_INET or AF_INET6 was passed as an argument.
6396 42. Exiwhat output was messed up when time zones were included in log
6399 43. Exiwhat now gives more information about the daemon's listening ports,
6400 and whether -tls-on-connect was used.
6402 44. The "port" option of the smtp transport is now expanded.
6404 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6405 silently ignored. Now an error message is written to the main and panic
6408 46. There's a new ACL modifier called "logwrite" which writes to a log file
6409 as soon as it is encountered.
6411 47. Added $local_user_uid and $local_user_gid at routing time.
6413 48. Exim crashed when trying to verify a sender address that was being
6416 49. Exim was recognizing only a space character after ".include". It now also
6417 recognizes a tab character.
6419 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6420 extracting the relevant information from the specification. The man page no
6421 longer contains scrambled data for the -d option, and I've added a section
6422 at the front about calling Exim under different names.
6424 51. Added "extra_headers" argument to the "mail" command in filter files.
6426 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6429 53. Installed eximstats 1.29.
6431 54. Added transport_filter_timeout as a generic transport option.
6433 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6434 Cc: header lines. This was required by RFC 822, but it not required by RFC
6437 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6438 incoming messages that did not have them. Now it does so only if the
6439 message originates locally, that is, if there is no associated remote host
6440 address. When Resent- header lines are present, this applies to the Resent-
6441 lines rather than the non-Resent- lines.
6443 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6444 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6446 58. Messages for configuration errors now include the name of the main
6447 configuration file - useful now that there may be more than one file in a
6448 list (.included file names were always shown).
6450 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6451 for those rare installations that do not start the daemon as root or run it
6452 setuid root. I've cut out the call to initgroups() if the daemon is not
6455 60. The Exim user and group can now be bound into the binary as text strings
6456 that are looked up at the start of Exim's processing.
6458 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6460 62. Added $mailstore_basename variable.
6462 63. Installed patch to sieve.c from Michael Haardt.
6464 64. When Exim failed to open the panic log after failing to open the main log,
6465 the original message it was trying to log was written to stderr and debug
6466 output, but if they were not available (the usual case in production), it
6467 was lost. Now it is written to syslog before the two lines that record the
6468 failures to open the logs.
6470 65. Users' Exim filters run in subprocesses under the user's uid. It is
6471 possible for a "deliver" command or an alias in a "personal" command to
6472 provoke an address rewrite. If logging of address rewriting is configured,
6473 this fails because the process is not running as root or exim. There may be
6474 a better way of dealing with this, but for the moment (because 4.30 needs
6475 to be released), I have disabled address rewrite logging when running a
6476 filter in a non-root, non-exim process.
6482 1. The buildconfig auxiliary program wasn't quoting the value set for
6483 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6484 not defined. This bug was masked in 4.22 by the effect that was fixed in
6487 2. Some messages that were rejected after a message id was allocated were
6488 shown as "incomplete" by exigrep. It no longer does this for messages that
6489 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6491 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6492 have allow_domain_literals set, the ID did not get logged in the <= line.
6493 Domain literals are now always recognized in Message-ID: header lines.
6495 4. The first argument for a ${extract expansion item is the key name or field
6496 number. Leading and trailing spaces in this item were not being ignored,
6497 causing some misleading effects.
6499 5. When deliver_drop_privilege was set, single queue runner processes started
6500 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6501 same command in the process it spins off) were not dropping privilege.
6503 6. When the daemon running as "exim" started a queue runner, it always
6504 re-executed Exim in the spun-off process. This is a waste of effort when
6505 deliver_drop_privilege is set. The new process now just calls the
6506 queue-runner function directly.
6512 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6515 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6516 ignored. Though the use of -oP was forcing the writing of a pid file, it
6517 was always written to the default place.
6519 3. If the message "no IP address found for host xxxx" is generated during
6520 incoming verification, it is now followed by identification of the incoming
6521 connection (so you can more easily find what provoked it).
6523 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6525 5. Added some features to "harden" Exim a bit more against certain attacks:
6527 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6528 be put in Local/Makefile. This is like the never_users runtime option,
6529 but it cannot be overridden. The default setting is "root".
6531 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6532 prefix string with which any file named in a -C command line option
6535 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6536 is retained for -C and -D only if the caller of Exim is root. Without
6537 it, the exim user may also use -C and -D and retain privilege.
6539 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6540 command line option is disabled.
