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.
184 Extend the restriction to also cover ARC-signing.
186 JH/35 Cutthrough: for a final-dot response timeout (and nonunderstood responses)
187 in defer=pass mode supply a 450 to the initiator. Previously the message
194 JH/01 Rework error string handling in TLS interface so that the caller in
195 more cases is responsible for logging. This permits library-sourced
196 string to be attached to addresses during delivery, and collapses
197 pairs of long lines into single ones.
199 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
200 during configuration. Wildcards are allowed and expanded.
202 JH/02 Rework error string handling in DKIM to pass more info back to callers.
203 This permits better logging.
205 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
206 do not close it down and have the child transport start it up again on
207 the passed-on TCP connection. Instead, proxy the child (and any
208 subsequent ones) for TLS via a unix-domain socket channel. Logging is
209 affected: the continued delivery log lines do not have any DNSSEC, TLS
210 Certificate or OCSP information. TLS cipher information is still logged.
212 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
213 identical IP addresses on different listening ports. Will also affect
216 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
217 add noisy ifdef guards to special-case this sillyness.
218 Patch from Bernd Kuhls.
220 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
221 than 255 are no longer allowed.
223 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
224 Disable the session-cache too, which might reduce our load. Since we
225 currrectly use a new context for every connection, both as server and
226 client, there is no benefit for these.
227 GnuTLS appears to not support tickets server-side by default (we don't
228 call gnutls_session_ticket_enable_server()) but client side is enabled
229 by default on recent versions (3.1.3 +) unless the PFS priority string
232 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
233 <https://reproducible-builds.org/specs/source-date-epoch/>.
235 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
236 the check for any unsuccessful recipients did not notice the limit, and
237 erroneously found still-pending ones.
239 JH/08 Pipeline CHUNKING command and data together, on kernels that support
240 MSG_MORE. Only in-clear (not on TLS connections).
242 JH/09 Avoid using a temporary file during transport using dkim. Unless a
243 transport-filter is involved we can buffer the headers in memory for
244 creating the signature, and read the spool data file once for the
245 signature and again for transmission.
247 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
248 4.77 as the kernel support then wasn't solid, having issues in 64bit
249 mode. Now, it's been long enough. Add support for FreeBSD also.
251 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
252 case where the routing stage had gathered several addresses to send to
253 a host before calling the transport for the first, we previously failed
254 to close down TLS in the old transport process before passing the TCP
255 connection to the new process. The new one sent a STARTTLS command
256 which naturally failed, giving a failed delivery and bloating the retry
257 database. Investigation and fix prototype from Wolfgang Breyha.
259 JH/12 Fix check on SMTP command input synchronisation. Previously there were
260 false-negatives in the check that the sender had not preempted a response
261 or prompt from Exim (running as a server), due to that code's lack of
262 awareness of the SMTP input buffering.
264 PP/04 Add commandline_checks_require_admin option.
265 Exim drops privileges sanely, various checks such as -be aren't a
266 security problem, as long as you trust local users with access to their
267 own account. When invoked by services which pass untrusted data to
268 Exim, this might be an issue. Set this option in main configuration
269 AND make fixes to the calling application, such as using `--` to stop
272 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
273 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
274 responses to those, into a single TLS record each way (this usually means
275 a single packet). As a side issue, smtp_enforce_sync now works on TLS
278 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
279 affects you only if you're dancing at the edge of the param size limits.
280 If you are, and this message makes sense to you, then: raise the
281 configured limit or use OpenSSL 1.1. Nothing we can do for older
284 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
285 cmdline element to get the documented default one. Previously it was
288 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
289 are unsuable for non-message-specific reasons - eg. connection timeout,
290 banner-time rejection.
292 JH/16 Fix logging of delivery remote port, when specified by router, under
295 PP/06 Repair manualroute's ability to take options in any order, even if one
296 is the name of a transport.
299 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
301 JH/17 Change the list-building routines interface to use the expanding-string
302 triplet model, for better allocation and copying behaviour.
304 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
305 Previously it was constructed the first time a possibly-matching string
306 was met in the configuration file input during startup; now it is done
309 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
310 compatible one, to avoid the (poorly documented) possibility of a config
311 file in the working directory redirecting the DB files, possibly correpting
312 some existing file. CVE-2017-10140 assigned for BDB.
314 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
315 cache-hot. Previously, although the result was properly cached, the
316 initial verify call returned a defer.
318 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
319 the main verify for receipient in uncached-mode.
321 JH/22 Retire historical build files to an "unsupported" subdir. These are
322 defined as "ones for which we have no current evidence of testing".
324 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
325 if present. Previously it was ignored.
327 JH/24 Start using specified-initialisers in C structure init coding. This is
328 a C99 feature (it's 2017, so now considered safe).
330 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
331 if was a fixed-sized field and bitmask ops via macros; it is now more
334 PP/07 GitHub PR 56: Apply MariaDB build fix.
335 Patch provided by Jaroslav Škarvada.
337 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
338 during Coverity cleanups [4.87 JH/47]
339 Diagnosis and fix provided by Michael Fischer v. Mollard.
341 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
342 the right size to place the terminating semicolon on its own folded
343 line, the header hash was calculated to an incorrect value thanks to
344 the (relaxed) space the fold became.
346 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
347 and confused the parent.
349 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
350 which could crash as a result. This could lead to undeliverable messages.
352 JH/28 Logging: "next input sent too soon" now shows where input was truncated
355 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
356 matters on fast-turnover and PID-randomising systems, which were getting
357 out-of-order delivery.
359 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
360 a possibly-overlapping copy. The symptom was that "Remote host closed
361 connection in response to HELO" was logged instead of the actual 4xx
364 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
365 Previously only that bufferd was discarded, resulting in SYMTP command
368 JH/32 DKIM: when a message has multiple signatures matching an identity given
369 in dkim_verify_signers, run the dkim acl once for each. Previously only
370 one run was done. Bug 2189.
372 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
373 "panic the current process" to "deliberately defer". The panic log is
374 still written with the problem list name; the mail and reject logs now
375 get a temp-reject line for the message that was being handled, saying
376 something like "domains check lookup or other defer". The SMTP 451
377 message is still "Temporary local problem".
379 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
380 A crafted sequence of BDAT commands could result in in-use memory beeing
381 freed. CVE-2017-16943.
383 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
384 from SMTP input. Previously it was always done; now only done for DATA
385 and not BDAT commands. CVE-2017-16944.
387 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
388 to the message (such as an overlong header line). Previously this was
389 not done and we did not exit BDAT mode. Followon from the previous item
390 though a different problem.
396 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
397 than -2003 did; needs libidn2 in addition to libidn.
399 JH/02 The path option on a pipe transport is now expanded before use.
401 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
402 Patch provided by "Björn", documentation fix added too.
404 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
405 missing a wire-to-host endian conversion.
407 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
408 close after a BDAT command line could be taken as a following command,
409 giving a synch failure. Fix by only checking for synch immediately
410 before acknowledging the chunk.
412 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
413 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
414 macro. Patches provided by Josh Soref.
416 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
417 Previously we did not; the RFC seems ambiguous and VRFY is not listed
418 by IANA as a service extension. However, John Klensin suggests that we
421 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
422 the dkim code may be unix-mode line endings rather than smtp wire-format
423 CRLF, so prepend a CR to any bare LF.
425 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
426 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
428 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
429 the first were themselves being wrongly included in the feed into dkim
430 processing; with most chunk sizes in use this resulted in an incorrect
431 body hash calculated value.
433 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
434 DKIM signature block, for verification. Although advised against by
435 standards it is specifically not ruled illegal.
437 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
439 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
440 missing a body hash (the bh= tag).
442 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
443 It seems that HAProxy sends the Proxy Protocol information in clear and
444 only then does a TLS startup, so do the same.
446 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
447 TCP connections (such as for Spamd) unless the daemon successfully set
448 Fast Open mode on its listening sockets. This fixes breakage seen on
449 too-old kernels or those not configured for Fast Open, at the cost of
450 requiring both directions being enabled for TFO, and TFO never being used
451 by non-daemon-related Exim processes.
453 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
454 endings, at least on the first header line. Try to canonify any that get
455 past that check, despite the cost.
457 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
458 now limited to an arbitrary five deep, while parsing addresses with the
459 strip_excess_angle_brackets option enabled.
461 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
462 instead leave the unprompted TLS handshake in socket buffer for the
463 TLS library to consume.
465 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
467 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
469 JH/16 Drop variables when they go out of scope. Memory management drops a whole
470 region in one operation, for speed, and this leaves assigned pointers
471 dangling. Add checks run only under the testsuite which checks all
472 variables at a store-reset and panics on a dangling pointer; add code
473 explicitly nulling out all the variables discovered. Fixes one known
474 bug: a transport crash, where a dangling pointer for $sending_ip_address
475 originally assigned in a verify callout, is re-used.
477 PP/06 Drop '.' from @INC in various Perl scripts.
479 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
481 PP/08 Reduce a number of compilation warnings under clang; building with
482 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
483 should be warning-free.
485 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
487 HS/01 Fix portability problems introduced by PP/08 for platforms where
488 realloc(NULL) is not equivalent to malloc() [SunOS et al].
490 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
491 chunk. This allows us to accept broken chunked messages. We need a more
492 general solution here.
494 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
495 already-broken messages in the queue.
497 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
499 JH/19 Fix reference counting bug in routing-generated-address tracking.
505 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
506 supports it and a size is available (ie. the sending peer gave us one).
508 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
509 years of being deprecated). The replacements are the ACLs
510 acl_smtp_mime and acl_not_smtp_mime.
512 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
513 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
514 or even an in-clear connection were permitted. Now, if the host lookup
515 was dnssec and dane was requested then the host is only used if the
516 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
517 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
518 if one fails this test.
519 This means that a poorly-configured remote DNS will make it incommunicado;
520 but it protects against a DNS-interception attack on it.
522 JH/04 Bug 1810: make continued-use of an open smtp transport connection
523 non-noisy when a race steals the message being considered.
525 JH/05 If main configuration option tls_certificate is unset, generate a
526 self-signed certificate for inbound TLS connections.
528 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
529 in rewrites and routers.
531 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
532 and logged a warning sing 4.83; now they are a configuration file error.
534 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
535 (lacking @domain). Apply the same qualification processing as RCPT.
537 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
539 JH/10 Support ${sha256:} applied to a string (as well as the previous
542 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
543 a cutthrough deliver is pending, as we always want to make a connection.
544 This also avoids re-routing the message when later placing the cutthrough
545 connection after a verify cache hit.
546 Do not update it with the verify result either.
548 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
549 when routing results in more than one destination address.
551 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
552 signing (which inhibits the cutthrough capability). Previously only
553 the presence of an option was tested; now an expansion evaluating as
554 empty is permissible (obviously it should depend only on data available
555 when the cutthrough connection is made).
557 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
558 the relevant preceding SMTP command did not note the pipelining mode.
560 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
561 Previously they were not counted.
563 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
564 as one having no matching records. Previously we deferred the message
565 that needed the lookup.
567 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
568 distinguished as "(=".
570 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
571 for missing MX records. Previously it only worked for missing A records.
573 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
575 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
576 after the data-go-ahead and data-ack. Patch from Jason Betts.
578 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
579 even for a "none" policy. Patch from Tony Meyer.
581 JH/22 Fix continued use of a connection for further deliveries. If a port was
582 specified by a router, it must also match for the delivery to be
585 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
586 When one of the recipients of a message was unsuitable for the connection
587 (has no matching addresses), we lost track of needing to mark it
588 deferred. As a result mail would be lost.
590 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
592 JH/25 Decoding ACL controls is now done using a binary search; the source code
593 takes up less space and should be simpler to maintain. Merge the ACL
594 condition decode tables also, with similar effect.
596 JH/26 Fix problem with one_time used on a redirect router which returned the
597 parent address unchanged. A retry would see the parent address marked as
598 delivered, so not attempt the (identical) child. As a result mail would
601 JH/27 Fix a possible security hole, wherein a process operating with the Exim
602 UID can gain a root shell. Credit to http://www.halfdog.net/ for
603 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
606 JH/28 Enable {spool,log} filesystem space and inode checks as default.
607 Main config options check_{log,spool}_{inodes,space} are now
608 100 inodes, 10MB unless set otherwise in the configuration.
610 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
611 Previously it only applied to the main-section connection policy
614 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
616 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
617 by me. Added RFC7919 DH primes as an alternative.
619 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
620 are not in the system include path.
622 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
623 GnuTLS, when a session startup failed (eg because the client disconnected)
624 Exim did stdio operations after fclose. This was exposed by a recent
625 change which nulled out the file handle after the fclose.
627 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
628 signed directly by the cert-signing cert, rather than an intermediate
629 OCSP-signing cert. This is the model used by LetsEncrypt.
631 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
633 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
634 an incoming connection.
636 HS/02 Bug 1802: Do not half-close the connection after sending a request
639 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
640 fallback to "prime256v1".
642 JH/34 SECURITY: Use proper copy of DATA command in error message.
643 Could leak key material. Remotely exploitable. CVE-2016-9963.
649 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
650 and 3.4.4 - once the server is enabled to respond to an OCSP request
651 it does even when not requested, resulting in a stapling non-aware
652 client dropping the TLS connection.
654 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
655 support variable-length bit vectors. No functional change.
657 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
658 The I= interface field on outgoing lines is now after the H= remote
659 host field, same as incoming lines. There is a separate
660 outgoing_interface log selector which allows you to disable the
663 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
664 If not running log_selector +smtp_connection the mainlog would be held
665 open indefinitely after a "too many connections" event, including to a
666 deleted file after a log rotate. Leave the per net connection logging
667 leaving it open for efficiency as that will be quickly detected by the
668 check on the next write.
670 HS/01 Bug 1671: Fix post transport crash.
671 Processing the wait-<transport> messages could crash the delivery
672 process if the message IDs didn't exist for some reason. When
673 using 'split_spool_directory=yes' the construction of the spool
674 file name failed already, exposing the same netto behaviour.
676 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
677 mime_regex ACL conditions.
679 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
680 to DSN fail messages (bounces): remote IP, remote greeting, remote response
681 to HELO, local diagnostic string.
683 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
684 log line to debug. Even when configured with a tls authenticator many
685 client connections are expected to not authenticate in this way, so
686 an authenticate fail is not an error.
688 HS/02 Add the Exim version string to the process info. This way exiwhat
689 gives some more detail about the running daemon.
691 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
692 matter for fast-change records such as DNSBLs.
694 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
695 retry record, even if constant. There may be multiple transports with
696 different interface settings and the retry behaviour needs to be kept
699 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
701 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
703 JH/10 Bug 840: fix log_defer_output option of pipe transport
705 JH/11 Bug 830: use same host for all RCPTS of a message, even under
706 hosts_randomize. This matters a lot when combined with mua_wrapper.
708 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
709 ${quote_pgsql:<string>} operator.
711 JH/13 Bug 1708: avoid misaligned access in cached lookup.
713 JH/14 Change header file name for freeradius-client. Relevant if compiling
714 with Radius support; from the Gentoo tree and checked under Fedora.
716 JH/15 Bug 1712: Introduce $prdr_requested flag variable
718 JH/16 Bug 1714: Permit an empty string as expansion result for transport
719 option transport_filter, meaning no filtering.
721 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
723 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
724 defaults to "*" (all hosts). The variable is now available when not built
725 with TLS, default unset, mainly to enable keeping the testsuite sane.
726 If a server certificate is not supplied (via tls_certificate) an error is
727 logged, and clients will find TLS connections fail on startup. Presumably
728 they will retry in-clear.
729 Packagers of Exim are strongly encouraged to create a server certificate
730 at installation time.
732 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
733 with the $config_file variable.
735 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
736 in transport context, after the attempt, and per-recipient. The latter type
737 is per host attempted. The event data is the error message, and the errno
738 information encodes the lookup type (A vs. MX) used for the (first) host,
739 and the trailing two digits of the smtp 4xx response.
741 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
742 to write to mainlog (or rejectlog, paniclog) in the window between file
743 creation and permissions/ownership being changed. Particularly affects
744 installations where exicyclog is run as root, rather than exim user;
745 result is that the running daemon panics and dies.
747 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
749 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
750 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
751 "pri" and "weight". Note that the previous implicit priority given by the
752 list order is no longer honoured.
754 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
757 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
758 by defining SUPPORT_SOCKS.
760 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
761 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
762 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
763 variables are renamed to proxy_{local,external}_{address,port}.
765 JH/27 Move Internationalisation support from Experimental to mainline, enabled
766 for a build by defining SUPPORT_I18N
768 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
769 of the query string, and make ${quote_redis:} do that quoting.
771 JH/29 Move Events support from Experimental to mainline, enabled by default
772 and removable for a build by defining DISABLE_EVENT.
774 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
776 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
777 cached by the daemon.
779 JH/32 Move Redis support from Experimental to mainline, enabled for a build
780 by defining LOOKUP_REDIS. The libhiredis library is required.
782 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
783 keys are given for lookup.
785 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
786 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
787 only supported when built with TLS support. The PolarSSL SHA routines
788 are still used when the TLS library is too old for convenient support.
790 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
791 openssl_options), for security. OpenSSL forces this from version 1.1.0
792 server-side so match that on older versions.
794 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
795 allocation for $value could be released as the expansion processing
796 concluded, but leaving the global pointer active for it.
798 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
799 and to use the domains and local_parts ACL conditions.
801 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
802 incorrectly not doubled on cutthrough transmission, hence seen as a
803 body-termination at the receiving system - resulting in truncated mails.
804 Commonly the sender saw a TCP-level error, and retransmitted the message
805 via the normal store-and-forward channel. This could result in duplicates
806 received - but deduplicating mailstores were liable to retain only the
807 initial truncated version.
809 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
811 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
813 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
814 we're in there, support oversigning also; bug 1309.
816 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
818 HS/04 Add support for keep_environment and add_environment options.
820 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
821 either intentional arithmetic overflow during PRNG, or testing config-
824 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
825 delivery resulted in actual delivery. Cancel cutthrough before DATA
828 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
829 rejects a recipient: pass the reject to the originator.
831 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
832 Many were false-positives and ignorable, but it's worth fixing the
835 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
836 for the new environment-manipulation done at startup. Move the routines
837 from being local to tls.c to being global via the os.c file.
839 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
840 an extract embedded as result-arg for a map, the first arg for extract
841 is unavailable so we cannot tell if this is a numbered or keyed
842 extraction. Accept either.
848 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
851 JH/02 The smtp transport option "multi_domain" is now expanded.
853 JH/03 The smtp transport now requests PRDR by default, if the server offers
856 JH/04 Certificate name checking on server certificates, when exim is a client,
857 is now done by default. The transport option tls_verify_cert_hostnames
858 can be used to disable this per-host. The build option
859 EXPERIMENTAL_CERTNAMES is withdrawn.
861 JH/05 The value of the tls_verify_certificates smtp transport and main options
862 default to the word "system" to access the system default CA bundle.
863 For GnuTLS, only version 3.0.20 or later.
865 JH/06 Verification of the server certificate for a TLS connection is now tried
866 (but not required) by default. The verification status is now logged by
867 default, for both outbound TLS and client-certificate supplying inbound
870 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
873 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
874 Status Notification (bounce) messages are now MIME format per RFC 3464.
875 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
876 under the control of the dsn_advertise_hosts option, and routers may
877 have a dsn_lasthop option.
879 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
880 default, modifiable by a malware= option. The list separator for
881 the options can now be changed in the usual way. Bug 68.
883 JH/10 The smtp_receive_timeout main option is now expanded before use.
885 JH/11 The incoming_interface log option now also enables logging of the
886 local interface on delivery outgoing connections.
888 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
889 if the interface and destination host and port all match.
891 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
894 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
895 Patch from Andrew Lewis.
897 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
898 now supports optional time-restrictions, weighting, and priority
899 modifiers per server. Patch originally by <rommer@active.by>.
901 JH/16 The spamd_address main option now supports a mixed list of local
902 and remote servers. Remote servers can be IPv6 addresses, and
903 specify a port-range.
905 JH/17 Bug 68: The spamd_address main option now supports an optional
906 timeout value per server.
908 JH/18 Bug 1581: Router and transport options headers_add/remove can
909 now have the list separator specified.
911 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
914 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
917 JH/21 Support for the A6 type of dns record is withdrawn.
919 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
920 rather than the verbs used.
922 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
923 from 255 to 1024 chars.
925 JH/24 Verification callouts now attempt to use TLS by default.
927 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
928 are generic router options now. The defaults didn't change.
930 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
931 Original patch from Alexander Shikoff, worked over by JH.
933 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
934 files if ZCAT_COMMAND is not executable.
936 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
938 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
940 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
941 Normally benign, it bites when the pair was led to by a CNAME;
942 modern usage is to not canonicalize the domain to a CNAME target
943 (and we were inconsistent anyway for A-only vs AAAA+A).
945 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
947 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
948 when evaluating $sender_host_dnssec.
950 JH/31 Check the HELO verification lookup for DNSSEC, adding new
951 $sender_helo_dnssec variable.
953 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
955 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
957 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
959 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
960 documented as working, but never had. Support all but $spam_report.
962 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
963 added for tls authenticator.
965 HS/03 Add perl_taintmode main config option
971 TL/01 When running the test suite, the README says that variables such as
972 no_msglog_check are global and can be placed anywhere in a specific
973 test's script, however it was observed that placement needed to be near
974 the beginning for it to behave that way. Changed the runtest perl
975 script to read through the entire script once to detect and set these
976 variables, reset to the beginning of the script, and then run through
977 the script parsing/test process like normal.
979 TL/02 The BSD's have an arc4random API. One of the functions to induce
980 adding randomness was arc4random_stir(), but it has been removed in
981 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
982 function when detected.
984 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
985 cause callback expansion.
987 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
988 syntax errors in an expansion can be treated as a string instead of
989 logging or causing an error, due to the internal use of bool_lax
990 instead of bool when processing it.
992 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
993 server certificates when making smtp deliveries.
995 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
997 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
999 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
1001 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
1002 Merged patch from Sebastian Wiedenroth.
1004 JH/05 Fix results-pipe from transport process. Several recipients, combined
1005 with certificate use, exposed issues where response data items split
1006 over buffer boundaries were not parsed properly. This eventually
1007 resulted in duplicates being sent. This issue only became common enough
1008 to notice due to the introduction of connection certificate information,
1009 the item size being so much larger. Found and fixed by Wolfgang Breyha.
1011 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
1012 size buffer was used, resulting in syntax errors when an expansion
1015 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1016 version 3.3.6 or later.
1018 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1019 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1020 both become "event_action", the variables become $event_name, $event_data
1021 and $event_defer_errno. There is a new variable $verify_mode, usable in
1022 routers, transports and related events. The tls:cert event is now also
1023 raised for inbound connections, if the main configuration event_action
1026 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1027 early OCSP support, but no stapling (appears to be less than 1.0.0).
1029 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1030 server certificate names available under the smtp transport option
1031 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1034 JH/10 Time-related extraction expansions from certificates now use the main
1035 option "timezone" setting for output formatting, and are consistent
1036 between OpenSSL and GnuTLS compilations. Bug 1541.
1038 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1039 encoded parameter in the incoming message. Bug 1558.
1041 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1042 include certificate info, eximon was claiming there were spoolfile
1045 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1047 JH/14 Log delivery-related information more consistently, using the sequence
1048 "H=<name> [<ip>]" wherever possible.
1050 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1051 are problematic for Debian distribution, omit them from the release
1054 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1056 JH/16 Fix string representation of time values on 64bit time_t architectures.
1059 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1060 output list separator was used.
1065 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1066 checkers that were complaining about end of non-void function with no
1069 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1070 This was a regression introduced in 4.83 by another bugfix.
1072 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1074 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1075 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1081 TF/01 Correctly close the server side of TLS when forking for delivery.
1083 When a message was received over SMTP with TLS, Exim failed to clear up
1084 the incoming connection properly after forking off the child process to
1085 deliver the message. In some situations the subsequent outgoing
1086 delivery connection happened to have the same fd number as the incoming
1087 connection previously had. Exim would try to use TLS and fail, logging
1088 a "Bad file descriptor" error.
1090 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1091 utilities have not been installed.
1093 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1094 temporary space as the ACL may create new global variables.
1096 TL/01 LDAP support uses per connection or global context settings, depending
1097 upon the detected version of the libraries at build time.
1099 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1100 to extract and use the src ip:port in logging and expansions as if it
1101 were a direct connection from the outside internet. PPv2 support was
1102 updated based on HAProxy spec change in May 2014.
1104 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1106 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1107 Properly escape header and check for NULL return.
1109 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1112 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1114 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1115 characters in header names, implemented as a verify condition.
1116 Contributed by Michael Fischer v. Mollard.
1118 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1119 results permerror and temperror. Previous values are deprecated but
1120 still accepted. In a future release, err_perm and err_temp will be
1121 completely removed, which will be a backward incompatibility if the
1122 ACL tests for either of these two old results. Patch contributed by
1123 user bes-internal on the mailing list.
1125 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1127 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1128 selectors, in both main and reject logs.
1130 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1133 JH/07 Add malware type "sock" for talking to simple daemon.
1135 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1137 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1138 routers/transports under cutthrough routing.
1140 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1141 numbers. Touch up "bool" conditional to keep the same definition.
1143 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1145 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1147 JH/12 Expand items in router/transport headers_add or headers_remove lists
1148 individually rather than the list as a whole. Bug 1452.
1150 Required for reasonable handling of multiple headers_ options when
1151 they may be empty; requires that headers_remove items with embedded
1152 colons must have them doubled (or the list-separator changed).
1154 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1155 view the policy declared in the DMARC record. Currently, $dmarc_status
1156 is a combined value of both the record presence and the result of the
1159 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1161 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1162 dnslookup router and the smtp transport (applying to the forward
1165 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1166 of ldap servers used for a specific lookup. Patch provided by Heiko
1169 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1170 New variable $lookup_dnssec_authenticated for observability.
1172 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1173 Patch submitted by Lars Timman.
1175 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1177 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1178 Requires trusted mode and valid format message id, aborts otherwise.
1179 Patch contributed by Heiko Schlichting.
1181 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1182 certextract with support for various fields. Bug 1358.
1184 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1185 is requested by default, modifiable by smtp transport option
1188 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1189 operate on certificate variables to give certificate fingerprints
1190 Also new ${sha256:cert_variable}.
1192 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1194 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1195 Christian Aistleitner.
1197 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1199 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1200 file. Patch from Wolfgang Breyha.
1202 JH/25 Expand the coverage of the delivery $host and $host_address to
1203 client authenticators run in verify callout. Bug 1476.
1205 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1206 align with daemon_smtp_ports. Bug 72.
1208 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1209 support and error reporting did not work properly.
1211 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1212 and is readable. Patch from Andrew Colin Kissa.
1214 TL/14 Enhance documentation of ${run expansion and how it parses the
1215 commandline after expansion, particularly in the case when an
1216 unquoted variable expansion results in an empty value.
1218 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1220 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1221 by renaming away from it. Addresses GH issue 15, reported by
1224 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1226 TL/15 SECURITY: prevent double expansion in math comparison functions
1227 (can expand unsanitized data). Not remotely exploitable.
1234 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1236 PP/02 Make -n do something, by making it not do something.
1237 When combined with -bP, the name of an option is not output.
1239 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1242 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1243 $sender_host_name and config options to manage this, and basic check
1246 PP/05 DSCP support for outbound connections and control modifier for inbound.
1248 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1249 (Only plugin which currently uses this is kerberos4, which nobody should
1250 be using, but we should make it available and other future plugins might
1251 conceivably use it, even though it would break NAT; stuff *should* be
1252 using channel bindings instead).
1254 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1255 name; added for Sendmail compatibility; requires admin caller.
1256 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1257 just ignore it); requires trusted caller.
1258 Also parse but ignore: -Ac -Am -X<logfile>
1261 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1263 TL/02 Add +smtp_confirmation as a default logging option.
1265 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1266 Patch by Magnus Holmgren from 2007-02-20.
1268 TL/04 Bugzilla 1281 - Spec typo.
1269 Bugzilla 1283 - Spec typo.
1270 Bugzilla 1290 - Spec grammar fixes.
1272 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1274 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1276 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1277 mailing list by by Dmitry Isaikin.
1279 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1281 JH/02 Support "G" suffix to numbers in ${if comparisons.
1283 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1285 NM/01 Bugzilla 1197 - Spec typo
1286 Bugzilla 1196 - Spec examples corrections
1288 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1290 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1291 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1294 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1295 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1297 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1298 "acl {{name}{arg}...}", and optional args on acl condition
1301 JH/05 Permit multiple router/transport headers_add/remove lines.
1303 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1305 JH/07 Avoid using a waiting database for a single-message-only transport.
1306 Performance patch from Paul Fisher. Bugzilla 1262.
1308 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1311 JH/09 Add $headers_added variable, with content from use of ACL modifier
1312 add_header (but not yet added to the message). Bugzilla 199.
1314 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1315 Pulled from Bugzilla 817 by Wolfgang Breyha.
1317 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1319 (nb: this is the same fix as in Exim 4.80.1)
1321 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1324 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1327 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1329 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1330 advertise SMTP AUTH mechanism to us, instead of a generic
1331 protocol violation error. Also, make Exim more robust to bad
1332 data from the Dovecot auth socket.
1334 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1336 When a queue runner is handling a message, Exim first routes the
1337 recipient addresses, during which it prunes them based on the retry
1338 hints database. After that it attempts to deliver the message to
1339 any remaining recipients. It then updates the hints database using
1342 So if a recipient address works intermittently, it can get repeatedly
1343 deferred at routing time. The retry hints record remains fresh so the
1344 address never reaches the final cutoff time.
1346 This is a fairly common occurrence when a user is bumping up against
1347 their storage quota. Exim had some logic in its local delivery code
1348 to deal with this. However it did not apply to per-recipient defers
1349 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1351 This change adds a proper retry rule check during routing so that the
1352 final cutoff time is checked against the message's age. We only do
1353 this check if there is an address retry record and there is not a
1354 domain retry record; this implies that previous attempts to handle
1355 the address had the retry_use_local_parts option turned on. We use
1356 this as an approximation for the destination being like a local
1357 delivery, as in LMTP.
1359 I suspect this new check makes the old local delivery cutoff check
1360 redundant, but I have not verified this so I left the code in place.
1362 TF/02 Correct gecos expansion when From: is a prefix of the username.
1364 Test 0254 submits a message to Exim with the header
1368 When I ran the test suite under the user fanf2, Exim expanded
1369 the header to contain my full name, whereas it should have added
1370 a Resent-Sender: header. It erroneously treats any prefix of the
1371 username as equal to the username.
1373 This change corrects that bug.
1375 GF/01 DCC debug and logging tidyup
1376 Error conditions log to paniclog rather than rejectlog.
1377 Debug lines prefixed by "DCC: " to remove any ambiguity.
1379 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1381 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1382 Bug spotted by Jeremy Harris; was flawed since initial commit.
1383 Would have resulted in OCSP responses post-SNI triggering an Exim
1384 NULL dereference and crash.
1386 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1388 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1389 Bug detection, analysis and fix by Samuel Thibault.
1390 Bugzilla 1331, Debian bug #698092.
1392 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1394 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1395 Server implementation by Todd Lyons, client by JH.
1396 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1397 config variable "prdr_enable" controls whether the server
1398 advertises the facility. If the client requests PRDR a new
1399 acl_data_smtp_prdr ACL is called once for each recipient, after
1400 the body content is received and before the acl_smtp_data ACL.
1401 The client is controlled by both of: a hosts_try_prdr option
1402 on the smtp transport, and the server advertisement.
1403 Default client logging of deliveries and rejections involving
1404 PRDR are flagged with the string "PRDR".
1406 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1407 fclose(). Diagnosis by Todd Lyons.
1409 PP/17 Update configure.default to handle IPv6 localhost better.
1410 Patch by Alain Williams (plus minor tweaks).
1413 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1414 This is now consistent with GnuTLS, and is now documented: the
1415 previous undocumented portable approach to treating the option as
1416 unset was to force an expansion failure. That still works, and
1417 an empty string is now equivalent.
1419 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1420 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1421 not performing validation itself.
1423 PP/20 Added force_command boolean option to pipe transport.
1424 Patch from Nick Koston, of cPanel Inc.
1426 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1429 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1431 PP/21 Fix eximon continuous updating with timestamped log-files.
1432 Broken in a format-string cleanup in 4.80, missed when I repaired the
1433 other false fix of the same issue.
1434 Report and fix from Heiko Schlichting.
1437 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1438 Report from Prashanth Katuri.
1440 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1441 It's SecureTransport, so affects any MacOS clients which use the
1442 system-integrated TLS libraries, including email clients.
1444 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1445 using a MIME ACL for non-SMTP local injection.
1446 Report and assistance in diagnosis by Warren Baker.
1448 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1450 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1452 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1453 last id that failed so it may be referenced in subsequent ACL's.
1455 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1458 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1459 ldap library initialization, allowing self-signed CA's to be
1460 used. Also properly sets require_cert option later in code by
1461 using NULL (global ldap config) instead of ldap handle (per
1462 session). Bug diagnosis and testing by alxgomz.
1464 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1465 the src/util/ subdirectory.
1467 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1468 renamed to Transport Post Delivery Action by Jeremy Harris, as
1471 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1472 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1473 redis_servers = needs to be configured which will be used by the redis
1474 lookup. Patch from Warren Baker, of The Packet Hub.
1476 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1478 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1479 hostname or reverse DNS when processing a host list. Used suggestions
1480 from multiple comments on this bug.
1482 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1484 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1485 Missed a few lines, added it to make the runtest require no keyboard
1488 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1489 contains upper case chars. Make router use caseful_local_part.
1491 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1492 support when GnuTLS has been built with p11-kit.
1498 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1500 This, or similar/improved, will also be change PP/11 of 4.82.
1506 PP/01 Handle short writes when writing local log-files.
1507 In practice, only affects FreeBSD (8 onwards).
1508 Bugzilla 1053, with thanks to Dmitry Isaikin.
1510 NM/01 Bugzilla 949 - Documentation tweak
1512 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1515 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1517 PP/02 Implemented gsasl authenticator.
1519 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1521 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1522 `pkg-config foo` for cflags/libs.
1524 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1525 with rest of GSASL and with heimdal_gssapi.
1527 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1528 `pkg-config foo` for cflags/libs for the TLS implementation.
1530 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1531 properties get this fed in as external SSF. A number of robustness
1532 and debugging improvements to the cyrus_sasl authenticator.
1534 PP/08 cyrus_sasl server now expands the server_realm option.
1536 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1537 Patch by Jeremy Harris.
1539 PP/10 Added dbmjz lookup type.
1541 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1543 PP/12 MAIL args handles TAB as well as SP, for better interop with
1544 non-compliant senders.
1545 Analysis and variant patch by Todd Lyons.
1547 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1548 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1549 Patch from Dirk Mueller <dmueller@suse.com>
1551 PP/13 tls_peerdn now print-escaped for spool files.
1552 Observed some $tls_peerdn in wild which contained \n, which resulted
1553 in spool file corruption.
1555 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1556 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1557 or write after TLS renegotiation, which otherwise led to messages
1560 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1561 as a tracking header (ie: a signed header comes before the signature).
1562 Patch from Wolfgang Breyha.
1564 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1565 comma-sep list; embedded commas doubled.
1567 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1569 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1571 Report and patch from Dmitry Banschikov.
1573 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1574 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1575 failures. We appear to now support TLS1.1+ with Exim.
1577 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1578 lets Exim select keys and certificates based upon TLS SNI from client.
1579 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1580 before an outbound SMTP session. New log_selector, +tls_sni.
1582 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1583 NULL dereference. Report and patch from Alun Jones.
1585 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1586 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1587 Not seeing resolver debug output on NetBSD, but suspect this is a
1588 resolver implementation change.
1590 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1591 Left warnings. Added "eximon gdb" invocation mode.
1593 PP/21 Defaulting "accept_8bitmime" to true, not false.
1595 PP/22 Added -bw for inetd wait mode support.
1597 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1598 locate the relevant includes and libraries. Made this the default.
1600 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1601 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1603 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1604 This may cause build issues on older platforms.
1606 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1607 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1608 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1609 Added SNI support via GnuTLS too.
1610 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1612 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1614 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1615 Applied second patch from Janne, implementing suggestion to default
1616 multiple-strings-in-record handling to match SPF spec.
1618 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1620 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1621 read-only, out of scope).
1622 Patch from Wolfgang Breyha, report from Stuart Northfield.
1624 PP/29 Fix three issues highlighted by clang analyser static analysis.
1625 Only crash-plausible issue would require the Cambridge-specific
1626 iplookup router and a misconfiguration.
1627 Report from Marcin Mirosław.
1629 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1631 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1632 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1633 As part of this, removing so much warning spew let me fix some minor
1634 real issues in debug logging.
1636 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1637 assignment on my part. Fixed.
1639 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1640 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1641 Janne Snabb (who went above and beyond: thank you).
1643 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1644 string otherwise requires a connection and a bunch more work and it's
1645 relatively easy to get wrong. Should also expose TLS library linkage
1648 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1649 64-bit ${eval} (JH/03).
1651 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1652 GNU libc to support some of the 64-bit stuff, should not lead to
1653 conflicts. Defined before os.h is pulled in, so if a given platform
1654 needs to override this, it can.
1656 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1657 protection layer was required, which is not implemented.
1658 Bugzilla 1254, patch from Wolfgang Breyha.
1660 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1661 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1662 tls_dhparam take prime identifiers. Also unbreak combination of
1663 OpenSSL+DH_params+TLSSNI.
1665 PP/39 Disable SSLv2 by default in OpenSSL support.
1671 PP/01 Solaris build fix for Oracle's LDAP libraries.
1672 Bugzilla 1109, patch from Stephen Usher.
1674 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1676 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1679 TF/02 Fix a couple more cases where we did not log the error message
1680 when unlink() failed. See also change 4.74-TF/03.
1682 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1683 lock up or crash if it happened to be inside a call to libc when it
1684 got a SIGUSR1 from exiwhat.
1686 The SIGUSR1 handler appends the current process status to the process
1687 log which is later printed by exiwhat. It used to use the general
1688 purpose logging code to do this, but several functions it calls are
1689 not safe for signals.
1691 The new output code in the SIGUSR1 handler is specific to the process
1692 log, and simple enough that it's easy to inspect for signal safety.
1693 Removing some special cases also simplifies the general logging code.
1694 Removing the spurious timestamps from the process log simplifies
1697 TF/04 Improved ratelimit ACL condition.
1699 The /noupdate option has been deprecated in favour of /readonly which
1700 has clearer semantics. The /leaky, /strict, and /readonly update modes
1701 are mutually exclusive. The update mode is no longer included in the
1702 database key; it just determines when the database is updated. (This
1703 means that when you upgrade Exim will forget old rate measurements.)
1705 Exim now checks that the per_* options are used with an update mode that
1706 makes sense for the current ACL. For example, when Exim is processing a
1707 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1708 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1709 must specify per_mail/readonly. If you omit the update mode it defaults to
1710 /leaky where that makes sense (as before) or /readonly where required.
1712 The /noupdate option is now undocumented but still supported for
1713 backwards compatibility. It is equivalent to /readonly except that in
1714 ACLs where /readonly is required you may specify /leaky/noupdate or
1715 /strict/noupdate which are treated the same as /readonly.
1717 A useful new feature is the /count= option. This is a generalization
1718 of the per_byte option, so that you can measure the throughput of other
1719 aggregate values. For example, the per_byte option is now equivalent
1720 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1722 The per_rcpt option has been generalized using the /count= mechanism
1723 (though it's more complicated than the per_byte equivalence). When it is
1724 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1725 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1726 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1727 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1728 Note that using per_rcpt with a non-readonly update mode in more than
1729 one ACL will cause the recipients to be double-counted. (The per_mail
1730 and per_byte options don't have this problem.)
1732 The handling of very low rates has changed slightly. If the computed rate
1733 is less than the event's count (usually one) then this event is the first
1734 after a long gap. In this case the rate is set to the same as this event's
1735 count, so that the first message of a spam run is counted properly.
1737 The major new feature is a mechanism for counting the rate of unique
1738 events. The new per_addr option counts the number of different
1739 recipients that someone has sent messages to in the last time period. It
1740 behaves like per_rcpt if all the recipient addresses are different, but
1741 duplicate recipient addresses do not increase the measured rate. Like
1742 the /count= option this is a general mechanism, so the per_addr option
1743 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1744 example, measure the rate that a client uses different sender addresses
1745 with the options per_mail/unique=$sender_address. There are further
1746 details in the main documentation.
1748 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1750 TF/06 Removed a few PCRE remnants.
1752 TF/07 Automatically extract Exim's version number from tags in the git
1753 repository when doing development or release builds.
1755 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1756 Bugzilla 879. Patch from Paul Fisher.
1758 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1759 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1762 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1764 PP/05 Variable $av_failed, true if the AV scanner deferred.
1765 Bugzilla 1078. Patch from John Horne.
1767 PP/06 Stop make process more reliably on build failure.
1768 Bugzilla 1087. Patch from Heiko Schlittermann.
1770 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1771 Bugzilla 1089. Patch from Heiko Schlittermann.
1773 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1774 Bugzilla 1131. Patch from Holger Weiß.
1776 PP/09 Handle IPv6 addresses with SPF.
1777 Bugzilla 860. Patch from Wolfgang Breyha.
1779 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1781 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1784 PP/11 match_* no longer expand right-hand-side by default.
1785 New compile-time build option, EXPAND_LISTMATCH_RHS.
1786 New expansion conditions, "inlist", "inlisti".
1788 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1790 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1792 PP/14 fix log_write() format string regression from TF/03.
1793 Bugzilla 1152. Patch from Dmitry Isaikin.
1799 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1801 PP/02 Harmonised TLS library version reporting; only show if debugging.
1802 Layout now matches that introduced for other libraries in 4.74 PP/03.
1804 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1806 PP/04 New "dns_use_edns0" global option.
1808 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1811 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1812 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1814 TK/01 Updated PolarSSL code to 0.14.2.
1815 Bugzilla 1097. Patch from Andreas Metzler.
1817 PP/07 Catch divide-by-zero in ${eval:...}.
1818 Fixes bugzilla 1102.
1820 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1823 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1824 format-string attack -- SECURITY: remote arbitrary code execution.
1826 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1827 time unintentionally subject to list matching rules, letting the header
1828 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1829 arbitrary string expansion). This allowed for information disclosure.
1831 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1832 INT_MIN/-1 -- value coerced to INT_MAX.
1838 NM/01 Workaround for PCRE version dependency in version reporting
1841 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1842 This fixes portability to compilers other than gcc, notably
1843 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1845 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1846 makefiles for portability to HP-UX and POSIX correctness.
1848 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1849 Also via indented variable definition in the Makefile.
1850 (Debugging by Oliver Heesakkers).
1852 PP/02 Restore caching of spamd results with expanded spamd_address.
1853 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1855 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1856 Improves build reliability. Fix from: Frank Elsner
1858 NM/02 Fix wide character breakage in the rfc2047 coding
1859 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1861 NM/03 Allow underscore in dnslist lookups
1862 Fixes bug 1026. Patch from Graeme Fowler
1864 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1865 Code patches from Adam Ciarcinski of NetBSD.
1867 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1870 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1871 is logged, to avoid truncation. Patch from John Horne.
1873 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1874 Patch from Jakob Hirsch.
1876 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1877 SQL string expansion failure details.
1878 Patch from Andrey Oktyabrski.
1880 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1881 Patch from Simon Arlott.
1883 PP/09 New lookups functionality failed to compile on old gcc which rejects
1884 extern declarations in function scope.
1885 Patch from Oliver Fleischmann
1887 PP/10 Use sig_atomic_t for flags set from signal handlers.
1888 Check getgroups() return and improve debugging.
1889 Fixed developed for diagnosis in bug 927 (which turned out to be
1892 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1893 Patch from Mark Zealey.
1895 PP/12 Bugzilla 1056: Improved spamd server selection.
1896 Patch from Mark Zealey.
1898 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1899 Based on patch from Heiko Schlittermann.
1901 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1902 Patch from Uwe Doering, sign-off by Michael Haardt.
1904 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1905 variable declaration deep within a block. Bug and patch from
1908 PP/15 lookups-Makefile IRIX compatibility coercion.
1910 PP/16 Make DISABLE_DKIM build knob functional.
1912 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1913 Patch by Simon Arlott
1915 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1916 variable argument macros. Our copy now differs from upstream.
1922 TF/01 Failure to get a lock on a hints database can have serious
1923 consequences so log it to the panic log.
1925 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1926 controlled using the smtp_confirmation log selector.
1928 TF/03 Include the error message when we fail to unlink a spool file.
1930 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1931 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1932 for maintaining out-of-tree patches for some time.
1934 PP/01 Bugzilla 139: Documentation and portability issues.
1935 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1936 Handle per-OS dynamic-module compilation flags.
1938 PP/02 Let /dev/null have normal permissions.
1939 The 4.73 fixes were a little too stringent and complained about the
1940 permissions on /dev/null. Exempt it from some checks.
1941 Reported by Andreas M. Kirchwitz.
1943 PP/03 Report version information for many libraries, including
1944 Exim version information for dynamically loaded libraries. Created
1945 version.h, now support a version extension string for distributors
1946 who patch heavily. Dynamic module ABI change.
1948 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1949 privilege escalation vulnerability whereby the Exim run-time user
1950 can cause root to append content of the attacker's choosing to
1953 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1956 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1957 If dropping privileges for untrusted macros, we disabled normal logging
1958 on the basis that it would fail; for the Exim run-time user, this is not
1959 the case, and it resulted in successful deliveries going unlogged.
1960 Fixed. Reported by Andreas Metzler.
1966 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1967 only prepend for the Resent-* case. Fixes regression introduced in
1968 Exim 4.70 by NM/22 for Bugzilla 607.
1970 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1971 increasing numbers of administrators be bitten by this.
1973 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1975 PP/03 Bugzilla 994: added openssl_options main configuration option.
1977 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1979 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1981 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1983 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1984 without a peer certificate, leading to a segfault because of an
1985 assumption that peers always have certificates. Be a little more
1986 paranoid. Problem reported by Martin Tscholak.
1988 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1989 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1990 NB: ClamAV planning to remove STREAM in "middle of 2010".
1991 CL also introduces -bmalware, various -d+acl logging additions and
1992 more caution in buffer sizes.
1994 PP/09 Implemented reverse_ip expansion operator.
1996 PP/10 Bugzilla 937: provide a "debug" ACL control.
1998 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
2000 PP/12 Bugzilla 973: Implement --version.
2002 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
2004 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
2006 PP/15 Bugzilla 816: support multiple condition rules on Routers.
2008 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
2009 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
2010 ignore trailing whitespace.
2012 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2014 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2017 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2018 Notification from Dr Andrew Aitchison.
2020 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2021 ExtendedDetectionInfo response format.
2022 Notification from John Horne.
2024 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2027 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2028 XSL and documented dependency on system catalogs, with examples of how
2031 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2034 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2035 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2036 configuration file which is writeable by the Exim user or group.
2038 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2039 of configuration files to cover files specified with the -C option if
2040 they are going to be used with root privileges, not just the default
2043 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2044 option (effectively making it always true).
2046 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2047 files to be used while preserving root privileges.
2049 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2050 that rogue child processes cannot use them.
2052 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2053 run-time user, instead of root.
2055 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2056 Exim run-time user without dropping privileges.
2058 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2059 result string, instead of calling string_vformat() twice with the same
2062 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2063 for other users. Others should always drop root privileges if they use
2064 -C on the command line, even for a whitelisted configure file.
2066 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2068 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2074 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2075 $data_path, and $header_path variables; fixed documentation bugs and
2078 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2079 exipick to access non-standard spools, including the "frozen" queue
2082 NM/01 Bugzilla 965: Support mysql stored procedures.
2083 Patch from Alain Williams
2085 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2087 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2088 Patch from Andreas Metzler
2090 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2091 Patch from Kirill Miazine
2093 NM/05 Bugzilla 671: Added umask to procmail example.
2095 JJ/03 installed exipick 20100323.0, fixing doc bug
2097 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2098 directory. Notification and patch from Dan Rosenberg.
2100 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2102 TK/02 Improve log output when DKIM signing operation fails.
2104 MH/01 Treat the transport option dkim_domain as a colon separated
2105 list, not as a single string, and sign the message with each element,
2106 omitting multiple occurences of the same signer.
2108 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2109 Bugzilla 985, 986. Patch by Simon Arlott
2111 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2112 Patch by Simon Arlott
2114 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2115 MBX locking. Notification from Dan Rosenberg.
2121 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2123 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2125 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2127 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2129 NM/04 Bugzilla 915: Flag broken perl installation during build.
2135 TK/01 Added patch by Johannes Berg that expands the main option
2136 "spamd_address" if it starts with a dollar sign.
2138 TK/02 Write list of recipients to X-Envelope-Sender header when building
2139 the mbox-format spool file for content scanning (suggested by Jakob
2142 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2143 (http://www.dcc-servers.net/) support via dccifd. Activated by
2144 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2146 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2147 by Mark Daniel Reidel <mr@df.eu>.
2149 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2150 When building exim an external PCRE library is now needed -
2151 PCRE is a system library on the majority of modern systems.
2152 See entry on PCRE_LIBS in EDITME file.
2154 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2155 conversation. Added nologin parameter to request.
2156 Patch contributed by Kirill Miazine.
2158 TF/01 Do not log submission mode rewrites if they do not change the address.
2160 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2162 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2163 log files in place. Contributed by Roberto Lima.
2165 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2167 TF/03 Bugzilla 615: When checking the local_parts router precondition
2168 after a local_part_suffix or local_part_prefix option, Exim now
2169 does not use the address's named list lookup cache, since this
2170 contains cached lookups for the whole local part.
2172 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2173 Robert Millan. Documentation is in experimental-spec.txt.
2175 TF/04 Bugzilla 668: Fix parallel build (make -j).
2177 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2179 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2180 Patch provided by Jan Srzednicki.
2182 TF/05 Leading white space used to be stripped from $spam_report which
2183 wrecked the formatting. Now it is preserved.
2185 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2186 that they are available at delivery time.
2188 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2190 TF/08 TLS error reporting now respects the incoming_interface and
2191 incoming_port log selectors.
2193 TF/09 Produce a more useful error message if an SMTP transport's hosts
2194 setting expands to an empty string.
2196 NM/06 Bugzilla 744: EXPN did not work under TLS.
2197 Patch provided by Phil Pennock.
2199 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2200 Patch provided by Richard Godbee.
2202 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2203 acl_smtp_notquit, added index entry.
2205 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2206 Patch provided by Eugene Bujak.
2208 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2209 accept(). Patch provided by Maxim Dounin.
2211 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2212 Patch provided by Phil Pennock.
2214 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2216 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2217 Patch provided by Brad "anomie" Jorsch.
2219 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2220 Patch provided by Dean Brooks.
2222 TK/05 Add native DKIM support (does not depend on external libraries).
2224 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2225 Patch provided by Graeme Fowler.
2227 NM/16 Bugzilla 851: Documentation example syntax fix.
2229 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2231 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2234 NM/19 Bugzilla 745: TLS version reporting.
2235 Patch provided by Phil Pennock.
2237 NM/20 Bugzilla 167: bool: condition support.
2238 Patch provided by Phil Pennock.
2240 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2241 clients. Patch provided by Phil Pennock.
2243 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2244 Patch provided by Brad "anomie" Jorsch.
2246 NM/23 Bugzilla 687: Fix misparses in eximstats.
2247 Patch provided by Heiko Schlittermann.
2249 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2250 Patch provided by Heiko Schlittermann.
2252 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2253 plus update to original patch.
2255 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2257 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2258 Patch provided by David Brownlee.
2260 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2262 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2264 NM/30 Bugzilla 888: TLS documentation bugfixes.
2266 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2268 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2269 Unlike the original bugzilla I have changed all shell scripts in src tree.
2271 NM/33 Bugzilla 898: Transport filter timeout fix.
2272 Patch by Todd Rinaldo.
2274 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2275 Patch by Serge Demonchaux.