6542 6. Macro names set by the -D option must start with an upper case letter, just
6543 like macro names defined in the configuration file.
6545 7. Added "dereference=" facility to LDAP.
6547 8. Two instances of the typo "uknown" in the source files are fixed.
6549 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6550 the Configure-Makefile script screwed up while processing it.
6552 10. Incorporated PCRE 4.4.
6554 11. The SMTP synchronization check was not operating right at the start of an
6555 SMTP session. For example, it could not catch a HELO sent before the client
6556 waited for the greeting. There is now a check for outstanding input at the
6557 point when the greeting is written. Because of the duplex, asynchronous
6558 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6559 way, but not yet received, when the check is performed.
6561 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6562 on TCP/IP sockets, because this apparently causes some broken clients to
6565 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6566 unchanged) from the Cygwin maintainer.
6568 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6569 was supported instead of testing for mbx. Effectively a typo.
6571 15. The spa authenticator server code was not checking that the input it
6572 received was valid base64.
6574 16. The debug output line for the "set" modifier in ACLs was not showing the
6575 name of the variable that was being set.
6577 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6579 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6580 lookup on its own. Something else was needed to trigger the lookup. For
6581 example, a match in host_lookup or the need for a host name in a host list.
6582 Now, if $sender_host_name is referenced and the host name has not yet been
6583 looked up, a lookup is performed. If the lookup fails, the variable remains
6584 empty, and $host_lookup_failed is set to "1".
6586 19. Added "eqi" as a case-independent comparison operator.
6588 20. The saslauthd authentication condition could segfault if neither service
6589 nor realm was specified.
6591 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6592 error message that was logged was misleading, and incoming SMTP
6593 connections were dropped. The message is now more accurate, and temporary
6594 errors are given to SMTP connections.
6596 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6597 (except RSET) with a 421 error, until QUIT is received. However, it was
6598 failing to send a response to QUIT.
6600 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6601 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6602 ACL is now run after the helo_try_verify_hosts code.
6604 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6605 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6606 case-independent, so other case variants are also recognized). Apparently
6607 some systems use these upper case variants.
6609 25. If more than two messages were waiting for the same host, and a transport
6610 filter was specified for the transport, Exim sent two messages over the
6611 same TCP/IP connection, and then failed with "socket operation on non-
6612 socket" when it tried to send the third.
6614 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6616 27. The extern definition of crypt16() in expand.c was not being excluded when
6617 the OS had its own crypt16() function.
6619 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6620 as a preferred synonym for return_size_limit, both as an option and as an
6623 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6625 30. Changed the default configuration ACL to relax the local part checking rule
6626 for addresses that are not in any local domains. For these addresses,
6627 slashes and pipe symbols are allowed within local parts, but the sequence
6628 /../ is explicitly forbidden.
6630 31. SPA server authentication was not clearing the challenge buffer before
6633 32. log_message in a "warn" ACL statement was writing to the reject log as
6634 well as to the main log, which contradicts the documentation and doesn't
6635 seem right (because no rejection is happening). So I have stopped it.
6637 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6638 However, I am unable to do any testing of this.
6640 34. Fixed an infelicity in the appendfile transport. When checking directories
6641 for a mailbox, to see if any needed to be created, it was accidentally
6642 using path names with one or more superfluous leading slashes; tracing
6643 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6645 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6646 contents are added to the log line that is written for every discarded
6647 recipient. (Previously a log_message setting was ignored.)
6649 36. The ${quote: operator now quotes the string if it is empty.
6651 37. The install script runs exim in order to find its version number. If for
6652 some reason other than non-existence or emptiness, which it checks, it
6653 could not run './exim', it was installing it with an empty version number,
6654 i.e. as "exim-". This error state is now caught, and the installation is
6657 38. An argument was missing from the function that creates an error message
6658 when Exim fails to connect to the socket for saslauthd authentication.
6659 This could cause Exim to crash, or give a corrupted message.
6661 39. Added isip, isip4, and isip6 to ${if conditions.
6663 40. The ACL variables $acl_xx are now saved with the message, and can be
6664 accessed later in routers, transports, and filters.
6666 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6667 strings in the file are not string-expanded.
6669 42. If a MAIL command specified a SIZE value that was too large to fit into an
6670 int variable, the check against message_size_limit failed. Such values are
6671 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6672 day this will have to be increased, but I don't think I want to be around
6673 when emails are that large.
6680 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6681 iconv() is not standard in FreeBSD.