2277 NM/35 Bugzilla 39: Base64 decode bug fixes.
2278 Patch by Jakob Hirsch.
2280 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2282 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2284 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2286 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2292 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2293 ALT-N's libdkim that I have put here:
2294 http://duncanthrax.net/exim-experimental/
2296 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2297 were called 'true' and it seems that C99 defines that as a reserved
2298 keyword to be used with 'bool' variable types. That means you could
2299 not include C99-style headers which use bools without triggering
2300 build errors in sieve.c.
2302 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2303 as mailq or other aliases. Changed the --help handling significantly
2304 to do whats expected. exim_usage() emits usage/help information.
2306 SC/01 Added the -bylocaldomain option to eximstats.
2308 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2310 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2312 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2318 PH/01 Another patch from the Sieve maintainer.
2320 PH/02 When an IPv6 address is converted to a string for single-key lookup
2321 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2322 dots are used instead of colons so that keys in lsearch files need not
2323 contain colons. This was done some time before quoting was made available
2324 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2325 (notated using the quote facility) so as to distinguish them from IPv4
2326 keys. This meant that lookups for IP addresses in host lists did not work
2327 for iplsearch lookups.
2329 This has been fixed by arranging for IPv6 addresses to be expressed with
2330 colons if the lookup type is iplsearch. This is not incompatible, because
2331 previously such lookups could never work.
2333 The situation is now rather anomalous, since one *can* have colons in
2334 ordinary lsearch keys. However, making the change in all cases is
2335 incompatible and would probably break a number of configurations.
2337 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2340 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2341 conversion specification without a maximum field width, thereby enabling
2342 a rogue spamd server to cause a buffer overflow. While nobody in their
2343 right mind would setup Exim to query an untrusted spamd server, an
2344 attacker that gains access to a server running spamd could potentially
2345 exploit this vulnerability to run arbitrary code as the Exim user.
2347 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2348 $primary_hostname instead of what libspf2 thinks the hosts name is.
2350 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2351 a directory entry by the name of the lookup key. Previously, if a
2352 symlink pointed to a non-existing file or a file in a directory that
2353 Exim lacked permissions to read, a lookup for a key matching that
2354 symlink would fail. Now it is enough that a matching directory entry
2355 exists, symlink or not. (Bugzilla 503.)
2357 PH/03 The body_linecount and body_zerocount variables are now exported in the
2360 PH/04 Added the $dnslist_matched variable.
2362 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2363 This means they are set thereafter only if the connection becomes
2366 PH/06 Added the client_condition to authenticators so that some can be skipped
2367 by clients under certain conditions.
2369 PH/07 The error message for a badly-placed control=no_multiline_responses left
2370 "_responses" off the end of the name.
2372 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2374 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2375 (without spaces) instead of just copying the configuration text.
2377 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2379 PH/11 Added $max_received_linelength.
2381 PH/12 Added +ignore_defer and +include_defer to host lists.
2383 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2384 way in which PCRE > 7.0 is built.
2386 PH/14 Implemented queue_only_load_latch.
2388 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2389 MAIL command. The effect was to mangle the value on 64-bit systems.
2391 PH/16 Another patch from the Sieve maintainer.
2393 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2395 PH/18 If a system quota error occurred while trying to create the file for
2396 a maildir delivery, the message "Mailbox is full" was not appended to the
2397 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2398 only to a quota excession during the actual writing of the file.
2400 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2401 characters?) which causes problems in log lines. The DN values are now
2402 passed through string_printing() before being added to log lines.
2404 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2405 and InterBase are left for another time.)
2407 PH/21 Added message_body_newlines option.
2409 PH/22 Guard against possible overflow in moan_check_errorcopy().
2411 PH/23 POSIX allows open() to be a macro; guard against that.
2413 PH/24 If the recipient of an error message contained an @ in the local part
2414 (suitably quoted, of course), incorrect values were put in $domain and
2415 $local_part during the evaluation of errors_copy.
2421 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2422 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2425 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2426 issue a MAIL command.
2428 PH/02 In an ACL statement such as
2430 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2432 if a client was not listed at all, or was listed with a value other than
2433 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2434 the condition was not true (as it should be), so access was not denied.
2435 The bug was that the ! inversion was incorrectly passed on to the second
2436 item. This has been fixed.
2438 PH/03 Added additional dnslists conditions == and =& which are different from
2439 = and & when the dns lookup returns more than one IP address.
2441 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2442 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2444 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2445 FSYNC, which compiles an option called disable_fsync that allows for
2446 bypassing fsync(). The documentation is heavily laced with warnings.
2448 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2450 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2451 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2452 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2453 including adding "make clean"; (3) Added -fPIC when compiling the test
2454 dynamically loaded module, to get rid of a warning.
2456 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2457 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2458 = 0s. The bug is otherwise harmless.
2460 PH/07 There was a bug in the dovecot authenticator such that the value of
2461 $auth1 could be overwritten, and so not correctly preserved, after a
2462 successful authentication. This usually meant that the value preserved by
2463 the server_setid option was incorrect.
2465 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2467 PH/09 Installed PCRE release 7.0.
2469 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2470 run for batched SMTP input. It is now run at the start of every message
2471 in the batch. While fixing this I discovered that the process information
2472 (output by running exiwhat) was not always getting set for -bs and -bS
2473 input. This is fixed, and it now also says "batched" for BSMTP.
2475 PH/11 Added control=no_pipelining.
2477 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2478 patch, slightly modified), and move the expansion of helo_data till after
2479 the connection is made in the smtp transport (so it can use these
2482 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2484 PH/14 Added log_selector = +pid.
2486 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2488 PH/16 Add ${if forany and ${if forall.
2490 PH/17 Added dsn_from option to vary the From: line in DSNs.
2492 PH/18 Flush SMTP output before performing a callout, unless control =
2493 no_callout_flush is set.
2495 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2496 was true (the default) a successful delivery failed to delete the retry
2497 item, thus causing premature timeout of the address. The bug is now
2500 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2502 PH/21 Long custom messages for fakedefer and fakereject are now split up
2503 into multiline responses in the same way that messages for "deny" and
2504 other ACL rejections are.
2506 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2507 with slight modification.
2509 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2510 draft, changing the syntax and factoring some duplicate code".
2512 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2513 for deliveries of the second and subsequent messages over the same SMTP
2516 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2517 ${reduce, with only minor "tidies".
2519 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2521 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2522 expansion side effects.
2524 PH/27 When a message times out after an over-quota error from an Exim-imposed
2525 quota, the bounce message says "mailbox is full". This message was not
2526 being given when it was a system quota that was exceeded. It now should
2529 MH/03 Made $recipients available in local_scan(). local_scan() already has
2530 better access to the recipient list through recipients_list[], but
2531 $recipients can be useful in postmaster-provided expansion strings.
2533 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2534 in the case of a MAIL command with additional options following the
2535 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2536 were accidentally chopped off.
2538 PH/29 SMTP synchronization checks are implemented when a command is read -
2539 there is a check that no more input is waiting when there shouldn't be
2540 any. However, for some commands, a delay in an ACL can mean that it is
2541 some time before the response is written. In this time, more input might
2542 arrive, invalidly. So now there are extra checks after an ACL has run for
2543 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2544 pipelining has not been advertised.
2546 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2548 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2549 returned in the SMTP response when smtp_return_error_details was false.
2550 This has been fixed.
2552 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2553 instead of the C I/O that was originally supplied, because problems were
2554 reported on Solaris.
2556 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2557 Exim which did not show up earlier: it was assuming that a call to
2558 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2559 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2560 was a macro that became an assignment, so it seemed to work. This has
2561 changed to a proper function call with a void return, hence the compile
2562 error. Exim's code has been fixed.
2564 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2567 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2569 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2571 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2572 because it turns out to be too big or there is a timeout) there is no
2573 "Completed" line in the log. When some messages of this type were
2574 selected by exigrep, they were listed as "not completed". Others were
2575 picked up by some special patterns. I have improved the selection
2576 criteria to be more general.
2578 PH/38 The host_find_failed option in the manualroute router can now be set
2579 to "ignore", to completely ignore a host whose IP address cannot be
2580 found. If all hosts are ignored, the behaviour is controlled by the new
2581 host_all_ignored option.
2583 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2584 homing or because of multiple MX records with /mx) generated more than
2585 one IP address, and the following item turned out to be the local host,
2586 all the secondary addresses of the first item were incorrectly removed
2587 from the list, along with the local host and any following hosts (which
2588 is what is supposed to happen).
2590 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2591 whoever called Exim into the -H file. In the case of the daemon it was
2592 behaving confusingly. When first started, it used values for whoever
2593 started the daemon, but after a SIGHUP it used the Exim user (because it
2594 calls itself on a restart). I have changed the code so that it now always
2597 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2598 message are rejected with the same error (e.g. no authentication or bad
2599 sender address), and a DATA command is nevertheless sent (as can happen
2600 with PIPELINING or a stupid MUA), the error message that was given to the
2601 RCPT commands is included in the rejection of the DATA command. This is
2602 intended to be helpful for MUAs that show only the final error to their
2605 PH/42 Another patch from the Sieve maintainer.
2607 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2608 Eximstats - Fixed some broken HTML links and added missing column headers
2610 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2611 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2613 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2615 PH/43 Yet another patch from the Sieve maintainer.
2617 PH/44 I found a way to check for a TCP/IP connection going away before sending
2618 the response to the final '.' that terminates a message, but only in the
2619 case where the client has not sent further data following the '.'
2620 (unfortunately, this is allowed). However, in many cases there won't be
2621 any further data because there won't be any more messages to send. A call
2622 to select() can be used: if it shows that the input is "ready", there is
2623 either input waiting, or the socket has been closed. An attempt to read
2624 the next input character can distinguish the two cases. Previously, Exim
2625 would have sent an OK response which the client would never have see.
2626 This could lead to message repetition. This fix should cure that, at
2627 least in a lot of common cases.
2629 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2630 advertised in response to EHLO.
2636 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2637 fixed by 4.65/MH/01 (is this a record?) are fixed:
2639 (i) An empty string was always treated as zero by the numeric comparison
2640 operators. This behaviour has been restored.
2642 (ii) It is documented that the numeric comparison operators always treat
2643 their arguments as decimal numbers. This was broken in that numbers
2644 starting with 0 were being interpreted as octal.
2646 While fixing these problems I realized that there was another issue that
2647 hadn't been noticed. Values of message_size_limit (both the global option
2648 and the transport option) were treated as octal if they started with 0.
2649 The documentation was vague. These values are now always treated as
2650 decimal, and I will make that clear in the documentation.
2656 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2657 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2660 MH/01 Don't check that the operands of numeric comparison operators are
2661 integers when their expansion is in "skipping" mode (fixes bug
2662 introduced by 4.64-PH/07).
2664 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2665 child addresses, Exim now panics and dies. Previously, because the count
2666 is held in a short int, deliveries were likely to be lost. As such a
2667 large number of recipients for a single message is ridiculous
2668 (performance will be very, very poor), I have chosen to impose a limit
2669 rather than extend the field.
2675 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2676 leftover -K file (the existence of which was triggered by #402).
2677 While we were at it, introduced process PID as part of the -K
2678 filename. This should rule out race conditions when creating
2681 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2682 processing considerably. Previous code took too long for large mails,
2683 triggering a timeout which in turn triggers #401.
2685 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2686 in the DK code in transports.c. sendfile() is not really portable,
2687 hence the _LINUX specificness.
2689 TF/01 In the add_headers option to the mail command in an Exim filter,
2690 there was a bug that Exim would claim a syntax error in any
2691 header after the first one which had an odd number of characters
2694 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2695 callout verification, Exim cached a "reject" for the entire domain. This
2696 is correct for most verifications, but it is not correct for a recipient
2697 verification with use_sender or use_postmaster set, because in that case
2698 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2699 case of MAIL FROM:<> rejection from other early rejections (e.g.
2700 rejection of HELO). When verifying a recipient using a non-null MAIL
2701 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2702 Whatever the result of the callout, the value of the domain cache is
2703 left unchanged (for any other kind of callout, getting as far as trying
2704 RCPT means that the domain itself is ok).
2706 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2709 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2710 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2711 session, the daemon ignores ECONNECT errors and logs others; it now
2712 ignores EPIPE as well.
2714 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2715 (quoted-printable decoding).
2717 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2718 later the small subsequent patch to fix an introduced bug.
2720 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2722 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2724 PH/08 An error is now given if message_size_limit is specified negative.
2726 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2727 to be given (somewhat) arbitrary names.
2729 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2732 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2733 miscellaneous code fixes
2735 PH/10 Added the log_reject_target ACL modifier to specify where to log
2738 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2739 hostname. This is wrong, because it relates to the incoming message (and
2740 probably the interface on which it is arriving) and not to the outgoing
2741 callout (which could be using a different interface). This has been
2742 changed to use the value of the helo_data option from the smtp transport
2743 instead - this is what is used when a message is actually being sent. If
2744 there is no remote transport (possible with a router that sets up host
2745 addresses), $smtp_active_hostname is used.
2747 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2748 tweaks were necessary in order to get it to work (see also 21 below):
2749 (a) The code assumed that strncpy() returns a negative number on buffer
2750 overflow, which isn't the case. Replaced with Exim's string_format()
2752 (b) There were several signed/unsigned issues. I just did the minimum
2753 hacking in of casts. There is scope for a larger refactoring.
2754 (c) The code used strcasecmp() which is not a standard C function.
2755 Replaced with Exim's strcmpic() function.
2756 (d) The code set only $1; it now sets $auth1 as well.
2757 (e) A simple test gave the error "authentication client didn't specify
2758 service in request". It would seem that Dovecot has changed its
2759 interface. Fortunately there's a specification; I followed it and
2760 changed what the client sends and it appears to be working now.
2762 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2765 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2766 address A is aliased to B and C, where B exists and C does not. Without
2767 -v the output is "A verified" because verification stops after a
2768 successful redirection if more than one address is generated. However,
2769 with -v the child addresses are also verified. Exim was outputting "A
2770 failed to verify" and then showing the successful verification for C,
2771 with its parentage. It now outputs "B failed to verify", showing B's
2772 parentage before showing the successful verification of C.
2774 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2775 look up a TXT record in a specific list after matching in a combined
2778 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2779 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2780 they consult the DNS. I had assumed they would set it the way they
2781 wanted; and indeed my experiments on Linux seem to show that in some
2782 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2783 To be on the safe side, however, I have now made the interface to
2784 host_find_byname() similar to host_find_bydns(), with an argument
2785 containing the DNS resolver options. The host_find_byname() function now
2786 sets these options at its start, just as host_find_bydns() does. The smtp
2787 transport options dns_qualify_single and dns_search_parents are passed to
2788 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2789 of host_find_byname() use the default settings of RES_DEFNAMES
2790 (qualify_single) but not RES_DNSRCH (search_parents).
2792 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2793 spool_read_header() do less string testing, by means of a preliminary
2794 switch on the second character of optional "-foo" lines. (This is
2795 overdue, caused by the large number of possibilities that now exist.
2796 Originally there were few.) While I was there, I also converted the
2797 str(n)cmp tests so they don't re-test the leading "-" and the first
2798 character, in the hope this might squeeze out yet more improvement.
2800 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2801 flag allowing group syntax was set by the header_syntax check but not
2802 turned off, possible causing trouble later; (2) The flag was not being
2803 set at all for the header_verify test, causing "group"-style headers to
2804 be rejected. I have now set it in this case, and also caused header_
2805 verify to ignore an empty address taken from a group. While doing this, I
2806 came across some other cases where the code for allowing group syntax
2807 while scanning a header line wasn't quite right (mostly, not resetting
2808 the flag correctly in the right place). These bugs could have caused
2809 trouble for malformed header lines. I hope it is now all correct.
2811 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2812 with the "reply" argument non-NULL. The code, however (which originally
2813 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2814 but it didn't always do it. This confused somebody who was copying the
2815 code for some other use. I have removed all the tests.
2817 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2818 feature that was used to support insecure browsers during the U.S. crypto
2819 embargo. It requires special client support, and Exim is probably the
2820 only MTA that supported it -- and would never use it because real RSA is
2821 always available. This code has been removed, because it had the bad
2822 effect of slowing Exim down by computing (never used) parameters for the
2823 RSA_EXPORT functionality.
2825 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2826 authenticator to fail if there's a tab character in the incoming data
2827 (there should never be unless someone is messing about, as it's supposed
2828 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2829 if the connection is using TLS or if the remote IP is the same as the
2830 local IP, and the "valid-client-cert option" if a client certificate has
2833 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2834 authenticators. This can be used for authorization after authentication
2835 succeeds. (In the case of plaintext, it servers for both authentication
2838 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2839 if any retry times were supplied.
2841 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2842 connection, where there is no client IP address to check. In this
2843 situation, the verify now always succeeds.
2845 PH/25 Applied John Jetmore's -Mset patch.
2847 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2849 PH/27 In a string expansion for a processed (not raw) header when multiple
2850 headers of the same name were present, leading whitespace was being
2851 removed from all of them, but trailing whitespace was being removed only
2852 from the last one. Now trailing whitespace is removed from each header
2853 before concatenation. Completely empty headers in a concatenation (as
2854 before) are ignored.
2856 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2857 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2859 PH/29 [Removed. This was a change that I later backed out, and forgot to
2860 correct the ChangeLog entry (that I had efficiently created) before
2861 committing the later change.]
2863 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2864 address errors (4xx response to RCPT) over the same connection as other
2865 messages routed to the same hosts. Such deliveries are always "forced",
2866 so retry times are not inspected. This resulted in far too many retries
2867 for the affected addresses. The effect occurred only when there were more
2868 hosts than the hosts_max_try setting in the smtp transport when it had
2869 the 4xx errors. Those hosts that it had tried were not added to the list
2870 of hosts for which the message was waiting, so if all were tried, there
2871 was no problem. Two fixes have been applied:
2873 (i) If there are any address or message errors in an SMTP delivery, none
2874 of the hosts (tried or untried) are now added to the list of hosts
2875 for which the message is waiting, so the message should not be a
2876 candidate for sending over the same connection that was used for a
2877 successful delivery of some other message. This seems entirely
2878 reasonable: after all the message is NOT "waiting for some host".
2879 This is so "obvious" that I'm not sure why it wasn't done
2880 previously. Hope I haven't missed anything, but it can't do any
2881 harm, as the worst effect is to miss an optimization.
2883 (ii) If, despite (i), such a delivery is accidentally attempted, the
2884 routing retry time is respected, so at least it doesn't keep
2885 hammering the server.
2887 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2888 in ${readsocket because some servers need this prod.
2890 PH/32 Added some extra debug output when updating a wait-xxx database.
2892 PH/33 The hint "could be header name not terminated by colon", which has been
2893 given for certain expansion errors for a long time, was not being given
2894 for the ${if def:h_colon_omitted{... case.
2896 PH/34 The spec says: "With one important exception, whenever a domain list is
2897 being scanned, $domain contains the subject domain." There was at least
2898 one case where this was not true.
2900 PH/35 The error "getsockname() failed: connection reset by peer" was being
2901 written to the panic log as well as the main log, but it isn't really
2902 panic-worthy as it just means the connection died rather early on. I have
2903 removed the panic log writing for the ECONNRESET error when getsockname()
2906 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2907 runs only) independently of the message's sender address. This meant
2908 that, if the 4xx error was in fact related to the sender, a different
2909 message to the same recipient with a different sender could confuse
2910 things. In particular, this can happen when sending to a greylisting
2911 server, but other circumstances could also provoke similar problems.
2912 I have changed the default so that the retry time for these errors is now
2913 based a combination of the sender and recipient addresses. This change
2914 can be overridden by setting address_retry_include_sender=false in the
2917 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2918 remote server are returned as part of bounce messages. This was not
2919 happening for LMTP over a pipe (the lmtp transport), but now it is the
2920 same for both kinds of LMTP.
2922 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2923 in header lines that were in fact CNAMEs. This is no longer the case.
2925 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2926 and queue runs started by the daemon processed all messages. This has
2927 been fixed so that -R and -S can now usefully be given with -q<time>.
2929 PH/40 Import PCRE release 6.7 (fixes some bugs).
2931 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2933 PH/42 Give an error if -q is specified more than once.
2935 PH/43 Renamed the variables $interface_address and $interface_port as
2936 $received_ip_address and $received_port, to make it clear that these
2937 values apply to message reception, and not to the outgoing interface when
2938 a message is delivered. (The old names remain recognized, of course.)
2940 PH/44 There was no timeout on the connect() call when using a Unix domain
2941 socket in the ${readsocket expansion. There now is.
2943 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2944 be meaningful with "accept".
2946 SC/01 Eximstats V1.43
2947 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2949 SC/02 Eximstats V1.44
2950 Use a glob alias rather than an array ref in the generated
2951 parser. This improves both readability and performance.
2953 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2954 Collect SpamAssassin and rejection statistics.
2955 Don't display local sender or destination tables unless
2956 there is data to show.
2957 Added average volumes into the top table text output.
2959 SC/04 Eximstats V1.46
2960 Collect data on the number of addresses (recipients)
2961 as well as the number of messages.
2963 SC/05 Eximstats V1.47
2964 Added 'Message too big' to the list of mail rejection
2965 reasons (thanks to Marco Gaiarin).
2967 SC/06 Eximstats V1.48
2968 Mainlog lines which have GMT offsets and are too short to
2969 have a flag are now skipped.
2971 SC/07 Eximstats V1.49 (Alain Williams)
2972 Added the -emptyok flag.
2974 SC/08 Eximstats V1.50
2975 Fixes for obtaining the IP address from reject messages.
2977 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2978 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2979 whitespace changes from 4.64-PH/27
2981 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2984 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2985 are found, allow negative numbers in numeric criteria)
2987 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2989 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2990 to match changes made in 4.64-PH/43
2992 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2994 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2995 "domain" is actually an IP address, and force a failure. This locks out
2996 those revolvers/nameservers that support "A-for-A" lookups, in
2997 contravention of the specifications.
2999 PH/48 When a host name was looked up from an IP address, and the subsequent
3000 forward lookup of the name timed out, the host name was left in
3001 $sender_host_name, contrary to the specification.
3003 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
3004 restricted to single-key lookups, Exim was not diagnosing an error if
3005 * or *@ was used with a query-style lookup.
3007 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
3009 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
3010 long ago, but noone interested enough thought of it. Let's just say that
3011 the "1.1" means that there are some new functions that weren't there at
3012 some point in the past.
3014 PH/51 Error processing for expansion failure of helo_data from an smtp
3015 transport during callout processing was broken.
3017 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3018 tested/used via the -bh/-bhc/-bs options.
3020 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3021 bug, fixed in subsequent PCRE releases).
3023 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3024 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3026 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3032 SC/01 Use a glob alias rather than an array ref in eximstats generated
3033 parser. This improves both readability and performance.
3035 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3036 Don't display local sender or destination tables in eximstats unless
3037 there is data to show.
3038 Added average volumes into the eximstats top table text output.
3040 SC/03 Collect data on the number of addresses (recipients) as well
3041 as the number of messages in eximstats.
3043 TF/01 Correct an error in the documentation for the redirect router. Exim
3044 does (usually) call initgroups() when daemonizing.
3046 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3047 with consistent privilege compared to when running as a daemon.
3049 TF/03 Note in the spec that $authenticated_id is not set for local
3050 submissions from trusted users.
3052 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3053 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3055 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3056 by adding some example configuration directives to the default
3057 configuration file. A little bit of work is required to uncomment the
3058 directives and define how usernames and passwords are checked, but
3059 there is now a framework to start from.
3061 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3062 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3063 without this. I don't know how relevant this is to other LDAP libraries.
3065 PH/02 Add the verb name to the "unknown ACL verb" error.
3067 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3069 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3071 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3072 directory not expanded when it should be if an expanded home directory
3073 was set for the address (which is overridden by the transport).
3075 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3078 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3079 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3080 because it is too late at that time, and has no effect.
3082 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3083 security issue with \' (bugzilla #107). I could not use the
3084 PQescapeStringConn() function, because it needs a PGconn value as one of
3087 PH/08 When testing addresses using -bt, indicate those final addresses that
3088 are duplicates that would not cause an additional delivery. At least one
3089 person was confused, thinking that -bt output corresponded to deliveries.
3090 (Suppressing duplicates isn't a good idea as you lose the information
3091 about possibly different redirections that led to the duplicates.)
3093 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3094 systems where poll() doesn't work, in particular OS X.
3096 PH/10 Added more information to debugging output for retry time not reached.
3098 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3099 operations in malware.c.
3101 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3104 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3105 syslog_duplication set false, log lines that would normally be written
3106 both the the main log and to the reject log were not written to syslog at
3109 PH/14 In the default configuration, change the use of "message" in ACL warn
3110 statements to "add_header".
3112 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3113 not followed by a command (e.g. "seen endif").
3115 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3116 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3119 PH/17 Added extra conditions to the default value of delay_warning_condition
3123 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3124 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3125 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3128 The Auto-Submitted: and various List- headers are standardised, whereas I
3129 don't think Precedence: ever was.
3131 PH/18 Refactored debugging code in route_finduser() to show more information,
3132 in particular, the error code if getpwnam() issues one.
3134 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3135 This is apparently needed in addition to the PH/07 change above to avoid
3136 any possible encoding problems.
3138 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3139 but not after initializing Perl.
3141 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3142 output them only if debugging. By default they are written stderr,
3143 apparently, which is not desirable.
3145 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3148 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3151 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3153 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3154 authenticated or an ident call has been made. Suppress the default
3155 values for $authenticated_id and $authenticated_sender (but permit -oMai
3156 and -oMas) when testing with -bh.
3158 PH/24 Re-jigged the order of the tests in the default configuration so that the
3159 tests for valid domains and recipients precede the DNS black list and CSA
3160 tests, on the grounds that those ones are more expensive.
3162 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3163 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3164 This bug exists in every version of Exim that I still have, right back to
3167 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3168 However, an attempt to turn on case-sensitivity in a regex key by
3169 including (?-i) didn't work because the subject string was already
3170 lowercased, and the effects were non-intuitive. It turns out that a
3171 one-line patch can be used to allow (?-i) to work as expected.
3177 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3178 other effects) broke the use of negated acl sub-conditions.
3180 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3183 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3184 "Deny" causes Exim to reject the incoming connection with a 554 error.