6683 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6684 IPv6 enabled. The observed symptom was a segmentation fault on return from
6685 the function os_common_find_running_interfaces() in src/os.c.
6687 3. In the check_special_case() function in daemon.c I had used "errno" as an
6688 argument name, which causes warnings on some systems. This was basically a
6689 typo, since it was named "eno" in the comments!
6691 4. The code that waits for the clock to tick (at a resolution of some fraction
6692 of a second) so as to ensure message-id uniqueness was always waiting for
6693 at least one whole tick, when it could have waited for less. [This is
6694 almost certainly not relevant at current processor speeds, where it is
6695 unlikely to ever wait at all. But we try to future-proof.]
6697 5. The function that sleeps for a time interval that includes fractions of a
6698 second contained a race. It did not block SIGALRM between setting the
6699 timer, and suspending (a couple of lines later). If the interval was short
6700 and the sigsuspend() was delayed until after it had expired, the suspension
6701 never ended. On busy systems this could lead to processes getting stuck for
6704 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6705 process, before it forks any delivery processes. The open lookup caching
6706 mechanism meant that the open file or database connection was passed into
6707 the delivery process. The problem was that delivery processes always tidy
6708 up cached lookup data. This could cause a problem for the next delivery
6709 process started by the queue runner, because the external queue runner
6710 process does not know about the closure. So the next delivery process
6711 still has data in the lookup cache. In the case of a file lookup, there was
6712 no problem because closing a file descriptor in a subprocess doesn't affect
6713 the parent. However, if the lookup was caching a connection to a database,
6714 the connection was closed, and the second delivery process was likely to
6715 see errors such as "PGSQL: query failed: server closed the connection
6716 unexpectedly". The problem has been fixed by closing all cached lookups
6717 in a queue runner before running a delivery process.
6719 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6720 seem to have the "const" qualifier which it has on other OS. I've
6723 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6724 *of the same type* (client or server) with the same public name that an
6725 error should be diagnosed.
6727 9. When Exim looked up a host name for an IP address, but failed to find the
6728 original IP address when looking up the host name (a safety check), it
6729 output the message "<ip address> does not match any IP for NULL", which was
6730 confusing, to say the least. The bug was that the host name should have
6731 appeared instead of "NULL".
6733 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6734 user that is built into the binary, and the -C or -D options is used, root
6735 privilege is dropped before the configuration file is read. In addition,
6736 logging is switched to stderr instead of the normal log files. If the
6737 configuration then re-defines the Exim user, the unprivileged environment
6738 is probably not what is expected, so Exim logs a panic warning message (but
6741 However, if deliver_drop_privilege is set, the unprivileged state may well
6742 be exactly what is intended, so the warning has been cut out in that case,
6743 and Exim is allowed to try to write to its normal log files.
6749 1. smtp_return_error_details was not giving details for temporary sender
6750 or receiver verification errors.
6752 2. Diagnose a configuration error if two authenticators have the same public
6755 3. Exim used not to create the message log file for a message until the first
6756 delivery attempt. This could be confusing when incoming messages were held
6757 for policy or load reasons. The message log file is now created at the time
6758 the message is received, and an initial "Received" line is written to it.
6760 4. The automatically generated man page for command line options had a minor
6761 bug that caused no ill effects; however, a more serious problem was that
6762 the procedure for building the man page automatically didn't always
6763 operate. Consequently, release 4.20 contains an out-of-date version. This
6764 shouldn't happen again.
6766 5. When building Exim with embedded Perl support, the script that builds the
6767 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6768 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6770 6. The freeze_tell option was not being used for messages that were frozen on
6771 arrival, either by an ACL or by local_scan().
6773 7. Added the smtp_incomplete_transaction log selector.
6775 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6776 was accepting AUTH without a new EHLO.
6778 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6779 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6782 10. Logging of TCP/IP connections (when configured) now happens in the main
6783 daemon process instead of the child process, so that the TCP/IP connection
6784 count is more accurate (but it can never be perfect).
6786 11. The use of "drop" in a nested ACL was not being handled correctly in the
6787 outer ACL. Now, if condition failure induced by the nested "drop" causes
6788 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6789 or "require"), the connection is dropped.
6791 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6792 that yield "discard" can now be used with an "accept" or a "discard" verb,
6793 but an error is generated for any others (because I can't see a useful way
6794 to define what should happen).
6796 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6797 are now processed in the same way as lines in the Exim configuration file.
6798 In particular, continuation lines are supported.
6800 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6802 15. Added -ti meaning -t -i.
6804 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6805 domains, and warn by logging if others are found.
6807 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6808 returned from GIFCONF when Exim is trying to find the list of interfaces on
6809 a host. The code in os.c has been modified to copy each ifreq to an aligned
6810 structure in all cases.
6812 Also, in some cases, the returned ifreq's were being copied to a 'struct
6813 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6814 means the last couple of bytes of an IPv6 address could be chopped if the
6815 ifreq contained only a normal sockaddr (14 bytes storage).
6817 18. Named domain lists were not supported in the hosts_treat_as_local option.
6818 An entry such as +xxxx was not recognized, and was treated as a literal
6821 19. Ensure that header lines added by a DATA ACL are included in the reject log
6822 if the ACL subsequently rejects the message.
6824 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6825 MD5 (which is deprecated).
6827 21. When testing a filter file using -bf, Exim was writing a message when it
6828 took the sender from a "From " line in the message, but it was not doing so
6829 when it took $return_path from a Return-Path: header line. It now does.
6831 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6832 with a valid header line field name (a series of printing characters
6833 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6835 23. Changed "disc" in the source to "disk" to conform to the documentation and
6836 the book and for uniformity.
6838 24. Ignore Sendmail's -Ooption=value command line item.
6840 25. When execve() failed while trying to run a command in a pipe transport,
6841 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6842 could be confused with a return value of 69 from the command itself. This
6843 has been changed to 127, the value the shell returns if it is asked to run
6844 a non-existent command. The wording for the related log line suggests a
6845 non-existent command as the problem.
6847 26. If received_header_text expands to an empty string, do not add a Received:
6848 header line to the message. (Well, it adds a token one on the spool, but
6849 marks it "old" so that it doesn't get used or transmitted.)
6851 27. Installed eximstats 1.28 (addition of -nt option).
6853 28. There was no check for failure on the call to getsockname() in the daemon
6854 code. This can fail if there is a shortage of resources on the system, with
6855 ENOMEM, for example. A temporary error is now given on failure.
6857 29. Contrary to the C standard, it seems that in some environments, the
6858 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6859 program. Exim now does this explicitly; it affects the formatting of
6860 timestamps using strftime().
6862 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6863 complaints. I've now initialized all the variables, to avoid this.
6865 32. Header lines added by a system filter were not being "seen" during
6866 transport-time rewrites.
6868 33. The info_callback() function passed to OpenSSL is set up with type void
6869 (*)(SSL *, int, int), as described somewhere. However, when calling the
6870 function (actually a macro) that sets it up, the type void(*)() is
6871 expected. I've put in a cast to prevent warnings from picky compilers.
6873 34. If a DNS black list lookup found a CNAME record, but there were no A
6874 records associated with the domain it pointed at, Exim crashed.
6876 35. If a DNS black list lookup returned more than one A record, Exim ignored
6877 all but the first. It now scans all returned addresses if a particular IP
6878 value is being sought. In this situation, the contents of the
6879 $dnslist_value variable are a list of all the addresses, separated by a
6882 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6883 to accept a host name and all its aliases if the forward lookup for any of
6884 them yielded the IP address of the incoming connection. Now it accepts only
6885 those names whose forward lookup yields the correct IP address. Any other
6886 names are discarded. This closes a loophole whereby a rogue DNS
6887 administrator could create reverse DNS records to break through a
6888 wildcarded host restriction in an ACL.
6890 37. If a user filter or a system filter that ran in a subprocess used any of
6891 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6892 the wrong values were passed to the pipe command ($thisaddress had the
6893 value of $0, $0 had the value of $1, etc). This bug was introduced by
6894 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6896 38. Improved the line breaking for long SMTP error messages from ACLs.
6897 Previously, if there was no break point between 40 and 75 characters, Exim
6898 left the rest of the message alone. Two changes have been made: (a) I've
6899 reduced the minimum length to 35 characters; (b) if it can't find a break
6900 point between 35 and 75 characters, it looks ahead and uses the first one
6901 that it finds. This may give the occasional overlong line, but at least the
6902 remaining text gets split now.
6904 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6905 file descriptors might be low, and that setting 1000 would always raise it.
6906 It turns out that in some environments, the limit is already over 1000 and
6907 that lowering it causes trouble. So now Exim takes care not to decrease it.