3185 Unfortunately, if there is a major crisis, such as a disk failure,
3186 tcp-wrappers gives "deny", whereas what one would like would be some
3187 kind of temporary error. A kludge has been added to help with this.
3188 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3189 554 error is used if errno is still zero or contains ENOENT (which occurs
3190 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3193 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3195 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3196 errors. Otherwise a message that provokes a temporary error (when other
3197 messages do not) can cause a whole host to time out.
3199 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3200 addresses were routed directly to files or pipes from a redirect router.
3201 File deliveries just didn't batch; pipe deliveries might have suffered
3204 PH/06 A failure to get a lock for a hints database would erroneously always say
3205 "Failed to get write lock", even when it was really a read lock.
3207 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3208 of 0600. This has been changed to use the value of the lockfile_mode
3209 option (which defaults to 0600).
3211 PH/08 Applied small patch from the Sieve maintainer.
3213 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3214 folder from quota calculations, a direct delivery into this folder messed
3215 up the contents of the maildirsize file. This was because the regex was
3216 used only to exclude .Trash (or whatever) when the size of the mailbox
3217 was calculated. There was no check that a delivery was happening into an
3218 excluded directory. This bug has been fixed by ignoring all quota
3219 processing for deliveries into excluded directories.
3221 PH/10 Added the maildirfolder_create_regex option to appendfile.
3227 PH/01 The code for finding all the local interface addresses on a FreeBSD
3228 system running IPv6 was broken. This may well have applied to all BSD
3229 systems, as well as to others that have similar system calls. The broken
3230 code found IPv4 interfaces correctly, but gave incorrect values for the
3231 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3232 that it would not match correctly against @[] and not recognize the IPv6
3235 PH/02 The ipliteral router was not recognizing addresses of the form user@
3236 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3238 PH/03 Added disable_ipv6.
3240 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3241 decoded form, because it is most often used to construct To: headers
3242 lines in autoreplies, and the decoded form may well be syntactically
3243 invalid. However, $reply_address has leading white space removed, and all
3244 newlines turned into spaces so that the autoreply transport does not
3247 PH/05 If group was specified without a user on a router, and no group or user
3248 was specified on a transport, the group from the router was ignored.
3250 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3251 for visible compile-time settings that can be used to change these
3252 numbers, for those that want even more. Backwards compatibility with old
3253 spool files has been maintained. However, going back to a previous Exim
3254 release will lost any variables that are in spool files.
3256 PH/07 Two small changes when running in the test harness: increase delay when
3257 passing a TCP/IP connection to a new process, in case the original
3258 process has to generate a bounce, and remove special handling of
3259 127.0.0.2 (sic), which is no longer necessary.
3261 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3262 be the same on different OS.
3264 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3267 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3268 whether --show-vars was specified or not
3270 JJ/02 exipick: Added support for new ACL variable spool format introduced
3273 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3274 syntactically invalid From: or Reply-to: line, and a filter used this to
3275 generate an autoreply, and therefore failed to obtain an address for the
3276 autoreply, Exim could try to deliver to a non-existent relative file
3277 name, causing unrelated and misleading errors. What now happens is that
3278 it logs this as a hard delivery error, but does not attempt to create a
3281 PH/11 The exinext utility has a -C option for testing purposes, but although
3282 the given file was scanned by exinext itself; it wasn't being passed on
3283 when Exim was called.
3285 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3286 an end-of-file indication when reading a command response.
3288 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3289 compiled. In many other places in Exim, IPv6 addresses are always
3290 recognized, so I have changed this. It also means that IPv4 domain
3291 literals of the form [IPV4:n.n.n.n] are now always recognized.
3293 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3294 used if the router is not running as root, for example, when verifying at
3295 ACL time, or when using -bh. The debugging output from this situation was
3296 non-existent - all you got was a failure to exec. I have made two
3299 (a) Failures to set uid/gid, the current directory, or a process leader
3300 in a subprocess such as that created by queryprogram now generate
3301 suitable debugging output when -d is set.
3303 (b) The queryprogram router detects when it is not running as root,
3304 outputs suitable debugging information if -d is set, and then runs
3305 the subprocess without attempting to change uid/gid.
3307 PH/15 Minor change to Makefile for building test_host (undocumented testing
3310 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3311 additional section of a DNS packet that returns MX or SRV records.
3312 Instead, it always explicitly searches for A/AAAA records. This avoids
3313 major problems that occur when a DNS server includes only records of one
3314 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3315 fixed another bug: if SRV records were looked up and the corresponding
3316 address records were *not* found in the additional section, the port
3317 values from the SRV records were lost.
3319 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3320 using the correct key (the original address) when searching the retry
3321 rules in order to find which one to use for generating the retry hint.
3323 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3324 adding the default one. Similarly, if it contains a Reply-To: header, the
3325 errors_reply_to option, if set, is not used.
3327 PH/19 When calculating a retry time, Exim used to measure the "time since
3328 failure" by looking at the "first failed" field in the retry record. Now
3329 it does not use this if it is later than than the arrival time of the
3330 message. Instead it uses the arrival time. This makes for better
3331 behaviour in cases where some deliveries succeed, thus re-setting the
3332 "first failed" field. An example is a quota failure for a huge message
3333 when small messages continue to be delivered. Without this change, the
3334 "time since failure" will always be short, possible causing more frequent
3335 delivery attempts for the huge message than are intended.
3336 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3338 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3339 $1, $2, $3) because the numerical variables can be reset during some
3340 expansion items (e.g. "match"), thereby losing the authentication data.
3342 PH/21 Make -bV show the size of off_t variables so that the test suite can
3343 decide whether to run tests for quotas > 2G.
3345 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3346 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3347 filecount value is greater than 2G or if a quota value is greater than 2G
3348 on a system where the size of off_t is not greater than 4, a panic error
3351 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3352 never match. The debug and -bh output now contains an explicit error
3353 message indicating a malformed IPv4 address or mask.
3355 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3356 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3357 PH/23 above applies.
3359 PH/25 Do not write to syslog when running in the test harness. The only
3360 occasion when this arises is a failure to open the main or panic logs
3361 (for which there is an explicit test).
3363 PH/26 Added the /no_tell option to "control=freeze".
3365 PH/27 If a host name lookup failed very early in a connection, for example, if
3366 the IP address matched host_lookup and the reverse lookup yielded a name
3367 that did not have a forward lookup, an error message of the form "no IP
3368 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3369 could be logged. Now it outputs the IP address instead of "NULL".
3371 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3372 allows the sender and the authenticated sender to be set when
3373 submitting a message from within Exim. Since child_open_exim() is
3374 documented for local_scan(), the new function should be too.
3376 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3377 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3378 results in an empty string is now treated as unset.
3380 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3382 PH/31 Added sender_verify_fail logging option.
3384 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3385 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3386 tidied the source and removed it altogether.
3388 PH/33 When a queue run was abandoned because the load average was too high, a
3389 log line was always written; now it is written only if the queue_run log
3390 selector is set. In addition, the log line for abandonment now contains
3391 information about the queue run such as the pid. This is always present
3392 in "start" and "stop" lines but was omitted from the "abandon" line.
3394 PH/34 Omit spaces between a header name and the colon in the error message that
3395 is given when verify = headers_syntax fails (if there are lots of them,
3396 the message gets confusing).
3398 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3399 names, as there are now some PTR records that contain slashes. This check
3400 is only to protect against broken name servers that fall over on strange
3401 characters, so the fact that it applies to all lookups doesn't matter.
3403 PH/36 Now that the new test suite is complete, we can remove some of the
3404 special code in Exim that was needed for the old test suite. For example,
3405 sorting DNS records because real resolvers return them in an arbitrary
3406 order. The new test suite's fake resolver always returns records in the
3409 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3410 bounces) except when queue_only is set, to avoid logging races between
3411 the different processes.
3413 PH/38 Panic-die if .include specifies a non-absolute path.
3415 PH/39 A tweak to the "H" retry rule from its user.
3417 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3418 a label. They prevented compilation on older perls.
3420 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3421 a warning to be raised on newish perls.
3423 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3424 on queue. Changes to match documented behaviour of showing count of
3425 messages matching specified criteria.
3427 PH/40 Changed the default ident timeout from 30s to 5s.
3429 PH/41 Added support for the use of login_cap features, on those BSD systems
3430 that have them, for controlling the resources used by pipe deliveries.
3432 PH/42 The content-scanning code uses fopen() to create files in which to put
3433 message data. Previously it was not paying any attention to the mode of
3434 the files. Exim runs with umask(0) because the rest of the code creates
3435 files with open(), and sets the required mode explicitly. Thus, these
3436 files were ending up world-writeable. This was not a big issue, because,
3437 being within the spool directory, they were not world-accessible. I have
3438 created a function called modefopen, which takes an additional mode
3439 argument. It sets umask(777), creates the file, chmods it to the required
3440 mode, then resets the umask. All the relevant calls to fopen() in the
3441 content scanning code have been changed to use this function.
3443 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3444 to 24 hours. This avoids potential overflow problems when processing G
3445 and H retry rules. I suspect nobody ever tinkers with this value.
3447 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3449 PH/45 When the plaintext authenticator is running as a client, the server's
3450 challenges are checked to ensure they are valid base64 strings. By
3451 default, the authentication attempt is cancelled if an invalid string is
3452 received. Setting client_ignore_invalid_base64 true ignores these errors.
3453 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3454 they are received. Thus, the responses can be made to depend on the
3455 challenges. If an invalid string is ignored, an empty string is placed in
3458 PH/46 Messages that are created by the autoreply transport now contains a
3459 References: header, in accordance with RFCs 2822 and 3834.
3461 PH/47 Added authenticated_sender_force to the smtp transport.
3463 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3465 PH/49 Installed latest patch from the Sieve maintainer.
3467 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3468 was also set, the appendfile transport was unnecessarily scanning a
3469 directory of message files (e.g. for maildir delivery) to find the count
3470 of files (along with the size), even though it did not need this
3471 information. It now does the scan only if it needs to find either the
3472 size of the count of files.
3474 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3476 PH/52 Two bugs concerned with error handling when the smtp transport is
3479 (i) Exim was not creating retry information for temporary errors given
3480 for individual recipients after the DATA command when the smtp transport
3481 was used in LMTP mode. This meant that they could be retried too
3482 frequently, and not timed out correctly.
3484 (ii) Exim was setting the flag that allows error details to be returned
3485 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3486 recipients that were returned after the DATA command.
3488 PH/53 This is related to PH/52, but is more general: for any failing address,
3489 when detailed error information was permitted to be returned to the
3490 sender, but the error was temporary, then after the final timeout, only
3491 "retry timeout exceeded" was returned. Now it returns the full error as
3492 well as "retry timeout exceeded".
3494 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3495 do this, and (what is worse) MTAs that accept it.
3497 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3498 will now be deprecated.
3500 PH/56 New os.c-cygwin from the Cygwin maintainer.
3502 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3503 formats (previously only available in exim formats via -bpr, -bpru,
3504 and -bpra. Now also available in native and exiqgrep formats)
3506 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3507 with very large, slow to parse queues
3509 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3511 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3513 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3514 responses to authentication challenges, though it was showing the
3515 challenges; (ii) I've removed the CR characters from the debug output for
3518 PH/58 Allow for the insertion of a newline as well as a space when a string
3519 is turned into more than one encoded-word during RFC 2047 encoding. The
3520 Sieve code now uses this.
3522 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3523 data_4xx, lost_connection, tls_required.
3525 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3526 message was being sent as an SMTP response.
3528 PH/61 Add -l and -k options to exicyclog.
3530 PH/62 When verifying, if an address was redirected to one new address, so that
3531 verification continued, and the new address failed or deferred after
3532 having set something in $address_data, the value of $address_data was not
3533 passed back to the ACL. This was different to the case when no
3534 redirection occurred. The value is now passed back in both cases.
3536 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3537 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3538 use login_cap.h, so on its own it isn't the distinguishing feature. The
3539 new name refers directly to the setclassresources() function.
3541 PH/65 Added configuration files for NetBSD3.
3543 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3545 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3546 is preferred over IPv4.
3548 PH/68 The bounce_return_message and bounce_return_body options were not being
3549 honoured for bounces generated during the reception of non-SMTP messages.
3550 In particular, this applied to messages rejected by the ACL. This bug has
3551 been fixed. However, if bounce_return_message is true and bounce_return_
3552 body is false, the headers that are returned for a non-SMTP message
3553 include only those that have been read before the error was detected.
3554 (In the case of an ACL rejection, they have all been read.)
3556 PH/69 The HTML version of the specification is now built in a directory called
3557 spec_html instead of spec.html, because the latter looks like a path with
3558 a MIME-type, and this confuses some software.
3560 PH/70 Catch two compiler warnings in sieve.c.
3562 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3563 function verify_get_ident() calls ip_connect() to connect a socket, but
3564 if the "connect()" function timed out, ip_connect() used to close the
3565 socket. However, verify_get_ident() also closes the socket later, and in
3566 between Exim writes to the log, which may get opened at this point. When
3567 the socket was closed in ip_connect(), the log could get the same file
3568 descriptor number as the socket. This naturally causes chaos. The fix is
3569 not to close the socket in ip_connect(); the socket should be closed by
3570 the function that creates it. There was only one place in the code where
3571 this was missing, in the iplookup router, which I don't think anybody now
3572 uses, but I've fixed it anyway.
3574 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3575 well as to direct DNS lookups. Otherwise the handling of names in host
3576 lists is inconsistent and therefore confusing.
3582 PH/01 Two changes to the default runtime configuration:
3584 (1) Move the checks for relay_from_hosts and authenticated clients from
3585 after to before the (commented out) DNS black list checks.
3587 (2) Add control=submission to the relay_from_hosts and authenticated
3588 clients checks, on the grounds that messages accepted by these
3589 statements are most likely to be submissions.
3591 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3593 (1) Generate an error if the third argument for the ${prvs expansion is
3596 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3599 (3) Reset the variables that are obtained from the first argument of
3600 ${prvscheck and used in the second argument before leaving the code,
3601 because their memory is reclaimed, so using them afterwards may do
3604 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3605 one (it's much easier than Tom thought :-).
3607 (5) Because of (4), we can now allow for the use of $prvscheck_result
3608 inside the third argument.
3610 PH/03 For some reason, the default setting of PATH when running a command from
3611 a pipe transport was just "/usr/bin". I have changed it to
3614 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3615 anything to be listed in the output from -bV.
3617 PH/05 When a filter generated an autoreply, the entire To: header line was
3618 quoted in the delivery log line, like this:
3620 => >A.N.Other <ano@some.domain> <original@ddress> ...
3622 This has been changed so that it extracts the operative address. There
3623 may be more than one such address. If so, they are comma-separated, like
3626 => >ano@some.domain,ona@other.domain <original@ddress> ...
3628 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3629 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3630 not being looked up in the rDNS to get a host name, Exim was showing the
3631 IP address twice in Received: lines, even though the IP addresses were
3632 identical. For example:
3634 Received: from [1.2.3.4] (helo=[1.2.3.4])
3636 However, if the real host name was known, it was omitting the HELO data
3637 if it matched the actual IP address. This has been tidied up so that it
3638 doesn't show the same IP address twice.
3640 PH/07 When both +timestamp and +memory debugging was on, the value given by
3641 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3642 called by the debug printing, thereby overwriting the timestamp buffer.
3643 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3645 PH/08 When the original message was included in an autoreply transport, it
3646 always said "this is a copy of the message, including all the headers",
3647 even if body_only or headers_only was set. It now gives an appropriate
3650 PH/09 Applied a patch from the Sieve maintainer which:
3652 o fixes some comments
3653 o adds the (disabled) notify extension core
3654 o adds some debug output for the result of if/elsif tests
3655 o points to the current vacation draft in the documentation
3656 and documents the missing references header update
3660 o fixes a bug in processing the envelope test (when testing
3661 multiple envelope elements, the last element determined the
3664 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3665 Electronic Mail") by including:
3667 Auto-submitted: auto-generated
3669 in the messages that it generates (bounce messages and others, such as
3670 warnings). In the case of bounce messages for non-SMTP messages, there was
3671 also a typo: it was using "Auto_submitted" (underscore instead of
3672 hyphen). Since every message generated by Exim is necessarily in response
3673 to another message, thes have all been changed to:
3675 Auto-Submitted: auto-replied
3677 in accordance with these statements in the RFC:
3679 The auto-replied keyword:
3681 - SHOULD be used on messages sent in direct response to another
3682 message by an automatic process,
3684 - MUST NOT be used on manually-generated messages,
3686 - MAY be used on Delivery Status Notifications (DSNs) and Message
3687 Disposition Notifications (MDNs),
3689 - MUST NOT be used on messages generated by automatic or periodic
3690 processes, except for messages which are automatic responses to
3693 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3694 to the default Received: header definition.
3696 PH/12 Added log selector acl_warn_skipped (default on).
3698 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3699 variables because (a) they are in the wrong storage pool and (b) even if
3700 they were copied, it wouldn't work properly because of the caching.
3702 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3703 checking when decoding. Apparently there are clients that generate
3704 overlong encoded strings. Why am I not surprised?
3706 PH/15 If the first argument of "${if match_address" was not empty, but did not
3707 contain an "@" character, Exim crashed. Now it writes a panic log message
3708 and treats the condition as false.
3710 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3712 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3713 extension "envelope-auth". The code is finished and in agreement with
3714 other implementations, but there is no documentation so far and in fact,
3715 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3716 not changing the active code.
3718 Print executed "if" and "elsif" statements when debugging is used. This
3719 helps a great deal to understand what a filter does.
3721 Document more things not specified clearly in RFC3028. I had all this
3722 sorted out, when out of a sudden new issues came to my mind. Oops."
3724 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3727 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3728 canonical form (as documented). However, after a host name lookup from
3729 the IP address, check_host() was doing a simple string comparison with
3730 addresses acquired from the DNS when checking that the found name did
3731 have the original IP as one of its addresses. Since any found IPv6
3732 addresses are likely to be in abbreviated form, the comparison could
3733 fail. Luckily, there already exists a function for doing the comparison
3734 by converting both addresses to binary, so now that is used instead of
3735 the text comparison.
3737 PH/20 There was another similar case to PH/19, when a complete host name was
3738 given in a host list; looking up its IP address could give an abbreviated
3739 form, whereas the current host's name might or might not be abbreviated.
3740 The same fix has been applied.
3746 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3747 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3750 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3751 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3753 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3755 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3756 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3757 Also, for "Auto-Submitted", treat anything other than "no" as
3758 non-personal, in accordance with RFC 3834. (Previously it treated
3759 anything starting "auto-" as non-personal.)
3761 TF/01 The control=submission/name=... option had a problem with syntax
3762 errors if the name included a slash character. The /name= option
3763 now slurps the rest of the string, so it can include any characters
3764 but it must come last in the list of options (after /sender_retain
3767 PH/05 Some modifications to the interface to the fake nameserver for the new
3775 TK/01 Added the "success_on_redirect" address verification option. See
3776 NewStuff for rationale and an example.
3778 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3780 PH/02 Patch to exigrep to allow it to work on syslog lines.
3782 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3783 fread() to skip over the body file's header line, because in Cygwin the
3784 header line is locked and is inaccessible.
3786 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3787 co-exist for some time) to make it clear that it is the Exim ID that is
3788 referenced, not the Message-ID: header line.
3790 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3791 string_format() function, because snprintf() does not exist on all
3794 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3795 ${stat: expansion item.
3797 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3798 protocol synchronization error", to keep the pedants happy.
3800 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3801 well as for IRIX systems, when gcc is being used. See the host.c source
3804 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3806 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3809 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3810 transport and to the smtp transport in LMTP mode.
3812 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3814 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3815 run a filter in a subprocess. This could lead to confusion in subsequent
3816 lookups in the parent process. There should also be a search_tidyup() at
3817 the end of the subprocess.
3819 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3820 only if the host matched helo_try_verify_hosts, which caused the
3821 verification to occur when the EHLO/HELO command was issued. The ACL just
3822 tested the remembered result. Now, if a previous verification attempt has
3823 not happened, "verify = helo" does it there and then.
3825 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3827 TK/03 Fix log output including CR from clamd.
3829 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3830 exist provoked a memory error which could cause a segfault.
3832 PH/15 Installed PCRE 6.2
3834 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3836 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3837 of the problem. Specifically, suggested +O2 rather than +O1 for the
3840 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3842 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3843 record, the port was not correctly logged when the outgoing_port log
3844 selector was set (it logged the transort's default port).
3846 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3847 fallback_hosts, and "hosts" in the smtp transport.
3849 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3850 host errors such as "Connection refused".
3852 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3853 authentication with radiusclient 0.4.9:
3855 - Error returned from rc_read_config was caught wrongly
3856 - Username/password not passed on to radius server due to wrong length.
3858 The presumption is that some radiusclient API changes for 4.51/PH/17
3859 were not taken care of correctly. The code is still untested by me (my
3860 Linux distribution still has 0.3.2 of radiusclient), but it was
3861 contributed by a Radius user.
3863 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3864 expanding the "port" option of the smtp transport.
3866 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3867 while reading a MIME header. Thanks to Tom Hughes for a patch.
3869 PH/24 Include config.h inside local_scan.h so that configuration settings are
3872 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3873 that in an ACL for RCPT (for example), you can examine exactly what was
3876 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3877 commands, but it was not correctly comparing the address with the actual
3878 client host address. Thus, it would show the EHLO address in Received:
3879 header lines when this was not necessary.
3881 PH/27 Added the % operator to ${eval:}.
3883 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3884 it should be ignoring failures (because with -C, for example, it has lost
3885 privilege). It wasn't ignoring creation failures other than "already
3888 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3891 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3892 because an input line was too long, either on its own, or by virtue of
3893 too many continuations, the temporary file was not being removed, and the
3894 return code was incorrect.
3896 PH/31 Missing "BOOL" in function definition in filtertest.c.
3898 PH/32 Applied Sieve patches from the maintainer.
3900 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3902 PH/33 Added "verify = not_blind".
3904 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3905 Local/Makefile (with some defaults set). These are used in built scripts
3906 such as exicyclog, but they have never been used in the exim_install
3907 script (though there are many overriding facilities there). I have
3908 arranged that the exim_install script now takes note of these two
3911 PH/35 Installed configuration files for Dragonfly.
3913 PH/36 When a locally submitted message by a trusted user did not contain a
3914 From: header, and the sender address was obtained from -f or from an SMTP
3915 MAIL command, and the trusted user did not use -F to supply a sender
3916 name, $originator_name was incorrectly used when constructing a From:
3917 header. Furthermore, $originator_name was used for submission mode
3918 messages from external hosts without From: headers in a similar way,
3919 which is clearly wrong.
3921 PH/37 Added control=suppress_local_fixups.
3923 PH/38 When log_selector = +received_sender was set, and the addition of the
3924 sender made the log line's construction buffer exactly full, or one byte
3925 less than full, an overflow happened when the terminating "\n" was
3928 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3929 when the result of a list match is failure because a DNS lookup failed.
3931 PH/40 RM_COMMAND is now used in the building process.
3933 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3934 the "build-* directories that it finds.
3936 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3937 address in a domain literal was a prefix of an interface address.
3939 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3940 when verifying a sender address, unless rewrite_headers is false.
3942 PH/44 Wrote a long comment about why errors_to addresses are verified as
3943 recipients, not senders.
3945 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3946 the ratelimit ACL was added.
3948 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3950 PH/46 Added extra information about PostgreSQL errors to the error string.
3952 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3953 avoiding the need to install special zones in a real server. This is
3954 backwards compatible; if it can't find the fake resolver, it drops back.
3955 Thus, both old and new test suites can be run.
3957 TF/02 Added util/ratelimit.pl
3959 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3960 clock is set back in time.
3962 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3963 Candler <B.Candler@pobox.com>.
3965 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3966 OK for addresses that are the result of redirections.
3968 PH/48 A number of further additions for the benefit of the new test suite,
3969 including a fake gethostbyname() that interfaces to the fake DNS resolver
3972 TF/06 The fix for widen_domains has also been applied to qualify_single and
3973 search_parents which are the other dnslookup options that can cause
3976 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3979 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3981 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3982 service" if it encountered an MX record with an empty target hostname.
3983 The message is now "an MX or SRV record indicated no SMTP service".
3985 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3986 if the DNS of the sending site is misconfigured. This is quite a
3987 common situation. This change restores the behaviour of treating a
3988 helo verification defer as a failure.
3990 PH/51 If self=fail was set on a router, the bounce message did not include the
3991 actual error message.
3997 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3999 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
4000 command itself ended in error, the underlying message about the transport
4001 filter timeout was being overwritten with the pipe command error. Now the
4002 underlying error message should be appended to the second error message.
4004 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
4006 PH/02 Reduce the amount of output that "make" produces by default. Full output
4007 can still be requested.
4009 PH/03 The warning log line about a condition test deferring for a "warn" verb
4010 was being output only once per connection, rather than after each
4011 occurrence (because it was using the same function as for successful
4012 "warn" verbs). This seems wrong, so I have changed it.
4014 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4015 it should not have, which might have caused a crash in the right
4016 circumstances, but probably never did.
4018 PH/04 Installed a modified version of Tony Finch's patch to make submission
4019 mode fix the return path as well as the Sender: header line, and to
4020 add a /name= option so that you can make the user's friendly name appear
4023 TF/03 Added the control = fakedefer ACL modifier.
4025 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4026 Mark Lowes for thorough testing.
4028 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4030 TK/03 Merged latest SRS patch from Miles Wilton.
4032 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4033 with the definition in sysexits.h (which is #included earlier).
4034 Fortunately, Exim does not actually use EX_OK. The code used to try to
4035 preserve the sysexits.h value, by assuming that macro definitions were
4036 scanned for macro replacements. I have been disabused of this notion,
4037 so now the code just undefines EX_OK before #including unistd.h.
4039 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4040 in the smtp transport. When a block could not be written in a single
4041 write() function, the timeout was being re-applied to each part-write.
4042 This seems wrong - if the receiver was accepting one byte at a time it
4043 would take for ever. The timeout is now adjusted when this happens. It
4044 doesn't have to be particularly precise.
4046 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4047 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4049 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4050 without a domain if the check to <postmaster@domain> fails.
4052 SC/01 Eximstats: added -xls and the ability to specify output files
4053 (patch written by Frank Heydlauf).