6909 40. When delivering a message, the value of $return_path is set to $sender_
6910 address at the start of routing (routers may change the value). By an
6911 oversight, this default was not being set up when an address was tested by
6912 -bt or -bv, which affected the outcome if any router or filter referred to
6915 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6916 log only when "message" or "log_message" are set. However, if one of the
6917 conditions was an address verification, or a call to a nested ACL, the
6918 messages generated by the underlying test were being passed through. This
6919 no longer happens. The underlying message is available in $acl_verify_
6920 message for both "message" and "log_message" expansions, so it can be
6921 passed through if needed.
6923 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6924 new expansion $bh_ to give the encoded byte string without charset
6925 translation. Translation happens only if iconv() is available; HAVE_ICONV
6926 indicates this at build time. HEADERS_CHARSET gives the charset to
6927 translate to; headers_charset can change it in the configuration, and
6928 "headers charset" can change it in an individual filter file.
6930 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6931 that creates RFC 2047 encoded "words" labels them as that charset instead
6932 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6933 expansion operator; (ii) when Exim creates a From: line for a local
6934 message; (iii) when a header line is rewritten to include a "phrase" part.
6936 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6937 buggy, causing it to skip the first lines of messages whose message ID
6938 ended in 'D'. This would not have bitten before Exim release 4.14, because
6939 message IDs were unlikely to end in 'D' before then. The effect was to have
6940 incorrect size information for certain domains.
6942 45. #include "config.h" was missing at the start of the crypt16.c module. This
6943 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6946 46. If there was a timeout during a "random" callout check, Exim treated it as
6947 a failure of the random address, and carried on sending RSET and the real
6948 address. If the delay was just some slowness somewhere, the response to the
6949 original RCPT would be taken as a response to RSET and so on, causing
6950 mayhem of various kinds.
6952 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6953 when I implemented it. It didn't allow for the fact that some option values
6954 may legitimately be negative (e.g. size_addition), and it didn't even do
6955 the right test for positive values.
6957 48. Domain names in DNS records are case-independent. Exim always looks them up
6958 in lower case. Some resolvers return domain names in exactly the case they
6959 appear in the zone file, that is, they may contain uppercase letters. Not
6960 all resolvers do this - some return always lower case. Exim was treating a
6961 change of case by a resolver as a change of domain, similar to a widening
6962 of a domain abbreviation. This triggered its re-routing code and so it was
6963 trying to route what was effectively the same domain again. This normally
6964 caused routing to fail (because the router wouldn't handle the domain
6965 twice). Now Exim checks for this case specially, and just changes the
6966 casing of the domain that it ultimately uses when it transmits the message
6969 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6972 50. If a filter generated a file delivery with a non-absolute name (possible if
6973 no home directory exists for the router), the forbid_file option was not
6976 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6977 the existing equality matching.
6979 52. Exim was using ints instead of ino_t variables in some places where it was
6980 dealing with inode numbers.
6982 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6983 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6984 called TMPDIR, and if it finds it is different, it changes its value.
6986 54. The smtp_printf() function is now made available to local_scan() so
6987 additional output lines can be written before returning. There is also an
6988 smtp_fflush() function to enable the detection of a dropped connection.
6989 The variables smtp_input and smtp_batched_input are exported to
6992 55. Changed the default runtime configuration: the message "Unknown user"
6993 has been removed from the ACL, and instead placed on the localuser router,
6994 using the cannot_route_message feature. This means that any verification
6995 failures that generate their own messages won't get overridden. Similarly,
6996 the "Unrouteable address" message that was in the ACL for unverifiable
6997 relay addresses has also been removed.
6999 56. Added hosts_avoid_esmtp to the smtp transport.
7001 57. The exicyclog script was not checking for the esoteric option
7002 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
7003 will work only if exicyclog is run under the appropriate euid.
7005 58. Following a discussion on the list, the rules by which Exim recognises line
7006 endings on incoming messages have been changed. The -dropcr and drop_cr
7007 options are now no-ops, retained only for backwards compatibility. The
7008 following line terminators are recognized: LF CRLF CR. However, special
7009 processing applies to CR:
7011 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7012 nor a local message in the state where . is a terminator.
7014 (ii) If a bare CR is encountered in a header line, an extra space is added
7015 after the line terminator so as not to end the header. The reasoning
7016 behind this is that bare CRs in header lines are most likely either
7017 to be mistakes, or people trying to play silly games.
7019 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7020 was being incorrectly given as 18 bytes larger than it should have been.