4055 SC/02 Eximstats: use FileHandles for outputting results.
4057 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4059 SC/04 Eximstats: fixed display of large numbers with -nvr option
4061 SC/05 Eximstats: fixed merging of reports with empty tables.
4063 SC/06 Eximstats: added the -include_original_destination flag
4065 SC/07 Eximstats: removed tabs and trailing whitespace.
4067 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4069 TK/06 MBOX spool code: Add real "From " MBOX separator line
4070 so the .eml file is really in mbox format (even though
4071 most programs do not really care). Patch from Alex Miller.
4073 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4074 The latter is generated from $received_to and is only set if the
4075 message has one envelope recipient. SA can use these headers,
4076 obviously out-of-the-box. Patch from Alex Miller.
4078 PH/08 The ${def test on a variable was returning false if the variable's
4079 value was "0", contrary to what the specification has always said!
4080 The result should be true unless the variable is empty.
4082 PH/09 The syntax error of a character other than { following "${if
4083 def:variable_name" (after optional whitespace) was not being diagnosed.
4084 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4085 accidental colon was present, for example, could give incorrect results.
4087 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4088 result is used (not including appendfile, where other changes are about
4091 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4092 This involved changing a lot of size variables from int to off_t. It
4093 should work with maildirs and everything.
4095 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4096 spamd dying while we are connected to it.
4098 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4101 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4102 function for BDB 4.3.
4104 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4106 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4107 the sources for PCRE have been re-arranged and more files are now
4110 PH/15 The code I had for printing potentially long long variables in PH/11
4111 above was not the best (it lost precision). The length of off_t variables
4112 is now inspected at build time, and an appropriate printing format (%ld
4113 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4114 to be "long long int" or "long int". This is needed for the internal
4115 formatting function string_vformat().
4117 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4118 the configuration file to be ":syslog", then the script "guesses" where
4119 the logs files are, rather than using the compiled in default. In our
4120 case the guess is not the same as the compiled default, so the script
4121 suddenly stopped working when I started to use syslog. The patch checks
4122 to see if log_file_path is "". If so, it attempts to read it from exim
4123 with no configuration file to get the compiled in version, before it
4124 falls back to the previous guessing code."
4126 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4127 implementing BATV in an Exim configuration. See NewStuff for the gory
4130 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4131 Makefile that are specific to HP-UX.
4133 PH/18 If the "use_postmaster" option was set for a recipient callout together
4134 with the "random" option, the postmaster address was used as the MAIL
4135 FROM address for the random test, but not for the subsequent recipient
4136 test. It is now used for both.
4138 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4139 patch removes a few documentation additions to RFC 3028, because the
4140 latest draft now contains them. It adds the new en;ascii-case comparator
4141 and a new error check for 8bit text in MIME parts. Comparator and
4142 require names are now matched exactly. I enabled the subaddress
4143 extension, but it is not well tested yet (read: it works for me)."
4145 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4146 rework some of the code of TK/09 above to avoid the hardwired use of
4147 "%lld" and "long long". Replaced the call to snprintf() with a call to
4150 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4151 records point to non-existent hosts", "retry timeout exceeded", and
4152 "retry time not reached for any host after a long failure period".
4154 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4155 experimental DomainKeys support:
4157 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4158 (2) On an error such as an illegally used "control", the wrong name for
4159 the control was given.
4161 These problems did NOT occur unless DomainKeys support was compiled.
4163 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4165 PH/24 Added ${if match_ip condition.
4167 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4168 databases so that it will be absolutely obvious if a crash occurs in the
4169 DB library. This is a regular occurrence (often caused by mis-matched
4172 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4173 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4174 picked up on a user's system that detects such things. There doesn't seem
4175 to be a gcc warning option for this - only an attribute that has to be
4176 put on the function's prototype. It seems that in Fedora Core 4 they have
4177 set this on a number of new functions. No doubt there will be more in due
4180 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4181 specify a transport. However, if an address that was verified by such a
4182 router was the subject of a callout, Exim crashed because it tried to
4183 read the rcpt_include_affixes from the non-existent transport. Now it
4184 just assumes that the setting of that option is false. This bug was
4185 introduced by 4.51/PH/31.
4187 PH/28 Changed -d+all to exclude +memory, because that information is very
4188 rarely of interest, but it makes the output a lot bigger. People tend to
4189 do -d+all out of habit.
4191 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4192 code in os-type was giving problems when libc.so lives in lib64, like on
4195 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4196 aren't the modern standard, and it seems that some systems' include files
4197 don't always have them. Exim was already checking for some of the newer
4198 ones like T_AAAA, and defining it itself. I've added checks for all the
4199 record types that Exim uses.
4201 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4202 not automatically generating a new one, as it is supposed to. This
4203 prevented TLS from working. If the file did exist, but contained invalid
4204 data, a new version was generated, as expected. It was only the case of a
4205 non-existent file that was broken.
4207 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4208 with a change in libdomainkeys > 0.64.
4210 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4211 from DNS. If the selector record carries the flag, it now has
4212 precedence over the domain-wide flag.
4214 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4216 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4217 the use of a port name that isn't defined in /etc/services) occurred, the
4218 message was deferred as in a normal delivery, and thus remained on the
4219 spool, instead of being failed because of the mua_wrapper setting. This
4220 is now fixed, and I tidied up some of the mua_wrapper messages at the
4223 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4224 the messages in a hash of arrays rather than using individual hashes.
4225 This is a bit cleaner and results in dramatic memory savings, albeit
4226 at a slight CPU cost.
4228 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4229 as requested by Marc Sherman.
4231 SC/10 Eximstats: added histograms for user specified patterns as requested
4234 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4236 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4237 fopen() in the content-scanning modules that did not already have it.
4243 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4244 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4246 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4248 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4250 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4251 to test Sieve filters that use "vacation".
4253 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4254 that changes the way the GnuTLS parameters are stored in the cache file.
4255 The new format can be generated externally. For backward compatibility,
4256 if the data in the cache doesn't make sense, Exim assumes it has read an
4257 old-format file, and it generates new data and writes a new file. This
4258 means that you can't go back to an older release without removing the
4261 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4262 work if there are any delivery delays because "one_time" forces the
4263 parent to be marked "delivered", so its unseen clone is never tried
4264 again. For this reason, Exim now forbids the simultaneous setting of
4267 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4268 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4269 entry if you want to know the details. The fix, however, neglected to
4270 consider the case when local delivery batching is involved. The test for
4271 "previously delivered" was not happening when checking to see if an
4272 address could be batched with a previous (undelivered) one; under
4273 certain circumstances this could lead to multiple deliveries to the same
4276 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4277 in its include files, and this causes problems building Exim.
4279 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4280 header_syntax) but Exim was just ignoring anything given after a slash.
4281 In particular, this caused confusion with an attempt to use "verify =
4282 reverse_host_lookup/defer_ok". An error is now given when options are
4283 supplied for verify items that do not have them. (Maybe reverse_host_
4284 lookup should have a defer_ok option, but that's a different point.)
4286 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4287 defined by RFC 821) to 2048, because there were problems with some AUTH
4288 commands, and RFC 1869 says the size should be increased for extended
4289 SMTP commands that take arguments.
4291 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4294 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4295 "unknown" error; now it says that the functionality isn't in the binary.
4297 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4298 an address' error message when a string expansion fails (syntax or
4299 whatever). Otherwise the password may appear in the log. Following change
4300 PH/42 below, there is no longer a chance of it appearing in a bounce
4303 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4305 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4306 of its addresses was ever tried. (Bugzilla bug #2.)
4308 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4309 the result incorrectly in the debug output. (It correctly added a newline
4310 to what was transported.)
4312 TF/01 Added $received_time.
4314 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4315 commented out examples of how to interface to a virus scanner and to
4316 SpamAssassin. Also added commented examples of av_scanner and
4317 spamd_address settings.
4319 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4320 and controls are allowed in which ACLs. There were a couple of minor
4321 errors. Some of the entries in the conditions table (which is a table of
4322 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4323 negation of where the condition IS allowed.
4325 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4327 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4328 header file does not have a version number, so I've had to invent a new
4329 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4330 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4331 radiusclient), but it was contributed by a Radius user.
4333 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4334 files or directories, for OpenSSL.
4336 PH/19 When an Exim process that is running as root has to create an Exim log
4337 file, it does so in a subprocess that runs as exim:exim so as to get the
4338 ownership right at creation (otherwise, other Exim processes might see
4339 the file with the wrong ownership). There was no test for failure of this
4340 fork() call, which would lead to the process getting stuck as it waited
4341 for a non-existent subprocess. Forks do occasionally fail when resources
4342 run out. I reviewed all the other calls to fork(); they all seem to check
4345 PH/20 When checking for unexpected SMTP input at connect time (before writing
4346 the banner), Exim was not dealing correctly with a non-positive return
4347 from the read() function. If the client had disconnected by this time,
4348 the result was a log entry for a synchronization error with an empty
4349 string after "input=" when read() returned zero. If read() returned -1
4350 (an event I could not check), uninitialized data bytes were printed.
4351 There were reports of junk text (parts of files, etc) appearing after
4354 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4356 PH/22 Added support for macro redefinition, and (re)definition in between
4357 driver and ACL definitions.
4359 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4360 forgetting to use the resulting value; it was using the unexpanded value.
4362 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4363 hadn't been configured. The fix is from Juergen Kreileder, who
4364 understands it better than I do:
4366 "Here's what I see happening with three configured cyrus_sasl
4367 authenticators configured (plain, login, cram-md5):
4369 On startup auth_cyrus_sasl_init() gets called for each of these.
4370 This means three calls to sasl_listmech() without a specified mech_list.
4371 => SASL tests which mechs of all available mechs actually work
4372 => three warnings about OTP not working
4373 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4375 With the patch, sasl_listmech() also gets called three times. But now
4376 SASL's mech_list option is set to the server_mech specified in the the
4377 authenticator. Or in other words, the answer from sasl_listmech()
4378 gets limited to just the mech you're testing for (which is different
4380 => the return list contains just 'plain' or 'login', 'cram-md5' or
4381 nothing depending on the value of ob->server_mech.
4383 I've just tested the patch: Authentication still works fine,
4384 unavailable mechs specified in the exim configuration are still
4385 caught, and the auth.log warnings about OTP are gone."
4387 PH/25 When debugging is enabled, the contents of the command line are added
4388 to the debugging output, even when log_selector=+arguments is not
4391 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4392 answer is "GNU", and only if the return is "GNU/something" is the answer
4395 PH/27 $acl_verify_message is now set immediately after the failure of a
4396 verification in an ACL, and so is available in subsequent modifiers. In
4397 particular, the message can be preserved by coding like this:
4399 warn !verify = sender
4400 set acl_m0 = $acl_verify_message
4402 Previously, $acl_verify_message was set only while expanding "message"
4403 and "log_message" when a very denied access.
4405 PH/28 Modified OS/os.c-Linux with
4407 -#ifndef OS_LOAD_AVERAGE
4408 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4410 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4411 nomenclature these days.)
4413 PH/29 Installed patch from the Sieve maintainer that adds the options
4414 sieve_useraddress and sieve_subaddress to the redirect router.
4416 PH/30 In these circumstances:
4417 . Two addresses routed to the same list of hosts;
4418 . First host does not offer TLS;
4419 . First host accepts first address;
4420 . First host gives temporary error to second address;
4421 . Second host offers TLS and a TLS session is established;
4422 . Second host accepts second address.
4423 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4424 and peerdn, if requested) that were in fact used only for the second
4427 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4428 attention to any local part prefix or suffix that was matched by the
4429 router that accepted the address. It now behaves in the same way as it
4430 does for delivery: the affixes are removed from the local part unless
4431 rcpt_include_affixes is set on the transport.
4433 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4434 timeout during the DATA phase of an incoming message.
4436 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4437 applied a patch sanctioned by the Sieve maintainer.
4439 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4440 the uid or gid is negative. A case of a negative gid caused this to be
4441 noticed. The fix allows for either to be negative.
4443 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4444 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4447 PH/36 The change PH/12 above was broken. Fixed it.
4449 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4450 the grounds that routing the same address twice would always produce the
4451 same answer. This might have been true once, but it is certainly no
4452 longer true now. Routing a child address may depend on the previous
4453 routing that produced that child. Some complicated redirection strategies
4454 went wrong when messages had multiple recipients, and made Exim's
4455 behaviour dependent on the order in which the addresses were given.
4457 I have moved the duplicate checking until after the routing is complete.
4458 Exim scans the addresses that are assigned to local and remote
4459 transports, and removes any duplicates. This means that more work will be
4460 done, as duplicates will always all be routed, but duplicates are
4461 presumably rare, so I don't expect this is of any significance.
4463 For deliveries to pipes, files, and autoreplies, the duplicate checking
4464 still happens during the routing process, since they are not going to be
4467 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4468 It corrects a timeout issue with spamd. This is Ian's comment: "The
4469 background is that sometimes spamd either never reads data from a
4470 connection it has accepted, or it never writes response data. The exiscan
4471 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4472 blindly assumes that writes won't block so it may never time out."
4474 PH/39 Allow G after quota size as well as K and M.
4476 PH/40 The value set for $authenticated_id in an authenticator may not contain
4477 binary zeroes or newlines because the value is written to log lines and
4478 to spool files. There was no check on this. Now the value is run through
4479 the string_printing() function so that such characters are converted to
4480 printable escape sequences.
4482 PH/41 $message_linecount is a new variable that contains the total number of
4483 lines in the message. Compare $body_linecount, which is the count for the
4486 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4487 bounce and delay warning messages, except in certain special cases, which
4490 (a) An SMTP error message from a remote host;
4491 (b) A message specified in a :fail: redirection;
4492 (c) A message specified in a "fail" command in a system filter;
4493 (d) A message specified in a FAIL return from the queryprogram router;
4494 (e) A message specified by the cannot_route_message router option.
4496 In these cases only, Exim does include the error details in bounce and
4497 warning messages. There are also a few cases where bland messages such
4498 as "unrouteable address" or "local delivery error" are given.
4500 PH/43 $value is now also set for the "else" part of a ${run expansion.
4502 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4503 being worked on, but at least Exim now implements the latest version to
4506 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4507 process to complete was treated as a delivery failure, a timeout while
4508 writing the message to the pipe was logged, but erroneously treated as a
4509 successful delivery. Such timeouts include transport filter timeouts. For
4510 consistency with the overall process timeout, these timeouts are now
4511 treated as errors, giving rise to delivery failures by default. However,
4512 there is now a new Boolean option for the pipe transport called
4513 timeout_defer, which, if set TRUE, converts the failures into defers for
4514 both kinds of timeout. A transport filter timeout is now identified in
4517 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4518 systems where "make" and "gmake" are different, calling "gmake" at top
4519 level broke things. I've arranged for the value of $(MAKE) to be passed
4520 from the Makefile to this script so that it can call the same version of
4524 A note about Exim versions 4.44 and 4.50
4525 ----------------------------------------
4527 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4528 changes of various kinds. As a consequence, a big documentation update was
4529 needed. This delayed the release for rather longer than seemed good, especially
4530 in the light of a couple of (minor) security issues. Therefore, the changes
4531 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4532 release. So 4.44 and 4.50 are in effect two different branches that both start
4535 I have left the 4.50 change log unchanged; it contains all the changes since
4536 4.43. The change log for 4.44 is below; many of its items are identical to
4537 those for 4.50. This seems to be the most sensible way to preserve the
4538 historical information.
4544 1. Minor wording change to the doc/README.SIEVE file.
4546 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4547 computation of the current number of files was incorrect.
4549 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4550 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4553 4. Give more explanation in the error message when the command for a transport
4554 filter fails to execute.
4556 5. There are several places where Exim runs a non-Exim command in a
4557 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4558 was being done only for the command run by the queryprogram router. It is
4559 now done for all such subprocesses. The other cases are: ${run, transport
4560 filters, and the commands run by the lmtp and pipe transports.
4562 6. Added CONFIGURE_GROUP build-time option.
4564 7. Some older OS have a limit of 256 on the maximum number of file
4565 descriptors. Exim was using setrlimit() to set 1000 as a large value
4566 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4567 systems. I've change it so that if it can't get 1000, it tries for 256.
4569 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4570 was an oversight, and furthermore, ever since the addition of extra
4571 controls (e.g. 4.43/32), the checks on when to allow different forms of
4572 "control" were broken. There should now be diagnostics for all cases when a
4573 control that does not make sense is encountered.
4575 9. Added the /retain_sender option to "control=submission".
4577 10. $recipients is now available in the predata ACL (oversight).
4579 11. Tidy the search cache before the fork to do a delivery from a message
4580 received from the command line. Otherwise the child will trigger a lookup
4581 failure and thereby defer the delivery if it tries to use (for example) a
4582 cached ldap connection that the parent has called unbind on.
4584 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4585 of $address_data from the recipient verification was clobbered by the
4586 sender verification.
4588 13. The value of address_data from a sender verification is now available in
4589 $sender_address_data in subsequent conditions in the ACL statement.
4591 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4593 15. Added a new option "connect=<time>" to callout options, to set a different
4596 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4597 was its contents. (It was OK if the option was not defined at all.)
4599 17. A "Completed" log line is now written for messages that are removed from
4600 the spool by the -Mrm option.
4602 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4603 information about exactly what failed.
4605 19. Added -dd to debug only the daemon process.
4607 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4608 handles timeouts, both on the server side and network timeouts. Renamed the
4609 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4611 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4612 It is now set to "smtps".
4614 22. $host_address is now set to the target address during the checking of
4615 ignore_target_hosts.
4617 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4618 being passed; this would have caused $sender_host_name to have been used if
4619 matching the list had actually called for a host name (not very likely,
4620 since this list is usually IP addresses). A host name is now passed as
4623 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4624 code that specifies a non-restarting handler (typically sigaction() in
4625 modern systems) in an attempt to fix a rare and obscure crash bug.
4627 25. Narrowed the window for a race in the daemon that could cause it to ignore
4628 SIGCHLD signals. This is not a major problem, because they are used only to
4629 wake it up if nothing else does.
4631 26. A malformed maildirsize file could cause Exim to calculate negative values
4632 for the mailbox size or file count. Odd effects could occur as a result.
4633 The maildirsize information is now recalculated if the size or filecount
4636 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4637 support for a long time. Removed HAVE_SYS_VFS_H.
4639 28. Installed the latest version of exipick from John Jetmore.
4641 29. In an address list, if the pattern was not a regular expression, an empty
4642 subject address (from a bounce message) matched only if the pattern was an
4643 empty string. Non-empty patterns were not even tested. This was the wrong
4644 because it is perfectly reasonable to use an empty address as part of a
4645 database query. An empty address is now tested by patterns that are
4646 lookups. However, all the other forms of pattern expect the subject to
4647 contain a local part and a domain, and therefore, for them, an empty
4648 address still always fails if the pattern is not itself empty.
4650 30. Exim went into a mad DNS loop when attempting to do a callout where the
4651 host was specified on an smtp transport, and looking it up yielded more
4652 than one IP address.
4654 31. Re-factored the code for checking spool and log partition space into a
4655 function that finds that data and another that does the check. The former
4656 is then used to implement four new variables: $spool_space, $log_space,
4657 $spool_inodes, and $log_inodes.
4659 32. The RFC2047 encoding function was originally intended for short strings
4660 such as real names; it was not keeping to the 75-character limit for
4661 encoded words that the RFC imposes. It now respects the limit, and
4662 generates multiple encoded words if necessary. To be on the safe side, I
4663 have increased the buffer size for the ${rfc2047: expansion operator from
4666 33. It is now permitted to omit both strings after an "if" condition; if the
4667 condition is true, the result is "true". As before, when the second string
4668 is omitted, a false condition yields an empty string. This makes it less
4669 cumbersome to write custom ACL and router conditions.
4671 34. Failure to deliver a bounce message always caused it to be frozen, even if
4672 there was an errors_to setting on the router. The errors_to setting is now
4675 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4676 canonical form (fully expanded) before being placed in
4677 $sender_host_address.
4679 36. The table in the code that translates DNS record types into text (T_A to
4680 "A" for instance) was missing entries for NS and CNAME. It is just possible
4681 that this could have caused confusion if both these types were looked up
4682 for the same domain, because the text type is used as part of Exim's
4683 per-process caching. But the chance of anyone hitting this buglet seems
4686 37. The dnsdb lookup has been extended in a number of ways.
4688 (1) There is a new type, "zns", which walks up the domain tree until it
4689 finds some nameserver records. It should be used with care.
4691 (2) There is a new type, "mxh", which is like "mx" except that it returns
4692 just the host names, not the priorities.
4694 (3) It is now possible to give a list of domains (or IP addresses) to be
4695 looked up. The behaviour when one of the lookups defers can be
4696 controlled by a keyword.
4698 (4) It is now possible to specify the separator character for use when
4699 multiple records are returned.
4701 38. The dnslists ACL condition has been extended: it is now possible to supply
4702 a list of IP addresses and/or domains to be looked up in a particular DNS
4705 39. Added log_selector=+queue_time_overall.
4707 40. When running the queue in the test harness, wait just a tad after forking a
4708 delivery process, to get repeatability of debugging output.
4710 41. Include certificate and key file names in error message when GnuTLS fails
4711 to set them up, because the GnuTLS error message doesn't include the name
4712 of the failing file when there is a problem reading it.
4714 42. Allow both -bf and -bF in the same test run.
4716 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4718 44. The "Exiscan patch" is now merged into the mainline Exim source.
4720 45. Sometimes the final signoff response after QUIT could fail to get
4721 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4722 < 0 before doing a fflush(). This bug looks as though it goes back to the
4723 introduction of TLS in release 3.20, but "sometimes" must have been rare
4724 because the tests only now provoked it.
4726 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4727 (this can affect the format of dates).
4729 47. exim_tidydb, when checking for the continued existence of a message for
4730 which it has found a message-specific retry record, was not finding
4731 messages that were in split spool directories. Consequently, it was
4732 deleting retry records that should have stayed in existence.
4734 48. Steve fixed some bugs in eximstats.
4736 49. The SPA authentication driver was not abandoning authentication and moving
4737 on to the next authenticator when an expansion was forced to fail,
4738 contradicting the general specification for all authenticators. Instead it
4739 was generating a temporary error. It now behaves as specified.
4741 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4742 (the order specifies the preference for clients). The order is now AES256,
4743 AES128, 3DES, ARCFOUR128.
4745 51. Small patch to Sieve code - explicitly set From: when generating an
4748 52. Exim crashed if a remote delivery caused a very long error message to be
4749 recorded - for instance if somebody sent an entire SpamAssassin report back
4750 as a large number of 550 error lines. This bug was coincidentally fixed by
4751 increasing the size of one of Exim's internal buffers (big_buffer) that
4752 happened as part of the Exiscan merge. However, to be on the safe side, I
4753 have made the code more robust (and fixed the comments that describe what
4756 53. Now that there can be additional text after "Completed" in log lines (if
4757 the queue_time_overall log selector is set), a one-byte patch to exigrep
4758 was needed to allow it to recognize "Completed" as not the last thing in
4761 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4762 patch that reportedly fixes this has been added. I am not expert enough to
4763 create a test for it. This is what the patch creator wrote:
4765 "I found a little strange behaviour of ldap code when working with
4766 Windows 2003 AD Domain, where users was placed in more than one
4767 Organization Units. When I tried to give exim partial DN, the exit code
4768 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4769 But simultaneously result of request was absolutely normal ldap result,
4770 so I produce this patch..."
4772 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4773 REFERENCE, so I have modified the code to exclude the patch when that macro
4776 55. Some experimental protocols are using DNS PTR records for new purposes. The
4777 keys for these records are domain names, not reversed IP addresses. The
4778 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4779 leaves it alone. Component reversal etc. now happens only for IP addresses.
4782 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4784 57. Double the size of the debug message buffer (to 2048) so that more of very
4785 long debug lines gets shown.
4787 58. The exicyclog utility now does better if the number of log files to keep
4788 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4790 59. Two changes related to the smtp_active_hostname option:
4792 (1) $smtp_active_hostname is now available as a variable.
4793 (2) The default for smtp_banner uses $smtp_active_hostname instead
4794 of $primary_hostname.
4796 60. The host_aton() function is supposed to be passed a string that is known
4797 to be a valid IP address. However, in the case of IPv6 addresses, it was
4798 not checking this. This is a hostage to fortune. Exim now panics and dies
4799 if the condition is not met. A case was found where this could be provoked
4800 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4801 components; fortuitously, this particular loophole had already been fixed
4802 by change 4.50/55 above.
4804 If there are any other similar loopholes, the new check in host_aton()
4805 itself should stop them being exploited. The report I received stated that
4806 data on the command line could provoke the exploit when Exim was running as
4807 exim, but did not say which command line option was involved. All I could
4808 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4809 running as the user.
4812 61. There was a buffer overflow vulnerability in the SPA authentication code
4813 (which came originally from the Samba project). I have added a test to the
4814 spa_base64_to_bits() function which I hope fixes it.
4817 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4818 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4820 63. The daemon start-up calls getloadavg() while still root for those OS that
4821 need the first call to be done as root, but it missed one case: when
4822 deliver_queue_load_max is set with deliver_drop_privilege. This is
4823 necessary for the benefit of the queue runner, because there is no re-exec
4824 when deliver_drop_privilege is set.
4826 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4827 This has been fixed.
4829 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4830 was in use, was not putting the data itself into the right store pool;
4831 consequently, it could be overwritten for a subsequent message in the same
4832 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4835 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4837 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4838 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4839 the function were treating the return as a boolean value, which happened to
4840 work because 0=false and not-0=true, but is not correct code.
4842 68. The host_aton() function was not handling scoped IPv6 addresses (those
4843 with, for example, "%eth0" on the end) correctly.
4845 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4846 negated items (that is, ~something) in unsigned ints. Some compilers
4847 apparently mutter when there is no cast.
4849 70. If an address verification called from an ACL failed, and did not produce a
4850 user-specific message (i.e. there was only a "system" message), nothing was
4851 put in $acl_verify_message. In this situation, it now puts the system
4854 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4855 session; change 4.31/43 added the unwanted input to the log line - except
4856 that it did not do this in the start of session case. It now does.
4858 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4859 This is wrong and can cause the other end to generate a synchronization
4860 error if it is another Exim or anything else that does the synchronization
4861 check. A QUIT command is no longer sent after a timeout.
4863 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4864 during host lookups.
4866 74. The defer_ok option of callout verification was not working if it was used
4867 when verifying addresses in header lines, that is, for this case:
4869 verify = header_sender/callout=defer_ok
4871 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4872 those file descriptors could be used for SMTP connections. If anything
4873 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4874 could be sent to the SMTP client, causing chaos. The daemon now opens
4875 stdin, stdout, and stderr to /dev/null when it puts itself into the
4878 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4879 log by default. The user can override this with suitable Perl magic.
4881 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4882 the log message when discard triggers, was not working for the DATA ACL or
4883 for the non-SMTP ACL.
4885 78. Error message wording change in sieve.c.
4887 79. If smtp_accept_max_per_host was set, the number of connections could be
4888 restricted to fewer than expected, because the daemon was trying to set up
4889 a new connection before checking whether the processes handling previous
4890 connections had finished. The check for completed processes is now done
4891 earlier. On busy systems, this bug wouldn't be noticed because something
4892 else would have woken the daemon, and it would have reaped the completed
4895 80. If a message was submitted locally by a user whose login name contained one
4896 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4897 It caused a spool format error. I have fixed the spool reading code. A
4898 related problem was that the "from" clause in the Received: line became
4899 illegal because of the space(s). It is now covered by ${quote_local_part.
4901 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4904 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4906 83. Patch from Sieve maintainer to fix unterminated string problem in
4907 "vacation" handling.
4909 84. Some minor changes to the Linux configuration files to help with other
4910 OS variants using glibc.
4912 85. One more patch for Sieve to update vacation handling to latest spec.
4915 ----------------------------------------------------
4916 See the note above about the 4.44 and 4.50 releases.
4917 ----------------------------------------------------
4923 1. Change 4.43/35 introduced a bug that caused file counts to be
4924 incorrectly computed when quota_filecount was set in an appendfile
4927 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4928 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4931 3. Give more explanation in the error message when the command for a transport
4932 filter fails to execute.
4934 4. There are several places where Exim runs a non-Exim command in a
4935 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4936 was being done only for the command run by the queryprogram router. It is
4937 now done for all such subprocesses. The other cases are: ${run, transport
4938 filters, and the commands run by the lmtp and pipe transports.
4940 5. Some older OS have a limit of 256 on the maximum number of file
4941 descriptors. Exim was using setrlimit() to set 1000 as a large value
4942 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4943 systems. I've change it so that if it can't get 1000, it tries for 256.
4945 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4946 was an oversight, and furthermore, ever since the addition of extra
4947 controls (e.g. 4.43/32), the checks on when to allow different forms of
4948 "control" were broken. There should now be diagnostics for all cases when a
4949 control that does not make sense is encountered.
4951 7. $recipients is now available in the predata ACL (oversight).
4953 8. Tidy the search cache before the fork to do a delivery from a message
4954 received from the command line. Otherwise the child will trigger a lookup
4955 failure and thereby defer the delivery if it tries to use (for example) a
4956 cached ldap connection that the parent has called unbind on.
4958 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4959 of $address_data from the recipient verification was clobbered by the
4960 sender verification.
4962 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4963 was its contents. (It was OK if the option was not defined at all.)
4965 11. A "Completed" log line is now written for messages that are removed from
4966 the spool by the -Mrm option.
4968 12. $host_address is now set to the target address during the checking of
4969 ignore_target_hosts.
4971 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4972 being passed; this would have caused $sender_host_name to have been used if
4973 matching the list had actually called for a host name (not very likely,
4974 since this list is usually IP addresses). A host name is now passed as
4977 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4978 code that specifies a non-restarting handler (typically sigaction() in
4979 modern systems) in an attempt to fix a rare and obscure crash bug.
4981 15. Narrowed the window for a race in the daemon that could cause it to ignore
4982 SIGCHLD signals. This is not a major problem, because they are used only to
4983 wake it up if nothing else does.
4985 16. A malformed maildirsize file could cause Exim to calculate negative values
4986 for the mailbox size or file count. Odd effects could occur as a result.
4987 The maildirsize information is now recalculated if the size or filecount
4990 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4991 support for a long time. Removed HAVE_SYS_VFS_H.
4993 18. Updated exipick to current release from John Jetmore.
4995 19. Allow an empty sender to be matched against a lookup in an address list.
4996 Previously the only cases considered were a regular expression, or an
4999 20. Exim went into a mad DNS lookup loop when doing a callout where the
5000 host was specified on the transport, if the DNS lookup yielded more than
5003 21. The RFC2047 encoding function was originally intended for short strings
5004 such as real names; it was not keeping to the 75-character limit for
5005 encoded words that the RFC imposes. It now respects the limit, and
5006 generates multiple encoded words if necessary. To be on the safe side, I
5007 have increased the buffer size for the ${rfc2047: expansion operator from
5010 22. Failure to deliver a bounce message always caused it to be frozen, even if
5011 there was an errors_to setting on the router. The errors_to setting is now
5014 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5015 canonical form (fully expanded) before being placed in
5016 $sender_host_address.
5018 24. Updated eximstats to version 1.33
5020 25. Include certificate and key file names in error message when GnuTLS fails
5021 to set them up, because the GnuTLS error message doesn't include the name
5022 of the failing file when there is a problem reading it.
5024 26. Expand error message when OpenSSL has problems setting up cert/key files.
5027 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5028 (this can affect the format of dates).
5030 28. exim_tidydb, when checking for the continued existence of a message for
5031 which it has found a message-specific retry record, was not finding
5032 messages that were in split spool directories. Consequently, it was
5033 deleting retry records that should have stayed in existence.
5035 29. eximstats updated to version 1.35
5036 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5037 1.35 - bugfix such that pie charts by volume are generated correctly
5039 30. The SPA authentication driver was not abandoning authentication and moving
5040 on to the next authenticator when an expansion was forced to fail,
5041 contradicting the general specification for all authenticators. Instead it
5042 was generating a temporary error. It now behaves as specified.
5044 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5045 (the order specifies the preference for clients). The order is now AES256,
5046 AES128, 3DES, ARCFOUR128.
5048 31. Small patch to Sieve code - explicitly set From: when generating an
5051 32. Exim crashed if a remote delivery caused a very long error message to be
5052 recorded - for instance if somebody sent an entire SpamAssassin report back
5053 as a large number of 550 error lines. This bug was coincidentally fixed by
5054 increasing the size of one of Exim's internal buffers (big_buffer) that
5055 happened as part of the Exiscan merge. However, to be on the safe side, I
5056 have made the code more robust (and fixed the comments that describe what
5059 33. Some experimental protocols are using DNS PTR records for new purposes. The
5060 keys for these records are domain names, not reversed IP addresses. The
5061 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5062 leaves it alone. Component reversal etc. now happens only for IP addresses.
5065 34. The host_aton() function is supposed to be passed a string that is known
5066 to be a valid IP address. However, in the case of IPv6 addresses, it was
5067 not checking this. This is a hostage to fortune. Exim now panics and dies
5068 if the condition is not met. A case was found where this could be provoked
5069 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5070 components; fortuitously, this particular loophole had already been fixed
5071 by change 4.50/55 or 4.44/33 above.
5073 If there are any other similar loopholes, the new check in host_aton()
5074 itself should stop them being exploited. The report I received stated that
5075 data on the command line could provoke the exploit when Exim was running as
5076 exim, but did not say which command line option was involved. All I could
5077 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5078 running as the user.
5081 35. There was a buffer overflow vulnerability in the SPA authentication code
5082 (which came originally from the Samba project). I have added a test to the
5083 spa_base64_to_bits() function which I hope fixes it.
5086 36. The daemon start-up calls getloadavg() while still root for those OS that
5087 need the first call to be done as root, but it missed one case: when
5088 deliver_queue_load_max is set with deliver_drop_privilege. This is
5089 necessary for the benefit of the queue runner, because there is no re-exec
5090 when deliver_drop_privilege is set.
5092 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5093 was in use, was not putting the data itself into the right store pool;
5094 consequently, it could be overwritten for a subsequent message in the same
5095 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5098 38. Sometimes the final signoff response after QUIT could fail to get
5099 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5100 < 0 before doing a fflush(). This bug looks as though it goes back to the
5101 introduction of TLS in release 3.20, but "sometimes" must have been rare
5102 because the tests only now provoked it.
5108 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5109 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5110 Now it returns an integer. A number of calls were still expecting a T/F
5111 return. Fortuitously, in all cases, the tests worked in OK situations,
5112 which is the norm. However, things would have gone wrong on any write
5113 failures on the smtp file descriptor. This function is used when sending
5114 messages over SMTP and also when doing verify callouts.
5116 2. When Exim is called to do synchronous delivery of a locally submitted
5117 message (the -odf or -odi options), it no longer closes stderr before doing
5120 3. Implemented the mua_wrapper option.
5122 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5124 5. Implemented the functions header_remove(), header_testname(),
5125 header_add_at_position(), and receive_remove_recipient(), and exported them
5128 6. If an ACL "warn" statement specified the addition of headers, Exim already
5129 inserted X-ACL-Warn: at the start if there was no header name. However, it
5130 was not making this test for the second and subsequent header lines if
5131 there were newlines in the string. This meant that an invalid header could
5132 be inserted if Exim was badly configured.
5134 7. Allow an ACL "warn" statement to add header lines at the start or after all
5135 the Received: headers, as well as at the end.
5137 8. Added the rcpt_4xx retry error code.
5139 9. Added postmaster_mailfrom=xxx to callout verification option.
5141 10. Added mailfrom=xxxx to the callout verification option, for verify=
5144 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5145 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5147 12. Inserted some casts to stop certain compilers warning when using pointer
5148 differences as field lengths or precisions in printf-type calls (mostly
5149 affecting debugging statements).
5151 13. Added optional readline() support for -be (dynamically loaded).
5153 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5154 same clock tick as a message's arrival, so that its received time was the
5155 same as the "first fail" time on the retry record, and that message
5156 remained on the queue past the ultimate address timeout, every queue runner
5157 would try a delivery (because it was past the ultimate address timeout) but
5158 after another failure, the ultimate address timeout, which should have then
5159 bounced the address, did not kick in. This was a "< instead of <=" error;
5160 in most cases the first failure would have been in the next clock tick
5161 after the received time, and all would be well.
5163 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5164 being recognized when the domain list was tested by the match_domain
5165 condition in an expansion string.
5167 16. Added the ${str2b64: operator.
5169 17. Exim was always calling setrlimit() to set a large limit for the number of
5170 processes, without checking whether the existing limit was already
5171 adequate. (It did check for the limit on file descriptors.) Furthermore,
5172 errors from getrlimit() and setrlimit() were being ignored. Now they are
5173 logged to the main and panic logs, but Exim does carry on, to try to do its
5174 job under whatever limits there are.
5176 18. Imported PCRE 5.0.
5178 19. Trivial typo in log message " temporarily refused connection" (the leading
5181 20. If the log selector return_path_on_delivery was set and an address was
5182 redirected to /dev/null, the delivery process crashed because it assumed
5183 that a return path would always be set for a "successful" delivery. In this
5184 case, the whole delivery is bypassed as an optimization, and therefore no
5187 21. Internal re-arrangement: the function for sending a challenge and reading
5188 a response while authentication was assuming a zero-terminated challenge
5189 string. It's now changed to take a pointer and a length, to allow for
5190 binary data in such strings.
5192 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5194 23. Exim was not respecting finduser_retries when seeking the login of the
5195 uid under which it was called; it was always trying 10 times. (The default
5196 setting of finduser_retries is zero.) Also, it was sleeping after the final
5197 failure, which is pointless.
5199 24. Implemented tls_on_connect_ports.
5201 25. Implemented acl_smtp_predata.
5203 26. If the domain in control=submission is set empty, Exim assumes that the
5204 authenticated id is a complete email address when it generates From: or
5205 Sender: header lines.
5207 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5208 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5209 chown and chgrp in /bin and hostname in /usr/bin.
5211 28. Exim was keeping the "process log" file open after each use, just as it
5212 does for the main log. This opens the possibility of it remaining open for
5213 long periods when the USR1 signal hits a daemon. Occasional processlog
5214 errors were reported, that could have been caused by this. Anyway, it seems
5215 much more sensible not to leave this file open at all, so that is what now
5218 29. The long-running daemon process does not normally write to the log once it
5219 has entered its main loop, and it closes the log before doing so. This is
5220 so that log files can straightforwardly be renamed and moved. However,
5221 there are a couple of unusual error situations where the daemon does write
5222 log entries, and I had neglected to close the log afterwards.
5224 30. The text of an SMTP error response that was received during a remote
5225 delivery was being truncated at 512 bytes. This is too short for some of
5226 the long messages that one sometimes sees. I've increased the limit to
5229 31. It is now possible to make retry rules that apply only when a message has a
5230 specific sender, in particular, an empty sender.
5232 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5233 it possible to be selective about when SMTP synchronization is enforced.
5235 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5237 32. Implemented hosts_connection_nolog.
5239 33. Added an ACL for QUIT.
5241 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5244 35. Added mailbox_size and mailbox_filecount to appendfile.
5246 36. Added control = no_multiline_responses to ACLs.
5248 37. There was a bug in the logic of the code that waits for the clock to tick
5249 in the case where the clock went backwards by a substantial amount such
5250 that the microsecond fraction of "now" was more than the microsecond
5251 fraction of "then" (but the whole seconds number was less).
5253 38. Added support for the libradius Radius client library this is found on
5254 FreeBSD (previously only the radiusclient library was supported).
5260 1. When certain lookups returned multiple values in the form name=value, the
5261 quoting of the values was not always being done properly. Specifically:
5262 (a) If the value started with a double quote, but contained no whitespace,
5264 (b) If the value contained whitespace other than a space character (i.e.
5265 tabs or newlines or carriage returns) it was not quoted.
5266 This fix has been applied to the mysql and pgsql lookups by writing a
5267 separate quoting function and calling it from the lookup code. The fix
5268 should probably also be applied to nisplus, ibase and oracle lookups, but
5269 since I cannot test any of those, I have not disturbed their existing code.
5271 2. A hit in the callout cache for a specific address caused a log line with no
5272 reason for rejecting RCPT. Now it says "Previous (cached) callout
5273 verification failure".
5275 3. There was an off-by-one bug in the queryprogram router. An over-long
5276 return line was truncated at 256 instead of 255 characters, thereby
5277 overflowing its buffer with the terminating zero. As well as fixing this, I
5278 have increased the buffer size to 1024 (and made a note to document this).
5280 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5281 when Exim is waiting for an SMTP response from a remote server, Exim
5282 restarts its select() call on the socket, thereby resetting its timeout.
5283 This is not a problem when such interrupts are rare. Somebody set up a cron
5284 job to run exiwhat every 2 minutes, which is less than the normal select()
5285 timeout (5 or 10 minutes). This meant that the select() timeout never
5286 kicked in because it was always reset. I have fixed this by comparing the
5287 time when an interrupt arrives with the time at the start of the first call
5288 to select(). If more time than the timeout has elapsed, the interrupt is
5289 treated as a timeout.
5291 5. Some internal re-factoring in preparation for the addition of Sieve
5292 extensions (by MH). In particular, the "personal" test is moved to a
5293 separate function, and given an option for scanning Cc: and Bcc: (which is
5294 not set for Exim filters).
5296 6. When Exim created an email address using the login of the caller as the
5297 local part (e.g. when creating a From: or Sender: header line), it was not
5298 quoting the local part when it contained special characters such as @.
5300 7. Installed new OpenBSD configuration files.
5302 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5303 try to make them clearer.
5305 9. Callout options, other than the timeout value, were being ignored when
5306 verifying sender addresses in header lines. For example, when using
5308 verify = header_sender/callout=no_cache
5310 the cache was (incorrectly) being used.
5312 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5313 only the Cygwin environment.
5315 11. When return_path_on_delivery was set as a log selector, if different remote
5316 addresses in the same message used different return paths and parallel
5317 remote delivery occurred, the wrong values would sometimes be logged.
5318 (Whenever a remote delivery process finished, the return path value from
5319 the most recently started remote delivery process was logged.)
5321 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5322 lines when AUTH and/or TLS are in use. This is the "received protocol"
5323 field. Exim used to use "asmtp" for authenticated SMTP, without any
5324 indication (in the protocol name) for TLS use. Now it follows the RFC and
5325 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5326 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5327 names appear in log lines as well as in Received: header lines.
5329 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5330 extensions, and comparison tests, and to fix some bugs.
5332 14. Changes to the "personal" filter test:
5334 (1) The test was buggy in that it was just doing the equivalent of
5335 "contains" tests on header lines. For example, if a user's address was
5336 anne@some.where, the "personal" test would incorrectly be true for
5338 To: susanne@some.where
5340 This test is now done by extracting each address from the header in turn,
5341 and checking the entire address. Other tests that are part of "personal"
5342 are now done using regular expressions (for example, to check local parts
5343 of addresses in From: header lines).
5345 (2) The list of non-personal local parts in From: addresses has been
5346 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5347 taken from the Sieve specification recommendations.
5349 (3) If the message contains any header line starting with "List-" it is
5350 treated as non-personal.
5352 (4) The test for "circular" in the Subject: header line has been removed
5353 because it now seems ill-conceived.
5355 15. Minor typos in src/EDITME comments corrected.
5357 16. Installed latest exipick from John Jetmore.
5359 17. If headers_add on a router specified a text string that was too long for
5360 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5361 of string_sprintf() is now avoided.
5363 18. $message_body_size was not set (it was always zero) when running the DATA
5364 ACL and the local_scan() function.
5366 19. For the "mail" command in an Exim filter, no default was being set for
5367 the once_repeat time, causing a random time value to be used if "once" was
5368 specified. (If the value happened to be <= 0, no repeat happened.) The
5369 default is now 0s, meaning "never repeat". The "vacation" command was OK
5370 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5371 (I found it when inspecting the code).
5373 20. There is now an overall timeout for performing a callout verification. It
5374 defaults to 4 times the callout timeout, which applies to individual SMTP
5375 commands during the callout. The overall timeout applies when there is more
5376 than one host that can be tried. The timeout is checked before trying the
5377 next host. This prevents very long delays if there are a large number of
5378 hosts and all are timing out (e.g. when the network connections are timing
5379 out). The value of the overall timeout can be changed by specifying an
5380 additional sub-option for "callout", called "maxwait". For example:
5382 verify = sender/callout=5s,maxwait=20s
5384 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5385 to the end before writing, but this should make it even safer).
5387 22. Exim was forgetting that it had advertised PIPELINING for the second and
5388 subsequent messages on an SMTP connection. It was also not resetting its
5389 memory on STARTTLS and an internal HELO.
5391 23. When Exim logs an SMTP synchronization error within a session, it now
5392 records whether PIPELINING has been advertised or not.
5394 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5395 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5396 rather than long int.
5398 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5400 26. Added the never_mail option to autoreply.
5406 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5407 crash if the getsockname() call failed; this can happen if a connection is
5408 closed very soon after it is established. The problem was simply in the
5409 order in which certain operations were done, causing Exim to try to write
5410 to the SMTP stream before it had set up the file descriptor. The bug has
5411 been fixed by making things happen in the correct order.
5417 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5418 before the connection was closed, thus losing the rejection response.
5420 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5421 some early Solaris releases, but causes trouble in current releases where
5422 socklen_t is defined.
5424 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5427 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5430 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5431 that releasing the top of it at the end releases what was used for sub-
5432 expansions (unless the block got too big). However, discard this block if
5433 the first thing is a variable or header, so that we can use its block when
5434 it is dynamic (useful for very large $message_headers, for example).
5436 6. Lookups now cache *every* query, not just the most recent. A new, separate
5437 store pool is used for this. It can be recovered when all lookup caches are
5438 flushed. Lookups now release memory at the end of their result strings.
5439 This has involved some general refactoring of the lookup sources.
5441 7. Some code has been added to the store_xxx() functions to reduce the amount
5442 of flapping under certain conditions.
5444 8. log_incoming_interface used to affect only the <= reception log lines. Now
5445 it causes the local interface and port to be added to several more SMTP log
5446 lines, for example "SMTP connection from", and rejection lines.
5448 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5450 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5452 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5453 could be overwritten at the end of the current message (or the start of a
5454 new message if it was set in a HELO ACL). The value is now preserved for
5455 the duration of the SMTP connection.
5457 12. If a transport had a headers_rewrite setting, and a matching header line
5458 contained an unqualified address, that address was qualified, even if it
5459 did not match any rewriting rules. The underlying bug was that the values
5460 of the flags that permit the existence of unqualified sender and recipient
5461 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5462 non-local messages, and by -bnq for local messages) were not being
5463 preserved with the message after it was received.
5465 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5466 "next input=" as part of the text comprising the host identity instead of
5467 the correct text. The code was using the same buffer for two different
5468 strings. However, depending on which order the printing function evaluated
5469 its arguments, the bug did not always show up. Under Linux, for example, my
5470 test suite worked just fine.
5472 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5473 doesn't work with some older versions of Perl. It has been changed to "my",
5474 which in any case is probably the better facility to use.
5476 15. A really picky compiler found some instances of statements for creating
5477 error messages that either had too many or two few arguments for the format
5480 16. The size of the buffer for calls to the DNS resolver has been increased
5481 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5482 for addresses that have a lot of PTR records. This alleviates a problem; it
5483 does not fully solve it.
5485 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5486 buffer now truncates the list and logs the incident, which is the same
5487 action as happens when Exim is looking up a host name and its aliases.
5488 Previously in this situation something unpredictable would happen;
5489 sometimes it was "internal error: store_reset failed".
5491 18. If a server dropped the connection unexpectedly when an Exim client was
5492 using GnuTLS and trying to read a response, the client delivery process
5493 crashed while trying to generate an error log message.
5495 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5496 string, for example:
5498 warn message = H1: something\nH2: something
5500 the text was added as a single header line from Exim's point of view
5501 though it ended up OK in the delivered message. However, searching for the
5502 second and subsequent header lines using $h_h2: did not work. This has been
5503 fixed. Similarly, if a system filter added multiple headers in this way,
5504 the routers could not see them.
5506 20. Expanded the error message when iplsearch is called with an invalid key to
5507 suggest using net-iplsearch in a host list.
5509 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5510 ACLs are no longer actually imposed (and a message to that effect is
5513 22. If a "gecos" field in a passwd entry contained escaped characters, in
5514 particular, if it contained a \" sequence, Exim got it wrong when building
5515 a From: or a Sender: header from that name. A second bug also caused
5516 incorrect handling when an unquoted " was present following a character
5517 that needed quoting.
5519 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5520 was not being matched caselessly.
5522 24. Arranged for all hyphens in the exim.8 source to be escaped with
5525 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5526 properly. Recipient callout cache records were still being keyed to include
5527 the sender, even when use_sender was set false. This led to far more
5528 callouts that were necessary. The sender is no longer included in the key
5529 when use_sender is false.
5531 26. Added "control = submission" modifier to ACLs.
5533 27. Added the ${base62d: operator to decode base 62 numbers.
5535 28. dnsdb lookups can now access SRV records.
5537 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5538 the configuration file.
5540 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5541 (-v) mode. This makes the output for a verbose queue run more intelligible.
5543 31. Added a use_postmaster feature to recipient callouts.
5545 32. Added the $body_zerocount variable, containing the number of binary zero
5546 bytes in the message body.
5548 33. The time of last modification of the "new" subdirectory is now used as the
5549 "mailbox time last read" when there is a quota error for a maildir
5552 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5554 35. Added +ignore_unknown as a special item in host lists.
5556 36. Code for decoding IPv6 addresses in host lists is now included, even if
5557 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5558 address was recognized as an IP address, but was then not correctly decoded
5559 into binary, causing unexpected and incorrect effects when compared with
5566 1. Very minor rewording of debugging text in manualroute to say "list of
5567 hosts" instead of "hostlist".
5569 2. If verify=header_syntax was set, and a header line with an unqualified
5570 address (no domain) and a large number of spaces between the end of the
5571 name and the colon was received, the reception process suffered a buffer
5572 overflow, and (when I tested it) crashed. This was caused by some obsolete
5573 code that should have been removed. The fix is to remove it!
5575 3. When running in the test harness, delay a bit after writing a bounce
5576 message to get a bit more predictability in the log output.
5578 4. Added a call to search_tidyup() just before forking a reception process. In
5579 theory, someone could use a lookup in the expansion of smtp_accept_max_
5580 per_host which, without the tidyup, could leave open a database connection.
5582 5. Added the variables $recipient_data and $sender_data which get set from a
5583 lookup success in an ACL "recipients" or "senders" condition, or a router
5584 "senders" option, similar to $domain_data and $local_part_data.
5586 6. Moved the writing of debug_print from before to after the "senders" test
5589 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5590 problems for message scanning, either using a data ACL, or using
5591 local_scan() because the Received: header was not generated till after they
5592 were called (in order to set the time as the time of reception completion).
5593 I have revised the way this works. The header is now generated after the
5594 body is received, but before the ACL or local_scan() are called. After they
5595 are run, the timestamp in the header is updated.
5601 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5602 before starting a queue runner without re-exec. This happened only when
5603 deliver_drop_privilege was set or when the Exim user was set to root. The
5604 effect of the bug was that timeouts during subsequent deliveries caused
5605 crashes instead of being properly handled. The handler is now left at its
5606 default (and expected) setting.
5608 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5609 message, again when deliver_drop_privilege is set or Exim is run as root.
5610 The bug described in (1) was not present in this case, but the tidying up
5611 of the other signals was missing. I have made the two cases consistent.
5613 3. The ignore_target_hosts setting on a manualroute router was being ignored
5614 for hosts that were looked up using the /MX notation.
5616 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5619 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5620 operated on the sender address. After changing the $sender_address to <>
5621 for the sender address verify, Exim was re-instated it as the original
5622 (before rewriting) address, but remembering that it had rewritten it, so it
5623 wasn't rewriting it again. This bug also had the effect of breaking the
5624 sender address verification caching when the sender address was rewritten.
5626 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5627 This has been changed so that if the ip literal address matches
5628 ignore_target_hosts, the router declines.
5630 7. Added expansion conditions match_domain, match_address, and match_local_
5631 part (NOT match_host).
5633 8. The placeholder for the Received: header didn't have a length field set.
5635 9. Added code to Exim itself and to exim_lock to test for a specific race
5636 condition that could lead to file corruption when using MBX delivery. The
5637 issue is with the lockfile that is created in /tmp. If this file is removed
5638 after a process has opened it but before that process has acquired a lock,
5639 there is the potential for a second process to recreate the file and also
5640 acquire a lock. This could lead to two Exim processes writing to the file
5641 at the same time. The added code performs the same test as UW imapd; it
5642 checks after acquiring the lock that its file descriptor still refers to
5643 the same named file.
5645 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5646 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5649 11. Added the smtp_active_hostname option. If used, this will typically be made
5650 to depend on the incoming interface address. Because $interface_address is
5651 not set up until the daemon has forked a reception process, error responses
5652 that can happen earlier (such as "too many connections") no longer contain
5655 12. If an expansion in a condition on a "warn" statement fails because a lookup
5656 defers, the "warn" statement is abandoned, and the next ACL statement is
5657 processed. Previously this caused the whole ACL to be aborted.
5659 13. Added the iplsearch lookup type.
5661 14. Added ident_timeout as a log selector.
5663 15. Added tls_certificate_verified as a log selector.
5665 16. Added a global option tls_require_ciphers (compare the smtp transport
5666 option of the same name). This controls incoming TLS connections.
5668 17. I finally figured out how to make tls_require_ciphers do a similar thing
5669 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5670 before starting the TLS session.
5672 18. Tabs are now shown as \t in -bP output.
5674 19. If the log selector return_path_on_delivery was set, Exim crashed when
5675 bouncing a message because it had too many Received: header lines.
5677 20. If two routers both had headers_remove settings, and the first one included
5678 a superfluous trailing colon, the final name in the first list and the
5679 first name in the second list were incorrectly joined into one item (with a
5680 colon in the middle).
5686 1. Added -C and -D options to the exinext utility, mainly to make it easier
5687 to include in the automated testing, but these could be helpful when
5688 multiple configurations are in use.
5690 2. The exinext utility was not formatting the output nicely when there was
5691 an alternate port involved in the retry record key, nor when there was a
5692 message id as well (for retries that were specific to a specific message
5693 and a specific host). It was also confused by IPv6 addresses, because of
5694 the additional colons they contain. I have fixed the IPv4 problem, and
5695 patched it up to do a reasonable job for IPv6.
5697 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5698 delivery, the log line now contains "pipelined" if PIPELINING was used.
5700 4. An SMTP transport process used to panic and die if the bind() call to set
5701 an explicit outgoing interface failed. This has been changed; it is now
5702 treated in the same way as a connect() failure.
5704 5. A reference to $sender_host_name in the part of a conditional expansion
5705 that was being skipped was still causing a DNS lookup. This no longer
5708 6. The def: expansion condition was not recognizing references to header lines
5709 that used bh_ and bheader_.
5711 7. Added the _cache feature to named lists.
5713 8. The code for checking quota_filecount in the appendfile transport was
5714 allowing one more file than it should have been.
5716 9. For compatibility with Sendmail, the command line option
5724 and sets the incoming protocol and host name (for trusted callers). The
5725 host name and its colon can be omitted when only the protocol is to be set.
5726 Note the Exim already has two private options, -pd and -ps, that refer to
5727 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5728 "s", but I don't think that's a major issue.
5730 10. A number of refactoring changes to the code, none of which should affect
5733 (a) The number of logging options was getting close to filling up the
5734 32-bit word that was used as a bit map. I have split them into two classes:
5735 those that are passed in the argument to log_write(), and those that are
5736 only ever tested independently outside of that function. These are now in
5737 separate 32-bit words, so there is plenty of room for expansion again.
5738 There is no change in the user interface or the logging behaviour.
5740 (b) When building, for example, log lines, the code previously used a
5741 macro that called string_cat() twice, in order to add two strings. This is
5742 not really sufficiently general. Furthermore, there was one instance where
5743 it was actually wrong because one of the argument was used twice, and in
5744 one call a function was used. (As it happened, calling the function twice
5745 did not affect the overall behaviour.) The macro has been replaced by a
5746 function that can join an arbitrary number of extra strings onto a growing
5749 (c) The code for expansion conditions now uses a table and a binary chop
5750 instead of a serial search (which was left over from when there were very
5751 few conditions). Also, it now recognizes conditions like "pam" even when
5752 the relevant support is not compiled in: a suitably worded error message is
5753 given if an attempt is made to use such a condition.
5755 11. Added ${time_interval:xxxxx}.
5757 12. A bug was causing one of the ddress fields not to be passed back correctly
5758 from remote delivery subprocesses. The field in question was not being
5759 subsequently used, so this caused to problems in practice.
5761 13. Added new log selectors queue_time and deliver_time.
5763 14. Might have fixed a bug in maildirsizefile handling that threw up
5764 "unexpected character" debug warnings, and recalculated the data
5765 unnecessarily. In any case, I expanded the warning message to give more
5768 15. Added the message "Restricted characters in address" to the statements in
5769 the default ACL that block characters like @ and % in local parts.
5771 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5772 Three changes have been made:
5774 (a) There was a serious bug; a negative response to MAIL caused the whole
5775 recipient domain to be cached as invalid, thereby blocking all messages
5776 to all local parts at the same domain, from all senders. This bug has
5777 been fixed. The domain is no longer cached after a negative response to
5778 MAIL if the sender used is not empty.
5780 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5783 (c) A new callout option, "use_sender" has been added for people who want
5784 the modified behaviour.
5790 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5793 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5794 indeed breaks things for older releases.
5796 3. Added additional logging to the case where there is a problem reading data
5797 from a filter that is running in a subprocess using a pipe, in order to
5798 try to track down a specific problem.
5800 4. Testing facility fudge: when running in the test harness and attempting
5801 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5802 getting "No route to host". Convert this to a timeout.
5804 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5807 6. Some OS don't have socklen_t but use size_t instead. This affects the
5808 fifth argument of getsockopt() amongst other things. This is now
5809 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5810 can be set for individual OS. I have set it for SunOS5, OSF1, and
5811 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5812 some earlier ones do not.
5814 7. Change 4.30/15 was not doing the test caselessly.
5816 8. The standard form for an IPv6 address literal was being rejected by address
5817 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5818 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5819 this, as well as the form without the "IPv6" on the front (but only when
5820 address literals are enabled, of course).
5822 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5824 10. Exim crashed if a message with an empty sender address specified by -f
5825 encountered a router with an errors_to setting. This could be provoked only
5826 by a command such as
5830 where an empty string was supplied; "<>" did not hit this bug.
5832 11. Installed PCRE release 4.5.
5834 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5835 remained set. It is now erased.
5837 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5838 times from message ids (which are base 36 rather than the normal 62).
5840 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5841 were being counted as actual protocol errors, and logged if the log
5842 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5843 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5844 and DATA following a set of rejected RCPTs do not count as protocol errors.
5845 In other words, Exim assumes they were pipelined, though this may not
5846 actually be the case. Of course, in all cases the client gets an
5847 appropriate error code.
5849 15. If a lookup fails in an ACL condition, a message about the failure may
5850 be available; it is used if testing the ACL cannot continue, because most
5851 such messages specify what the cause of the deferral is. However, some
5852 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5853 that caused an old message to be retained and used if a later statement
5854 caused a defer, replacing the real cause of the deferral.
5856 16. If an IP address had so many PTR records that the DNS lookup buffer
5857 was not large enough to hold them, Exim could crash while trying to process
5858 the truncated data. It now detects and logs this case.
5860 17. Further to 4.21/58, another change has been made: if (and only if) the
5861 first line of a message (the first header line) ends with CRLF, a bare LF
5862 in a subsequent header line has a space inserted after it, so as not to
5863 terminate the header.
5865 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5866 unnecessarily, used atoi() instead of strtol(), and didn't check the
5867 termination when getting file sizes from file names by regex.
5869 19. Completely re-implemented the support for maildirsize files, in the light
5870 of a number of problems with the previous contributed implementation
5871 (4.30/29). In particular:
5873 . If the quota is zero, the maildirsize file is maintained, but no quota is
5876 . If the maildir directory does not exist, it is created before any attempt
5877 to write a maildirsize file.
5879 . The quota value in the file is just a cache; if the quota is changed in
5880 the transport, the new value overrides.
5882 . A regular expression is available for excluding directories from the
5885 20. The autoreply transport checks the characters in options that define the
5886 message's headers; it allows continued headers, but it was checking with
5887 isspace() after an embedded newline instead of explicitly looking for a
5890 21. If all the "regular" hosts to which an address was routed had passed their
5891 expiry times, and had not reached their retry times, the address was
5892 bounced, even if fallback hosts were defined. Now Exim should go on to try
5895 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5896 equivalent code in the SMTP transport. Some hosts send humungous responses
5897 to HELO/EHLO, more than 1024 it seems.
5899 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5900 gives compiler warnings in some environments. I've now done it "properly",
5903 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5904 (because of problems with the built-in one) was declared to return uschar *
5905 instead of char *, causing compiler failure.
5907 25. Fixed a file descriptor leak when processing alias/forward files.
5909 26. Fixed a minor format string issue in dbfn.c.
5911 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5913 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5914 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5917 29. When a sender address is verified, it is cached, to save repeating the test
5918 when there is more than one recipient in a message. However, when the
5919 verification involves a callout, it is possible for different callout
5920 options to be set for different recipients. It is too complicated to keep
5921 track of this in the cache, so now Exim always runs a verification when a
5922 callout is required, relying on the callout cache for the optimization.
5923 The overhead is duplication of the address routing, but this should not be
5926 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5927 to be verified with callout=postmaster, and the main callout worked but the
5928 postmaster check failed, the verification correctly failed. However, if a
5929 subsequent RCPT command asked for sender verification *without* the
5930 postmaster check, incorrect caching caused this verification also to fail,
5933 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5934 it was not caching the DNS options (qualify_single, search_parents) that
5935 were used when the lookup failed. A subsequent lookup with different
5936 options therefore always gave the same answer, though there were cases
5937 where it should not have. (Example: a "domains = !$mx_any" option on a
5938 dnslookup router: the "domains" option is always processed without any
5939 widening, but the router might have qualify_single set.) Now Exim uses the
5940 cached value only when the same options are set.
5942 32. Added John Jetmore's "exipick" utility to the distribution.
5944 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5945 than a timeout (e.g. a certificate is required, and is not provided), an
5946 Exim server now closes the connection immediately. Previously it waited for
5947 the client to close - but if the client is SSL, it seems that they each
5948 wait for each other, leading to a delay before one of them times out.
5950 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5951 maintained 0.8.x compatibility because I don't think many are using it, and
5952 it is clearly obsolete.
5954 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5957 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5958 client certificate was expired. A simple patch fixes this, though I don't
5959 understand the full logic of why the verify callback is called multiple
5962 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5963 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5964 which causes problems with some clients (such as the Certicom SSL Plus
5965 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5966 disables the coutermeasure allowing Eudora to connect."
5968 38. Exim was not checking that a write() to a log file succeeded. This could
5969 lead to Bad Things if a log got too big, in particular if it hit a file
5970 size limit. Exim now panics and dies if it cannot write to a log file, just
5971 as it does if it cannot open a log file.
5973 39. Modified OS/Makefile-Linux so that it now contains
5975 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5977 The two -D definitions ensure that Exim is compiled with large file
5978 support, which makes it possible to handle log files that are bigger than
5981 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5982 instance) a domain was checked against a named list that involved a lookup,
5983 causing $domain_data to be set, then another domain was checked against the
5984 same list, then the first domain was re-checked, the value of $domain_data
5985 after the final check could be wrong. In particular, if the second check
5986 failed, it could be set empty. This bug probably also applied to
5989 41. The strip_trailing_dot option was not being applied to the address given
5990 with the -f command-line option.
5992 42. The code for reading a message's header from the spool was incrementing
5993 $received_count, but never initializing it. This meant that the value was
5994 incorrect (doubled) while delivering a message in the same process in which
5995 it was received. In the most common configuration of Exim, this never
5996 happens - a fresh exec is done - but it can happen when
5997 deliver_drop_privilege is set.
5999 43. When Exim logs an SMTP synchronization error - client data sent too soon -
6000 it now includes up to 150 characters of the unexpected data in the log
6003 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
6004 and building data strings. The size of both of these buffers was 10 000
6005 bytes - far larger than anybody would *ever* want, thought I. Needless to
6006 say, somebody hit the limit. I have increased the maximum line length to
6007 20 000 and the maximum data length of concatenated lines to 100 000. I have
6008 also fixed two bugs, because there was no checking on these buffers. Tsk,
6009 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
6010 buffer is too small.
6012 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6013 lsearch lookups. Now it does.
6015 46. When parsing a route_list item in a manualroute router, a fixed-length
6016 buffer was used for the list of hosts. I made this 1024 bytes long,
6017 thinking that nobody would ever have a list of hosts that long. Wrong.
6018 Somebody had a whole pile of complicated expansion conditions, and the
6019 string was silently truncated, leading to an expansion error. It turns out
6020 that it is easier to change to an unlimited length (owing to other changes
6021 that have happened since this code was originally written) than to build
6022 structure for giving a limitation error. The length of the item that
6023 expands into the list of hosts is now unlimited.
6025 47. The lsearch lookup could not handle data where the length of text line was
6026 more than 4095 characters. Such lines were truncated, leading to shortened
6027 data being returned. It should now handle lines of any length.
6029 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6030 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6033 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6034 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6035 no longer happens in comment lines. A list of replacements is now placed
6036 at the head of all of the source files, except those whose only change is
6037 to replace PERL_COMMAND in the very first #! line.
6039 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6040 messages on the queue, with a bottom-up merge sort, using code contributed
6041 by Michael Haardt. This should make operations like -bp somewhat faster on
6042 large queues. It won't affect queue runners, except when queue_run_in_order
6045 51. Installed eximstats 1.31 in the distribution.
6047 52. Added support for SRV lookups to the dnslookup router.
6049 53. If an ACL referred to $message_body or $message_body_end, the value was not
6050 reset for any messages that followed in the same SMTP session.
6052 54. The store-handling optimization for building very long strings was not
6053 differentiating between the different store pools. I don't think this
6054 actually made any difference in practice, but I've tidied it.
6056 55. While running the routers to verify a sender address, $sender_address
6057 was still set to the sender address. This is wrong, because when routing to
6058 send a bounce to the sender, it would be empty. Therefore, I have changed
6059 it so that, while verifying a sender address, $sender_address is set to <>.
6060 (There is no change to what happens when verifying a recipient address.)
6062 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6063 target A or AAAA records (if not already returned) without resetting the
6064 qualify_single or search_parents options of the DNS resolver. These are
6065 inappropriate in this case because the targets of MX and SRV records must
6066 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6067 target that, when qualified, matched something in the local domain. These
6068 two options are now turned off when doing these lookups.
6070 57. It seems that at least some releases of Reiserfs (which does not have the
6071 concept of a fixed number of inodes) returns zero and not -1 for the
6072 number of available inodes. This interacted badly with check_spool_inodes,
6073 which assumed that -1 was the "no such thing" setting. What I have done is
6074 to check that the total number of inodes is greater than zero before doing
6075 the test of how many are available.
6077 58. When a "warn" ACL statement has a log_message modifier, the message is
6078 remembered, and not repeated. This is to avoid a lot of repetition when a
6079 message has many recipients that cause the same warning to be written.
6080 However, Exim was preserving the list of already written lines for an
6081 entire SMTP session, which doesn't seem right. The memory is now reset if a
6082 new message is started.
6084 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6085 debugging output unless log_rewrite was also set.
6087 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6088 of (int)(handle) when we know that handle contains (void *)(-1).
6090 61. The Exim daemon panic-logs an error return when it closes the incoming
6091 connection. However "connection reset by peer" seems to be common, and
6092 isn't really an error worthy of noting specially, so that particular error
6095 62. When Exim is trying to find all the local interfaces, it used to panic and
6096 die if the ioctl to get the interface flags failed. However, it seems that
6097 on at least one OS (Solaris 9) it is possible to have an interface that is
6098 included in the list of interfaces, but for which you get a failure error
6099 for this call. This happens when the interface is not "plumbed" into a
6100 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6101 failure of the "get flags" call assumes that the interface is down.
6103 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6104 makes life easier for people who are doing arithmetic on fields extracted
6105 from dates, where you often get leading zeros that should not be
6106 interpreted as octal.
6108 64. Added qualify_domain to the redirect router, to override the global
6111 65. If a pathologically long header line contained very many addresses (the
6112 report of this problem mentioned 10 000) and each of them was rewritten,
6113 Exim could use up a very large amount of memory. (It kept on making new
6114 copies of the header line as it rewrote, and never released the old ones.)
6115 At the expense of a bit more processing, the header rewriting function has
6116 been changed so that it no longer eats memory in this way.
6118 66. The generation of the Received: header has been moved from the time that a
6119 message starts to be received, to the time that it finishes. The timestamp
6120 in the Received: header should now be very close to that of the <= log
6121 line. There are two side-effects of this change:
6123 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6124 logged header lines no longer include the local Received: line, because
6125 it has not yet been created. The same applies to a copy of the message
6126 that is returned to a non-SMTP sender when a message is rejected.
6128 (b) When a filter file is tested using -bf, no additional Received: header
6129 is added to the test message. After some thought, I decided that this
6132 This change does not affect the value of $received_for. It is still set
6133 after address rewriting, but before local_scan() is called.
6135 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6137 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6138 gave an unhelpful panic error message, and a defer error. I have managed to
6139 change this behaviour so that it now rejects any supplied certificate,
6140 which seems right, as the list of acceptable certificates is empty.
6142 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6143 gave an unhelpful defer error. I have not managed to make this reject any
6144 supplied certificates, but the error message it gives is "no certificate
6145 supplied", which is not helpful.
6147 70. exigrep's output now also includes lines that are not associated with any
6148 message, but which match the given pattern. Implemented by a patch from
6149 Martin Sluka, which also tidied up the Perl a bit.
6151 71. Recipient callout verification, like sender verification, was using <> in
6152 the MAIL FROM command. This isn't really the right thing, since the actual
6153 sender may affect whether the remote host accepts the recipient or not. I
6154 have changed it to use the actual sender in the callout; this means that
6155 the cache record is now keyed on a recipient/sender pair, not just the
6156 recipient address. There doesn't seem to be a real danger of callout loops,
6157 since a callout by the remote host to check the sender would use <>.
6158 [SEE ABOVE: changed after hitting problems.]
6160 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6161 temporary errors. However, in the case of such a code being given after
6162 the end of a data transmission (i.e. after ".") Exim was failing to write
6163 a retry record for the message. (Yes, there was some broken host that was
6164 actually sending 8xx at this point.)
6166 73. An unknown lookup type in a host list could cause Exim to panic-die when
6167 the list was checked. (An example that provoked this was putting <; in the
6168 middle of a list instead of at the start.) If this happened during a DATA
6169 ACL check, a -D file could be left lying around. This kind of configuration
6170 error no longer causes Exim to die; instead it causes a defer error. The
6171 incident is still logged to the main and panic logs.
6173 74. Buglet left over from Exim 3 conversion. The message "too many messages
6174 in one connection" was written to the rejectlog but not the mainlog, except
6175 when address rewriting (yes!) was being logged.
6177 75. Added write_rejectlog option.
6179 76. When a system filter was run not as root (that is, when system_filter_user
6180 was set), the values of the $n variables were not being returned to the
6181 main process; thus, they were not subsequently available in the $sn
6184 77. Added +return_path_on_delivery log selector.
6186 78. A connection timeout was being treated differently from recipients deferred
6187 when testing hosts_max_try with a message that was older than the host's
6188 retry timeout. (The host should not be counted, thus allowing all hosts to
6189 be tried at least once before bouncing.) This may have been the cause of an
6190 occasionally reported bug whereby a message would remain on the queue
6191 longer than the retry timeout, but would be bounced if a delivery was
6192 forced. I say "may" because I never totally pinned down the problem;
6193 setting up timeout/retry tests is difficult. See also the next item.
6195 79. The ultimate address timeout was not being applied to errors that involved
6196 a combination of host plus message (for example, a timeout on a MAIL
6197 command). When an address resolved to a number of possible hosts, and they
6198 were not all tried for each delivery (e.g. because of hosts_max_try), a
6199 message could remain on the queue longer than the retry timeout.
6201 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6204 81. Fixed an obscure SMTP outgoing bug which required at least the following
6205 conditions: (a) there was another message waiting for the same server;
6206 (b) the server returned 5xx to all RCPT commands in the first message so
6207 that the message was not completed; (c) the server dropped the connection
6208 or gave a negative response to the RSET that Exim sends to abort the
6209 transaction. The observed case was a dropped connection after DATA that had
6210 been sent in pipelining mode. That is, the server had advertised PIPELINING
6211 but was not implementing it correctly. The effect of the bug was incorrect
6212 behaviour, such as trying another host, and this could lead to a crash.
6218 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6219 and daemon.c were passed as pointers to ints; they should have been
6220 pointers to socklen_t variables (which are typically unsigned ints).
6222 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6225 3. Fixed a really odd bug that affected only the testing scheme; patching a
6226 certain fixed string in the binary changed the value of another string that
6227 happened to be identical to the end of the original first string.
6229 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6230 name", it returns that address as the IP address. On some operating
6231 systems (e.g. Solaris), it also passes back the IP address string as the
6232 "host name". However, on others (e.g. Linux), it passes back an empty
6233 string. Exim wasn't checking for this, and was changing the host name to an
6234 empty string, assuming it had been canonicalized.
6236 5. Although rare, it is permitted to have more than one PTR record for a given
6237 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6238 all the names associated with an address, because they do in Solaris.
6239 However, it seems that they do not in Linux for data that comes from the
6240 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6241 I found this out when I moved to a new Linux workstation and tried to run
6242 the Exim test suite.
6244 To get round this problem I have changed the code so that it now does its
6245 own call to the DNS to look up PTR records when searching for a host name.
6246 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6247 addresses that are only in /etc/hosts are still found.
6249 This behaviour is, however, controlled by an option called host_lookup_
6250 order, which defaults to "bydns:byaddr". If people want to use the other
6251 order, or indeed, just use one or the other means of lookup, they can
6252 specify it in this variable.
6254 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6255 some operating systems, this comes back from gethostbyaddr() as an empty
6256 string, and this is what Exim used to test for. However, it seems that in
6257 other systems, "." is yielded. Exim now tests for this case too.
6259 7. The values of check_spool_space and check_log_space are now held internally
6260 as a number of kilobytes instead of an absolute number of bytes. If a
6261 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6262 kilobyte. This means that much larger values can be stored.
6264 8. Exim monitor: an attempt to get the action menu when not actually pointing
6265 at a message produces an empty menu entitled "No message selected". This
6266 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6267 no entries in it ("Shell widget menu has zero width and/or height"). So I
6268 have added a single, blank menu entry in this case.
6270 9. Added ${quote_local_part.
6272 10. MIME decoding is now applied to the contents of Subject: header lines when
6275 11. Now that a reference to $sender_host_address automatically causes a reverse
6276 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6277 host lookup before query-style lookups in lists that might use this
6278 variable. This has therefore been abolished, and the "net-" prefix is no
6279 longer necessary for query-style lookups.
6281 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6282 have been a typo for LFLAGS, so it has been changed.
6284 13. The install script calls Exim with "-C /dev/null" in order to find the
6285 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6286 to be output. However, since Exim outputs its version number before the
6287 error, it didn't break the script. It just looked ugly. I fixed this by
6288 always allowing "-C /dev/null" if the caller is root.
6290 14. Ignore overlarge ACL variable number when reading spool file - insurance
6291 against a later release with more variables having written the file.
6293 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6294 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6295 this, as well as the form without the "IPv6" on the front.
6297 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6298 OS/Makefile-Darwin file.
6300 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6301 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6303 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6304 variable. However, some OS use ints for inodes. I've added cast to long int
6305 to get rid of the compiler warning.
6307 19. I had forgotten to lock out "/../" in configuration file names when
6308 ALT_CONFIG_PREFIX was set.
6310 20. Routers used for verification do not need to specify transports. However,
6311 if such a router generated a host list, and callout was configured, Exim
6312 crashed, because it could not find a port number from the (non-existent)
6313 transport. It now assumes port 25 in this circumstance.
6315 21. Added the -t option to exigrep.
6317 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6318 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6319 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6320 them is set, LOOKUP_LSEARCH is forced.
6322 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6324 24. Added sender and host information to the "rejected by local_scan()" log
6325 line; previously there was no indication of these.
6327 25. Added .include_if_exists.
6329 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6330 Linux. It turns out that not all file systems support this. Apparently some
6331 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6332 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6333 not supported on the file descriptor, is now ignored when Exim is trying to
6334 sync a directory. This applies only to Linux.
6336 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6338 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6339 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6342 29. Added support for maildirsize files from supplied patch (modified a bit).
6344 30. The use of :fail: followed by an empty string could lead Exim to respond to
6345 sender verification failures with (e.g.):
6347 550 Verification failed for <xxx>
6348 550 Sender verify failed
6350 where the first response line was missing the '-' that indicates it is not
6351 the final line of the response.
6353 31. The loop for finding the name of the user that called Exim had a hardwired
6354 limit of 10; it now uses the value of finduser_retries, which is used for
6355 all other user lookups.
6357 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6360 33. Exim was neglecting to zero errno before one call of strtol() when
6361 expanding a string and expecting an integer value. On some systems this
6362 resulted in spurious "integer overflow" errors. Also, it was casting the
6363 result into an int without checking.
6365 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6366 did not work. The code looks as if it has *never* worked, though it appears
6367 to have been documented since at least release 1.62. I have made it work.
6369 35. The "timeout_DNS" error in retry rules, also documented since at least
6370 1.62, also never worked. As it isn't clear exactly what this means, and
6371 clearly it isn't a major issue, I have abolished the feature by treating it
6372 as "timeout", and writing a warning to the main and panic logs.
6374 36. The display of retry rules for -brt wasn't always showing the error code
6377 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6378 timeout_connect_A, timeout_connect_MX.
6380 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6381 to the empty sender.
6383 39. The daemon was not analysing the content of -oX till after it had closed
6384 stderr and disconnected from the controlling terminal. This meant that any
6385 syntax errors were only noted on the panic log, and the return code from
6386 the command was 0. By re-arranging the code a little, I've made the
6387 decoding happen first, so such errors now appear on stderr, and the return
6388 code is 1. However, the actual setting up of the sockets still happens in
6389 the disconnected process, so errors there are still only recorded on the
6392 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6393 connections (as happens on some IP stacks) was logged at start up time as
6394 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6395 it from "IPv6 and IPv4", which means that two separate sockets are being
6398 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6399 says whether AF_INET or AF_INET6 was passed as an argument.
6401 42. Exiwhat output was messed up when time zones were included in log
6404 43. Exiwhat now gives more information about the daemon's listening ports,
6405 and whether -tls-on-connect was used.
6407 44. The "port" option of the smtp transport is now expanded.
6409 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6410 silently ignored. Now an error message is written to the main and panic
6413 46. There's a new ACL modifier called "logwrite" which writes to a log file
6414 as soon as it is encountered.
6416 47. Added $local_user_uid and $local_user_gid at routing time.
6418 48. Exim crashed when trying to verify a sender address that was being
6421 49. Exim was recognizing only a space character after ".include". It now also
6422 recognizes a tab character.
6424 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6425 extracting the relevant information from the specification. The man page no
6426 longer contains scrambled data for the -d option, and I've added a section
6427 at the front about calling Exim under different names.
6429 51. Added "extra_headers" argument to the "mail" command in filter files.
6431 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6434 53. Installed eximstats 1.29.
6436 54. Added transport_filter_timeout as a generic transport option.
6438 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6439 Cc: header lines. This was required by RFC 822, but it not required by RFC
6442 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6443 incoming messages that did not have them. Now it does so only if the
6444 message originates locally, that is, if there is no associated remote host
6445 address. When Resent- header lines are present, this applies to the Resent-
6446 lines rather than the non-Resent- lines.
6448 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6449 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6451 58. Messages for configuration errors now include the name of the main
6452 configuration file - useful now that there may be more than one file in a
6453 list (.included file names were always shown).
6455 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6456 for those rare installations that do not start the daemon as root or run it
6457 setuid root. I've cut out the call to initgroups() if the daemon is not
6460 60. The Exim user and group can now be bound into the binary as text strings
6461 that are looked up at the start of Exim's processing.
6463 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6465 62. Added $mailstore_basename variable.
6467 63. Installed patch to sieve.c from Michael Haardt.
6469 64. When Exim failed to open the panic log after failing to open the main log,
6470 the original message it was trying to log was written to stderr and debug
6471 output, but if they were not available (the usual case in production), it
6472 was lost. Now it is written to syslog before the two lines that record the
6473 failures to open the logs.
6475 65. Users' Exim filters run in subprocesses under the user's uid. It is
6476 possible for a "deliver" command or an alias in a "personal" command to
6477 provoke an address rewrite. If logging of address rewriting is configured,
6478 this fails because the process is not running as root or exim. There may be
6479 a better way of dealing with this, but for the moment (because 4.30 needs
6480 to be released), I have disabled address rewrite logging when running a
6481 filter in a non-root, non-exim process.
6487 1. The buildconfig auxiliary program wasn't quoting the value set for
6488 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6489 not defined. This bug was masked in 4.22 by the effect that was fixed in
6492 2. Some messages that were rejected after a message id was allocated were
6493 shown as "incomplete" by exigrep. It no longer does this for messages that
6494 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6496 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6497 have allow_domain_literals set, the ID did not get logged in the <= line.
6498 Domain literals are now always recognized in Message-ID: header lines.
6500 4. The first argument for a ${extract expansion item is the key name or field
6501 number. Leading and trailing spaces in this item were not being ignored,
6502 causing some misleading effects.
6504 5. When deliver_drop_privilege was set, single queue runner processes started
6505 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6506 same command in the process it spins off) were not dropping privilege.
6508 6. When the daemon running as "exim" started a queue runner, it always
6509 re-executed Exim in the spun-off process. This is a waste of effort when
6510 deliver_drop_privilege is set. The new process now just calls the
6511 queue-runner function directly.
6517 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6520 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6521 ignored. Though the use of -oP was forcing the writing of a pid file, it
6522 was always written to the default place.
6524 3. If the message "no IP address found for host xxxx" is generated during
6525 incoming verification, it is now followed by identification of the incoming
6526 connection (so you can more easily find what provoked it).
6528 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6530 5. Added some features to "harden" Exim a bit more against certain attacks:
6532 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6533 be put in Local/Makefile. This is like the never_users runtime option,
6534 but it cannot be overridden. The default setting is "root".
6536 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6537 prefix string with which any file named in a -C command line option
6540 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6541 is retained for -C and -D only if the caller of Exim is root. Without
6542 it, the exim user may also use -C and -D and retain privilege.
6544 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6545 command line option is disabled.
6547 6. Macro names set by the -D option must start with an upper case letter, just
6548 like macro names defined in the configuration file.
6550 7. Added "dereference=" facility to LDAP.
6552 8. Two instances of the typo "uknown" in the source files are fixed.
6554 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6555 the Configure-Makefile script screwed up while processing it.
6557 10. Incorporated PCRE 4.4.
6559 11. The SMTP synchronization check was not operating right at the start of an
6560 SMTP session. For example, it could not catch a HELO sent before the client
6561 waited for the greeting. There is now a check for outstanding input at the
6562 point when the greeting is written. Because of the duplex, asynchronous
6563 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6564 way, but not yet received, when the check is performed.
6566 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6567 on TCP/IP sockets, because this apparently causes some broken clients to
6570 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6571 unchanged) from the Cygwin maintainer.
6573 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6574 was supported instead of testing for mbx. Effectively a typo.
6576 15. The spa authenticator server code was not checking that the input it
6577 received was valid base64.
6579 16. The debug output line for the "set" modifier in ACLs was not showing the
6580 name of the variable that was being set.
6582 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6584 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6585 lookup on its own. Something else was needed to trigger the lookup. For
6586 example, a match in host_lookup or the need for a host name in a host list.
6587 Now, if $sender_host_name is referenced and the host name has not yet been
6588 looked up, a lookup is performed. If the lookup fails, the variable remains
6589 empty, and $host_lookup_failed is set to "1".
6591 19. Added "eqi" as a case-independent comparison operator.
6593 20. The saslauthd authentication condition could segfault if neither service
6594 nor realm was specified.
6596 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6597 error message that was logged was misleading, and incoming SMTP
6598 connections were dropped. The message is now more accurate, and temporary
6599 errors are given to SMTP connections.
6601 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6602 (except RSET) with a 421 error, until QUIT is received. However, it was
6603 failing to send a response to QUIT.
6605 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6606 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6607 ACL is now run after the helo_try_verify_hosts code.
6609 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6610 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6611 case-independent, so other case variants are also recognized). Apparently
6612 some systems use these upper case variants.
6614 25. If more than two messages were waiting for the same host, and a transport
6615 filter was specified for the transport, Exim sent two messages over the
6616 same TCP/IP connection, and then failed with "socket operation on non-
6617 socket" when it tried to send the third.
6619 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6621 27. The extern definition of crypt16() in expand.c was not being excluded when
6622 the OS had its own crypt16() function.
6624 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6625 as a preferred synonym for return_size_limit, both as an option and as an
6628 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6630 30. Changed the default configuration ACL to relax the local part checking rule
6631 for addresses that are not in any local domains. For these addresses,
6632 slashes and pipe symbols are allowed within local parts, but the sequence
6633 /../ is explicitly forbidden.
6635 31. SPA server authentication was not clearing the challenge buffer before
6638 32. log_message in a "warn" ACL statement was writing to the reject log as
6639 well as to the main log, which contradicts the documentation and doesn't
6640 seem right (because no rejection is happening). So I have stopped it.
6642 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6643 However, I am unable to do any testing of this.
6645 34. Fixed an infelicity in the appendfile transport. When checking directories
6646 for a mailbox, to see if any needed to be created, it was accidentally
6647 using path names with one or more superfluous leading slashes; tracing
6648 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6650 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6651 contents are added to the log line that is written for every discarded
6652 recipient. (Previously a log_message setting was ignored.)
6654 36. The ${quote: operator now quotes the string if it is empty.
6656 37. The install script runs exim in order to find its version number. If for
6657 some reason other than non-existence or emptiness, which it checks, it
6658 could not run './exim', it was installing it with an empty version number,
6659 i.e. as "exim-". This error state is now caught, and the installation is
6662 38. An argument was missing from the function that creates an error message
6663 when Exim fails to connect to the socket for saslauthd authentication.
6664 This could cause Exim to crash, or give a corrupted message.
6666 39. Added isip, isip4, and isip6 to ${if conditions.
6668 40. The ACL variables $acl_xx are now saved with the message, and can be
6669 accessed later in routers, transports, and filters.
6671 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6672 strings in the file are not string-expanded.
6674 42. If a MAIL command specified a SIZE value that was too large to fit into an
6675 int variable, the check against message_size_limit failed. Such values are
6676 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6677 day this will have to be increased, but I don't think I want to be around
6678 when emails are that large.
6685 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6686 iconv() is not standard in FreeBSD.
6688 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6689 IPv6 enabled. The observed symptom was a segmentation fault on return from
6690 the function os_common_find_running_interfaces() in src/os.c.
6692 3. In the check_special_case() function in daemon.c I had used "errno" as an
6693 argument name, which causes warnings on some systems. This was basically a
6694 typo, since it was named "eno" in the comments!
6696 4. The code that waits for the clock to tick (at a resolution of some fraction
6697 of a second) so as to ensure message-id uniqueness was always waiting for
6698 at least one whole tick, when it could have waited for less. [This is
6699 almost certainly not relevant at current processor speeds, where it is
6700 unlikely to ever wait at all. But we try to future-proof.]
6702 5. The function that sleeps for a time interval that includes fractions of a
6703 second contained a race. It did not block SIGALRM between setting the
6704 timer, and suspending (a couple of lines later). If the interval was short
6705 and the sigsuspend() was delayed until after it had expired, the suspension
6706 never ended. On busy systems this could lead to processes getting stuck for
6709 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6710 process, before it forks any delivery processes. The open lookup caching
6711 mechanism meant that the open file or database connection was passed into
6712 the delivery process. The problem was that delivery processes always tidy
6713 up cached lookup data. This could cause a problem for the next delivery
6714 process started by the queue runner, because the external queue runner
6715 process does not know about the closure. So the next delivery process
6716 still has data in the lookup cache. In the case of a file lookup, there was
6717 no problem because closing a file descriptor in a subprocess doesn't affect
6718 the parent. However, if the lookup was caching a connection to a database,
6719 the connection was closed, and the second delivery process was likely to
6720 see errors such as "PGSQL: query failed: server closed the connection
6721 unexpectedly". The problem has been fixed by closing all cached lookups
6722 in a queue runner before running a delivery process.
6724 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6725 seem to have the "const" qualifier which it has on other OS. I've
6728 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6729 *of the same type* (client or server) with the same public name that an
6730 error should be diagnosed.
6732 9. When Exim looked up a host name for an IP address, but failed to find the
6733 original IP address when looking up the host name (a safety check), it
6734 output the message "<ip address> does not match any IP for NULL", which was
6735 confusing, to say the least. The bug was that the host name should have
6736 appeared instead of "NULL".
6738 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6739 user that is built into the binary, and the -C or -D options is used, root
6740 privilege is dropped before the configuration file is read. In addition,
6741 logging is switched to stderr instead of the normal log files. If the
6742 configuration then re-defines the Exim user, the unprivileged environment
6743 is probably not what is expected, so Exim logs a panic warning message (but
6746 However, if deliver_drop_privilege is set, the unprivileged state may well
6747 be exactly what is intended, so the warning has been cut out in that case,
6748 and Exim is allowed to try to write to its normal log files.
6754 1. smtp_return_error_details was not giving details for temporary sender
6755 or receiver verification errors.
6757 2. Diagnose a configuration error if two authenticators have the same public
6760 3. Exim used not to create the message log file for a message until the first
6761 delivery attempt. This could be confusing when incoming messages were held
6762 for policy or load reasons. The message log file is now created at the time
6763 the message is received, and an initial "Received" line is written to it.
6765 4. The automatically generated man page for command line options had a minor
6766 bug that caused no ill effects; however, a more serious problem was that
6767 the procedure for building the man page automatically didn't always
6768 operate. Consequently, release 4.20 contains an out-of-date version. This
6769 shouldn't happen again.
6771 5. When building Exim with embedded Perl support, the script that builds the
6772 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6773 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6775 6. The freeze_tell option was not being used for messages that were frozen on
6776 arrival, either by an ACL or by local_scan().
6778 7. Added the smtp_incomplete_transaction log selector.
6780 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6781 was accepting AUTH without a new EHLO.
6783 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6784 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6787 10. Logging of TCP/IP connections (when configured) now happens in the main
6788 daemon process instead of the child process, so that the TCP/IP connection
6789 count is more accurate (but it can never be perfect).
6791 11. The use of "drop" in a nested ACL was not being handled correctly in the
6792 outer ACL. Now, if condition failure induced by the nested "drop" causes
6793 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6794 or "require"), the connection is dropped.
6796 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6797 that yield "discard" can now be used with an "accept" or a "discard" verb,
6798 but an error is generated for any others (because I can't see a useful way
6799 to define what should happen).
6801 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6802 are now processed in the same way as lines in the Exim configuration file.
6803 In particular, continuation lines are supported.
6805 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6807 15. Added -ti meaning -t -i.
6809 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6810 domains, and warn by logging if others are found.
6812 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6813 returned from GIFCONF when Exim is trying to find the list of interfaces on
6814 a host. The code in os.c has been modified to copy each ifreq to an aligned
6815 structure in all cases.
6817 Also, in some cases, the returned ifreq's were being copied to a 'struct
6818 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6819 means the last couple of bytes of an IPv6 address could be chopped if the
6820 ifreq contained only a normal sockaddr (14 bytes storage).
6822 18. Named domain lists were not supported in the hosts_treat_as_local option.
6823 An entry such as +xxxx was not recognized, and was treated as a literal
6826 19. Ensure that header lines added by a DATA ACL are included in the reject log
6827 if the ACL subsequently rejects the message.
6829 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6830 MD5 (which is deprecated).
6832 21. When testing a filter file using -bf, Exim was writing a message when it
6833 took the sender from a "From " line in the message, but it was not doing so
6834 when it took $return_path from a Return-Path: header line. It now does.
6836 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6837 with a valid header line field name (a series of printing characters
6838 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6840 23. Changed "disc" in the source to "disk" to conform to the documentation and
6841 the book and for uniformity.
6843 24. Ignore Sendmail's -Ooption=value command line item.
6845 25. When execve() failed while trying to run a command in a pipe transport,
6846 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6847 could be confused with a return value of 69 from the command itself. This
6848 has been changed to 127, the value the shell returns if it is asked to run
6849 a non-existent command. The wording for the related log line suggests a
6850 non-existent command as the problem.
6852 26. If received_header_text expands to an empty string, do not add a Received:
6853 header line to the message. (Well, it adds a token one on the spool, but
6854 marks it "old" so that it doesn't get used or transmitted.)
6856 27. Installed eximstats 1.28 (addition of -nt option).
6858 28. There was no check for failure on the call to getsockname() in the daemon
6859 code. This can fail if there is a shortage of resources on the system, with
6860 ENOMEM, for example. A temporary error is now given on failure.
6862 29. Contrary to the C standard, it seems that in some environments, the
6863 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6864 program. Exim now does this explicitly; it affects the formatting of
6865 timestamps using strftime().
6867 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6868 complaints. I've now initialized all the variables, to avoid this.
6870 32. Header lines added by a system filter were not being "seen" during
6871 transport-time rewrites.
6873 33. The info_callback() function passed to OpenSSL is set up with type void
6874 (*)(SSL *, int, int), as described somewhere. However, when calling the
6875 function (actually a macro) that sets it up, the type void(*)() is
6876 expected. I've put in a cast to prevent warnings from picky compilers.
6878 34. If a DNS black list lookup found a CNAME record, but there were no A
6879 records associated with the domain it pointed at, Exim crashed.
6881 35. If a DNS black list lookup returned more than one A record, Exim ignored
6882 all but the first. It now scans all returned addresses if a particular IP
6883 value is being sought. In this situation, the contents of the
6884 $dnslist_value variable are a list of all the addresses, separated by a
6887 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6888 to accept a host name and all its aliases if the forward lookup for any of
6889 them yielded the IP address of the incoming connection. Now it accepts only
6890 those names whose forward lookup yields the correct IP address. Any other
6891 names are discarded. This closes a loophole whereby a rogue DNS
6892 administrator could create reverse DNS records to break through a
6893 wildcarded host restriction in an ACL.
6895 37. If a user filter or a system filter that ran in a subprocess used any of
6896 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6897 the wrong values were passed to the pipe command ($thisaddress had the
6898 value of $0, $0 had the value of $1, etc). This bug was introduced by
6899 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6901 38. Improved the line breaking for long SMTP error messages from ACLs.
6902 Previously, if there was no break point between 40 and 75 characters, Exim
6903 left the rest of the message alone. Two changes have been made: (a) I've
6904 reduced the minimum length to 35 characters; (b) if it can't find a break
6905 point between 35 and 75 characters, it looks ahead and uses the first one
6906 that it finds. This may give the occasional overlong line, but at least the
6907 remaining text gets split now.
6909 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6910 file descriptors might be low, and that setting 1000 would always raise it.
6911 It turns out that in some environments, the limit is already over 1000 and
6912 that lowering it causes trouble. So now Exim takes care not to decrease it.
6914 40. When delivering a message, the value of $return_path is set to $sender_
6915 address at the start of routing (routers may change the value). By an
6916 oversight, this default was not being set up when an address was tested by
6917 -bt or -bv, which affected the outcome if any router or filter referred to
6920 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6921 log only when "message" or "log_message" are set. However, if one of the
6922 conditions was an address verification, or a call to a nested ACL, the
6923 messages generated by the underlying test were being passed through. This
6924 no longer happens. The underlying message is available in $acl_verify_
6925 message for both "message" and "log_message" expansions, so it can be
6926 passed through if needed.
6928 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6929 new expansion $bh_ to give the encoded byte string without charset
6930 translation. Translation happens only if iconv() is available; HAVE_ICONV
6931 indicates this at build time. HEADERS_CHARSET gives the charset to
6932 translate to; headers_charset can change it in the configuration, and
6933 "headers charset" can change it in an individual filter file.
6935 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6936 that creates RFC 2047 encoded "words" labels them as that charset instead
6937 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6938 expansion operator; (ii) when Exim creates a From: line for a local
6939 message; (iii) when a header line is rewritten to include a "phrase" part.
6941 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6942 buggy, causing it to skip the first lines of messages whose message ID
6943 ended in 'D'. This would not have bitten before Exim release 4.14, because
6944 message IDs were unlikely to end in 'D' before then. The effect was to have
6945 incorrect size information for certain domains.
6947 45. #include "config.h" was missing at the start of the crypt16.c module. This
6948 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6951 46. If there was a timeout during a "random" callout check, Exim treated it as
6952 a failure of the random address, and carried on sending RSET and the real
6953 address. If the delay was just some slowness somewhere, the response to the
6954 original RCPT would be taken as a response to RSET and so on, causing
6955 mayhem of various kinds.
6957 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6958 when I implemented it. It didn't allow for the fact that some option values
6959 may legitimately be negative (e.g. size_addition), and it didn't even do
6960 the right test for positive values.
6962 48. Domain names in DNS records are case-independent. Exim always looks them up
6963 in lower case. Some resolvers return domain names in exactly the case they
6964 appear in the zone file, that is, they may contain uppercase letters. Not
6965 all resolvers do this - some return always lower case. Exim was treating a
6966 change of case by a resolver as a change of domain, similar to a widening
6967 of a domain abbreviation. This triggered its re-routing code and so it was
6968 trying to route what was effectively the same domain again. This normally
6969 caused routing to fail (because the router wouldn't handle the domain
6970 twice). Now Exim checks for this case specially, and just changes the
6971 casing of the domain that it ultimately uses when it transmits the message
6974 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6977 50. If a filter generated a file delivery with a non-absolute name (possible if
6978 no home directory exists for the router), the forbid_file option was not
6981 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6982 the existing equality matching.
6984 52. Exim was using ints instead of ino_t variables in some places where it was
6985 dealing with inode numbers.
6987 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6988 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6989 called TMPDIR, and if it finds it is different, it changes its value.
6991 54. The smtp_printf() function is now made available to local_scan() so
6992 additional output lines can be written before returning. There is also an
6993 smtp_fflush() function to enable the detection of a dropped connection.
6994 The variables smtp_input and smtp_batched_input are exported to
6997 55. Changed the default runtime configuration: the message "Unknown user"
6998 has been removed from the ACL, and instead placed on the localuser router,
6999 using the cannot_route_message feature. This means that any verification
7000 failures that generate their own messages won't get overridden. Similarly,
7001 the "Unrouteable address" message that was in the ACL for unverifiable
7002 relay addresses has also been removed.
7004 56. Added hosts_avoid_esmtp to the smtp transport.
7006 57. The exicyclog script was not checking for the esoteric option
7007 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
7008 will work only if exicyclog is run under the appropriate euid.
7010 58. Following a discussion on the list, the rules by which Exim recognises line
7011 endings on incoming messages have been changed. The -dropcr and drop_cr
7012 options are now no-ops, retained only for backwards compatibility. The
7013 following line terminators are recognized: LF CRLF CR. However, special
7014 processing applies to CR:
7016 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7017 nor a local message in the state where . is a terminator.
7019 (ii) If a bare CR is encountered in a header line, an extra space is added
7020 after the line terminator so as not to end the header. The reasoning
7021 behind this is that bare CRs in header lines are most likely either
7022 to be mistakes, or people trying to play silly games.
7024 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7025 was being incorrectly given as 18 bytes larger than it should have been.
7026 This is a VOB (very old bug).
7028 60. This may never have affected anything current, but just in case it has:
7029 When the local host is found other than at the start of a list of hosts,
7030 the local host, those with the same MX, and any that follow, are discarded.
7031 When the list in question was part of a longer list of hosts, the following
7032 hosts (not currently being processed) were also being discarded. This no
7033 longer happens. I'm not sure if this situation could ever has previously
7036 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7039 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7040 header. This is something that is recommended in a new Internet Draft, and
7041 is something that is documented as being done by Sendmail. There are two
7042 possible values. For messages generated by the autoreply transport, Exim
7045 Auto-Submitted: auto-replied
7047 whereas for all other generated messages (e.g. bounces) it adds
7049 Auto-Submitted: auto-generated
7051 63. The "personal" condition in filters now includes a test for the
7052 Auto-Submitted: header. If it contains the string "auto-" the message it
7053 not considered personal.
7055 64. Added rcpt_include_affixes as a generic transport option.
7057 65. Added queue_only_override (default true).
7059 66. Added the syslog_duplication option.
7061 67. If what should have been the first header line of a message consisted of
7062 a space followed by a colon, Exim was mis-interpreting it as a header line.
7063 It isn't of course - it is syntactically invalid and should therefore be
7064 treated as the start of the message body. The misbehaviour could have
7065 caused a number of strange effects, including loss of data in subsequent
7066 header lines, and spool format errors.
7068 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7069 client host had authenticated. This control can now be exercised by an ACL
7070 for more flexibility.
7072 69. By default, callouts do not happen when testing with -bh. There is now a
7073 variant, -bhc, which does actually run the callout code, including
7074 consulting and updating the callout cache.
7076 70. Added support for saslauthd authentication, courtesy of Alexander
7079 71. If statvfs() failed on the spool or log directories while checking their
7080 size for availability, Exim confusingly gave the error "space shortage".
7081 Furthermore, in debugging mode it crashed with a floating point exception.
7082 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7083 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7084 serious problem, Exim now writes to the main and panic logs when this
7085 happens, with details of the failure. It then refuses to accept the
7086 incoming message, giving the message "spool directory problem" or "log
7087 directory problem" with a 421 code for SMTP messages.
7089 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7090 0, 1, and 2 exist, because some OS complain for execs without them (see
7091 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7092 descriptors. However, the code omitted to check that the open succeeded,
7093 causing mysterious errors if for some reason the permissions on /dev/null
7094 got screwed. Now Exim writes a message to the main and panic logs, and
7095 bombs out if it can't open /dev/null.
7097 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7098 interact so that it is all more flexible. It is supposed to remain
7099 backwards compatible. Also added extra_local_interfaces.
7101 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7102 to bomb out with an assertion failure - to the client this appears as a
7103 connection drop. This problem occurs in the part of the code that was taken
7104 from the Samba project. Fortunately, the assertion is in a very simple
7105 function, so I have fixed this by reproducing the function inline in the
7106 one place where it is called, and arranging for authentication to fail
7107 instead of killing the process with assert().
7109 75. The SPA client code was not working when the server requested OEM rather
7110 than Unicode encoding.
7112 76. Added code to make require_files with a specific uid setting more usable in
7113 the case where statting the file as root fails - usually a non-root-mounted
7114 NFS file system. When this happens and the failure is EACCES, Exim now
7115 forks a subprocess and does the per-uid checking as the relevant uid.
7117 77. Added process_log_path.
7119 78. If log_file_path was not explicitly set, a setting of check_log_space or
7120 check_log_inodes was ignored.
7122 79. If a space check for the spool or log partitions fails, the incident is now
7123 logged. Of course, in the latter case the data may get lost...
7125 80. Added the %p formatting code to string_format() so that it can be used to
7126 print addresses in debug_print(). Adjusted all the address printing in the
7127 debugging in store.c to use %p rather than %d.
7129 81. There was a concern that a line of code in smtp_in.c could overflow a
7130 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7131 initially expressed, the concern was not well-founded, because trailing
7132 spaces are removed early. However, if the trailing spaces were followed by
7133 a NULL, they did not get removed, so the overflow was possible. Two fixes
7136 (a) I re-wrote the offending code in a cleaner fashion.
7137 (b) If an incoming SMTP command contains a NULL character, it is rejected
7140 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7141 runs initgroups(), so that if the Exim user is in any additional groups,
7142 they will be used during message reception.
7148 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / jgh / exim.git / blob