7021 This is a VOB (very old bug).
7023 60. This may never have affected anything current, but just in case it has:
7024 When the local host is found other than at the start of a list of hosts,
7025 the local host, those with the same MX, and any that follow, are discarded.
7026 When the list in question was part of a longer list of hosts, the following
7027 hosts (not currently being processed) were also being discarded. This no
7028 longer happens. I'm not sure if this situation could ever has previously
7031 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7034 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7035 header. This is something that is recommended in a new Internet Draft, and
7036 is something that is documented as being done by Sendmail. There are two
7037 possible values. For messages generated by the autoreply transport, Exim
7040 Auto-Submitted: auto-replied
7042 whereas for all other generated messages (e.g. bounces) it adds
7044 Auto-Submitted: auto-generated
7046 63. The "personal" condition in filters now includes a test for the
7047 Auto-Submitted: header. If it contains the string "auto-" the message it
7048 not considered personal.
7050 64. Added rcpt_include_affixes as a generic transport option.
7052 65. Added queue_only_override (default true).
7054 66. Added the syslog_duplication option.
7056 67. If what should have been the first header line of a message consisted of
7057 a space followed by a colon, Exim was mis-interpreting it as a header line.
7058 It isn't of course - it is syntactically invalid and should therefore be
7059 treated as the start of the message body. The misbehaviour could have
7060 caused a number of strange effects, including loss of data in subsequent
7061 header lines, and spool format errors.
7063 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7064 client host had authenticated. This control can now be exercised by an ACL
7065 for more flexibility.
7067 69. By default, callouts do not happen when testing with -bh. There is now a
7068 variant, -bhc, which does actually run the callout code, including
7069 consulting and updating the callout cache.
7071 70. Added support for saslauthd authentication, courtesy of Alexander
7074 71. If statvfs() failed on the spool or log directories while checking their
7075 size for availability, Exim confusingly gave the error "space shortage".
7076 Furthermore, in debugging mode it crashed with a floating point exception.
7077 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7078 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7079 serious problem, Exim now writes to the main and panic logs when this
7080 happens, with details of the failure. It then refuses to accept the
7081 incoming message, giving the message "spool directory problem" or "log
7082 directory problem" with a 421 code for SMTP messages.
7084 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7085 0, 1, and 2 exist, because some OS complain for execs without them (see
7086 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7087 descriptors. However, the code omitted to check that the open succeeded,
7088 causing mysterious errors if for some reason the permissions on /dev/null
7089 got screwed. Now Exim writes a message to the main and panic logs, and
7090 bombs out if it can't open /dev/null.
7092 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7093 interact so that it is all more flexible. It is supposed to remain
7094 backwards compatible. Also added extra_local_interfaces.
7096 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7097 to bomb out with an assertion failure - to the client this appears as a
7098 connection drop. This problem occurs in the part of the code that was taken
7099 from the Samba project. Fortunately, the assertion is in a very simple
7100 function, so I have fixed this by reproducing the function inline in the
7101 one place where it is called, and arranging for authentication to fail
7102 instead of killing the process with assert().
7104 75. The SPA client code was not working when the server requested OEM rather
7105 than Unicode encoding.
7107 76. Added code to make require_files with a specific uid setting more usable in
7108 the case where statting the file as root fails - usually a non-root-mounted
7109 NFS file system. When this happens and the failure is EACCES, Exim now
7110 forks a subprocess and does the per-uid checking as the relevant uid.
7112 77. Added process_log_path.
7114 78. If log_file_path was not explicitly set, a setting of check_log_space or
7115 check_log_inodes was ignored.
7117 79. If a space check for the spool or log partitions fails, the incident is now
7118 logged. Of course, in the latter case the data may get lost...
7120 80. Added the %p formatting code to string_format() so that it can be used to
7121 print addresses in debug_print(). Adjusted all the address printing in the
7122 debugging in store.c to use %p rather than %d.
7124 81. There was a concern that a line of code in smtp_in.c could overflow a
7125 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7126 initially expressed, the concern was not well-founded, because trailing
7127 spaces are removed early. However, if the trailing spaces were followed by
7128 a NULL, they did not get removed, so the overflow was possible. Two fixes
7131 (a) I re-wrote the offending code in a cleaner fashion.
7132 (b) If an incoming SMTP command contains a NULL character, it is rejected
7135 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7136 runs initgroups(), so that if the Exim user is in any additional groups,
7137 they will be used during message reception.
7143 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob