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.
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
190 PP/02 DANE: add dane_require_tls_ciphers SMTP Transport option; if unset,
191 tls_require_ciphers is used as before.
193 HS/03 Malware Avast: Better match the Avast multiline protocol.
194 Only tmpfails from the scanner are written to the paniclog, as
195 they may require admin intervention (permission denied, license
196 issues). Other scanner errors (like decompression bombs) do not
197 cause a paniclog entry.
203 JH/01 Rework error string handling in TLS interface so that the caller in
204 more cases is responsible for logging. This permits library-sourced
205 string to be attached to addresses during delivery, and collapses
206 pairs of long lines into single ones.
208 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
209 during configuration. Wildcards are allowed and expanded.
211 JH/02 Rework error string handling in DKIM to pass more info back to callers.
212 This permits better logging.
214 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
215 do not close it down and have the child transport start it up again on
216 the passed-on TCP connection. Instead, proxy the child (and any
217 subsequent ones) for TLS via a unix-domain socket channel. Logging is
218 affected: the continued delivery log lines do not have any DNSSEC, TLS
219 Certificate or OCSP information. TLS cipher information is still logged.
221 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
222 identical IP addresses on different listening ports. Will also affect
225 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
226 add noisy ifdef guards to special-case this sillyness.
227 Patch from Bernd Kuhls.
229 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
230 than 255 are no longer allowed.
232 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
233 Disable the session-cache too, which might reduce our load. Since we
234 currrectly use a new context for every connection, both as server and
235 client, there is no benefit for these.
236 GnuTLS appears to not support tickets server-side by default (we don't
237 call gnutls_session_ticket_enable_server()) but client side is enabled
238 by default on recent versions (3.1.3 +) unless the PFS priority string
241 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
242 <https://reproducible-builds.org/specs/source-date-epoch/>.
244 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
245 the check for any unsuccessful recipients did not notice the limit, and
246 erroneously found still-pending ones.
248 JH/08 Pipeline CHUNKING command and data together, on kernels that support
249 MSG_MORE. Only in-clear (not on TLS connections).
251 JH/09 Avoid using a temporary file during transport using dkim. Unless a
252 transport-filter is involved we can buffer the headers in memory for
253 creating the signature, and read the spool data file once for the
254 signature and again for transmission.
256 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
257 4.77 as the kernel support then wasn't solid, having issues in 64bit
258 mode. Now, it's been long enough. Add support for FreeBSD also.
260 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
261 case where the routing stage had gathered several addresses to send to
262 a host before calling the transport for the first, we previously failed
263 to close down TLS in the old transport process before passing the TCP
264 connection to the new process. The new one sent a STARTTLS command
265 which naturally failed, giving a failed delivery and bloating the retry
266 database. Investigation and fix prototype from Wolfgang Breyha.
268 JH/12 Fix check on SMTP command input synchronisation. Previously there were
269 false-negatives in the check that the sender had not preempted a response
270 or prompt from Exim (running as a server), due to that code's lack of
271 awareness of the SMTP input buffering.
273 PP/04 Add commandline_checks_require_admin option.
274 Exim drops privileges sanely, various checks such as -be aren't a
275 security problem, as long as you trust local users with access to their
276 own account. When invoked by services which pass untrusted data to
277 Exim, this might be an issue. Set this option in main configuration
278 AND make fixes to the calling application, such as using `--` to stop
281 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
282 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
283 responses to those, into a single TLS record each way (this usually means
284 a single packet). As a side issue, smtp_enforce_sync now works on TLS
287 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
288 affects you only if you're dancing at the edge of the param size limits.
289 If you are, and this message makes sense to you, then: raise the
290 configured limit or use OpenSSL 1.1. Nothing we can do for older
293 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
294 cmdline element to get the documented default one. Previously it was
297 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
298 are unsuable for non-message-specific reasons - eg. connection timeout,
299 banner-time rejection.
301 JH/16 Fix logging of delivery remote port, when specified by router, under
304 PP/06 Repair manualroute's ability to take options in any order, even if one
305 is the name of a transport.
308 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
310 JH/17 Change the list-building routines interface to use the expanding-string
311 triplet model, for better allocation and copying behaviour.
313 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
314 Previously it was constructed the first time a possibly-matching string
315 was met in the configuration file input during startup; now it is done
318 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
319 compatible one, to avoid the (poorly documented) possibility of a config
320 file in the working directory redirecting the DB files, possibly correpting
321 some existing file. CVE-2017-10140 assigned for BDB.
323 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
324 cache-hot. Previously, although the result was properly cached, the
325 initial verify call returned a defer.
327 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
328 the main verify for receipient in uncached-mode.
330 JH/22 Retire historical build files to an "unsupported" subdir. These are
331 defined as "ones for which we have no current evidence of testing".
333 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
334 if present. Previously it was ignored.
336 JH/24 Start using specified-initialisers in C structure init coding. This is
337 a C99 feature (it's 2017, so now considered safe).
339 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
340 if was a fixed-sized field and bitmask ops via macros; it is now more
343 PP/07 GitHub PR 56: Apply MariaDB build fix.
344 Patch provided by Jaroslav Škarvada.
346 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
347 during Coverity cleanups [4.87 JH/47]
348 Diagnosis and fix provided by Michael Fischer v. Mollard.
350 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
351 the right size to place the terminating semicolon on its own folded
352 line, the header hash was calculated to an incorrect value thanks to
353 the (relaxed) space the fold became.
355 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
356 and confused the parent.
358 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
359 which could crash as a result. This could lead to undeliverable messages.
361 JH/28 Logging: "next input sent too soon" now shows where input was truncated
364 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
365 matters on fast-turnover and PID-randomising systems, which were getting
366 out-of-order delivery.
368 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
369 a possibly-overlapping copy. The symptom was that "Remote host closed
370 connection in response to HELO" was logged instead of the actual 4xx
373 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
374 Previously only that bufferd was discarded, resulting in SYMTP command
377 JH/32 DKIM: when a message has multiple signatures matching an identity given
378 in dkim_verify_signers, run the dkim acl once for each. Previously only
379 one run was done. Bug 2189.
381 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
382 "panic the current process" to "deliberately defer". The panic log is
383 still written with the problem list name; the mail and reject logs now
384 get a temp-reject line for the message that was being handled, saying
385 something like "domains check lookup or other defer". The SMTP 451
386 message is still "Temporary local problem".
388 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
389 A crafted sequence of BDAT commands could result in in-use memory beeing
390 freed. CVE-2017-16943.
392 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
393 from SMTP input. Previously it was always done; now only done for DATA
394 and not BDAT commands. CVE-2017-16944.
396 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
397 to the message (such as an overlong header line). Previously this was
398 not done and we did not exit BDAT mode. Followon from the previous item
399 though a different problem.
405 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
406 than -2003 did; needs libidn2 in addition to libidn.
408 JH/02 The path option on a pipe transport is now expanded before use.
410 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
411 Patch provided by "Björn", documentation fix added too.
413 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
414 missing a wire-to-host endian conversion.
416 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
417 close after a BDAT command line could be taken as a following command,
418 giving a synch failure. Fix by only checking for synch immediately
419 before acknowledging the chunk.
421 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
422 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
423 macro. Patches provided by Josh Soref.
425 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
426 Previously we did not; the RFC seems ambiguous and VRFY is not listed
427 by IANA as a service extension. However, John Klensin suggests that we
430 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
431 the dkim code may be unix-mode line endings rather than smtp wire-format
432 CRLF, so prepend a CR to any bare LF.
434 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
435 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
437 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
438 the first were themselves being wrongly included in the feed into dkim
439 processing; with most chunk sizes in use this resulted in an incorrect
440 body hash calculated value.
442 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
443 DKIM signature block, for verification. Although advised against by
444 standards it is specifically not ruled illegal.
446 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
448 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
449 missing a body hash (the bh= tag).
451 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
452 It seems that HAProxy sends the Proxy Protocol information in clear and
453 only then does a TLS startup, so do the same.
455 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
456 TCP connections (such as for Spamd) unless the daemon successfully set
457 Fast Open mode on its listening sockets. This fixes breakage seen on
458 too-old kernels or those not configured for Fast Open, at the cost of
459 requiring both directions being enabled for TFO, and TFO never being used
460 by non-daemon-related Exim processes.
462 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
463 endings, at least on the first header line. Try to canonify any that get
464 past that check, despite the cost.
466 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
467 now limited to an arbitrary five deep, while parsing addresses with the
468 strip_excess_angle_brackets option enabled.
470 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
471 instead leave the unprompted TLS handshake in socket buffer for the
472 TLS library to consume.
474 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
476 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
478 JH/16 Drop variables when they go out of scope. Memory management drops a whole
479 region in one operation, for speed, and this leaves assigned pointers
480 dangling. Add checks run only under the testsuite which checks all
481 variables at a store-reset and panics on a dangling pointer; add code
482 explicitly nulling out all the variables discovered. Fixes one known
483 bug: a transport crash, where a dangling pointer for $sending_ip_address
484 originally assigned in a verify callout, is re-used.
486 PP/06 Drop '.' from @INC in various Perl scripts.
488 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
490 PP/08 Reduce a number of compilation warnings under clang; building with
491 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
492 should be warning-free.
494 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
496 HS/01 Fix portability problems introduced by PP/08 for platforms where
497 realloc(NULL) is not equivalent to malloc() [SunOS et al].
499 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
500 chunk. This allows us to accept broken chunked messages. We need a more
501 general solution here.
503 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
504 already-broken messages in the queue.
506 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
508 JH/19 Fix reference counting bug in routing-generated-address tracking.
514 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
515 supports it and a size is available (ie. the sending peer gave us one).
517 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
518 years of being deprecated). The replacements are the ACLs
519 acl_smtp_mime and acl_not_smtp_mime.
521 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
522 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
523 or even an in-clear connection were permitted. Now, if the host lookup
524 was dnssec and dane was requested then the host is only used if the
525 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
526 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
527 if one fails this test.
528 This means that a poorly-configured remote DNS will make it incommunicado;
529 but it protects against a DNS-interception attack on it.
531 JH/04 Bug 1810: make continued-use of an open smtp transport connection
532 non-noisy when a race steals the message being considered.
534 JH/05 If main configuration option tls_certificate is unset, generate a
535 self-signed certificate for inbound TLS connections.
537 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
538 in rewrites and routers.
540 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
541 and logged a warning sing 4.83; now they are a configuration file error.
543 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
544 (lacking @domain). Apply the same qualification processing as RCPT.
546 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
548 JH/10 Support ${sha256:} applied to a string (as well as the previous
551 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
552 a cutthrough deliver is pending, as we always want to make a connection.
553 This also avoids re-routing the message when later placing the cutthrough
554 connection after a verify cache hit.
555 Do not update it with the verify result either.
557 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
558 when routing results in more than one destination address.
560 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
561 signing (which inhibits the cutthrough capability). Previously only
562 the presence of an option was tested; now an expansion evaluating as
563 empty is permissible (obviously it should depend only on data available
564 when the cutthrough connection is made).
566 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
567 the relevant preceding SMTP command did not note the pipelining mode.
569 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
570 Previously they were not counted.
572 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
573 as one having no matching records. Previously we deferred the message
574 that needed the lookup.
576 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
577 distinguished as "(=".
579 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
580 for missing MX records. Previously it only worked for missing A records.
582 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
584 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
585 after the data-go-ahead and data-ack. Patch from Jason Betts.
587 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
588 even for a "none" policy. Patch from Tony Meyer.
590 JH/22 Fix continued use of a connection for further deliveries. If a port was
591 specified by a router, it must also match for the delivery to be
594 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
595 When one of the recipients of a message was unsuitable for the connection
596 (has no matching addresses), we lost track of needing to mark it
597 deferred. As a result mail would be lost.
599 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
601 JH/25 Decoding ACL controls is now done using a binary search; the source code
602 takes up less space and should be simpler to maintain. Merge the ACL
603 condition decode tables also, with similar effect.
605 JH/26 Fix problem with one_time used on a redirect router which returned the
606 parent address unchanged. A retry would see the parent address marked as
607 delivered, so not attempt the (identical) child. As a result mail would
610 JH/27 Fix a possible security hole, wherein a process operating with the Exim
611 UID can gain a root shell. Credit to http://www.halfdog.net/ for
612 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
615 JH/28 Enable {spool,log} filesystem space and inode checks as default.
616 Main config options check_{log,spool}_{inodes,space} are now
617 100 inodes, 10MB unless set otherwise in the configuration.
619 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
620 Previously it only applied to the main-section connection policy
623 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
625 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
626 by me. Added RFC7919 DH primes as an alternative.
628 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
629 are not in the system include path.
631 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
632 GnuTLS, when a session startup failed (eg because the client disconnected)
633 Exim did stdio operations after fclose. This was exposed by a recent
634 change which nulled out the file handle after the fclose.
636 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
637 signed directly by the cert-signing cert, rather than an intermediate
638 OCSP-signing cert. This is the model used by LetsEncrypt.
640 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
642 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
643 an incoming connection.
645 HS/02 Bug 1802: Do not half-close the connection after sending a request
648 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
649 fallback to "prime256v1".
651 JH/34 SECURITY: Use proper copy of DATA command in error message.
652 Could leak key material. Remotely exploitable. CVE-2016-9963.
658 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
659 and 3.4.4 - once the server is enabled to respond to an OCSP request
660 it does even when not requested, resulting in a stapling non-aware
661 client dropping the TLS connection.
663 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
664 support variable-length bit vectors. No functional change.
666 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
667 The I= interface field on outgoing lines is now after the H= remote
668 host field, same as incoming lines. There is a separate
669 outgoing_interface log selector which allows you to disable the
672 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
673 If not running log_selector +smtp_connection the mainlog would be held
674 open indefinitely after a "too many connections" event, including to a
675 deleted file after a log rotate. Leave the per net connection logging
676 leaving it open for efficiency as that will be quickly detected by the
677 check on the next write.
679 HS/01 Bug 1671: Fix post transport crash.
680 Processing the wait-<transport> messages could crash the delivery
681 process if the message IDs didn't exist for some reason. When
682 using 'split_spool_directory=yes' the construction of the spool
683 file name failed already, exposing the same netto behaviour.
685 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
686 mime_regex ACL conditions.
688 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
689 to DSN fail messages (bounces): remote IP, remote greeting, remote response
690 to HELO, local diagnostic string.
692 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
693 log line to debug. Even when configured with a tls authenticator many
694 client connections are expected to not authenticate in this way, so
695 an authenticate fail is not an error.
697 HS/02 Add the Exim version string to the process info. This way exiwhat
698 gives some more detail about the running daemon.
700 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
701 matter for fast-change records such as DNSBLs.
703 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
704 retry record, even if constant. There may be multiple transports with
705 different interface settings and the retry behaviour needs to be kept
708 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
710 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
712 JH/10 Bug 840: fix log_defer_output option of pipe transport
714 JH/11 Bug 830: use same host for all RCPTS of a message, even under
715 hosts_randomize. This matters a lot when combined with mua_wrapper.
717 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
718 ${quote_pgsql:<string>} operator.
720 JH/13 Bug 1708: avoid misaligned access in cached lookup.
722 JH/14 Change header file name for freeradius-client. Relevant if compiling
723 with Radius support; from the Gentoo tree and checked under Fedora.
725 JH/15 Bug 1712: Introduce $prdr_requested flag variable
727 JH/16 Bug 1714: Permit an empty string as expansion result for transport
728 option transport_filter, meaning no filtering.
730 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
732 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
733 defaults to "*" (all hosts). The variable is now available when not built
734 with TLS, default unset, mainly to enable keeping the testsuite sane.
735 If a server certificate is not supplied (via tls_certificate) an error is
736 logged, and clients will find TLS connections fail on startup. Presumably
737 they will retry in-clear.
738 Packagers of Exim are strongly encouraged to create a server certificate
739 at installation time.
741 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
742 with the $config_file variable.
744 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
745 in transport context, after the attempt, and per-recipient. The latter type
746 is per host attempted. The event data is the error message, and the errno
747 information encodes the lookup type (A vs. MX) used for the (first) host,
748 and the trailing two digits of the smtp 4xx response.
750 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
751 to write to mainlog (or rejectlog, paniclog) in the window between file
752 creation and permissions/ownership being changed. Particularly affects
753 installations where exicyclog is run as root, rather than exim user;
754 result is that the running daemon panics and dies.
756 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
758 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
759 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
760 "pri" and "weight". Note that the previous implicit priority given by the
761 list order is no longer honoured.
763 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
766 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
767 by defining SUPPORT_SOCKS.
769 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
770 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
771 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
772 variables are renamed to proxy_{local,external}_{address,port}.
774 JH/27 Move Internationalisation support from Experimental to mainline, enabled
775 for a build by defining SUPPORT_I18N
777 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
778 of the query string, and make ${quote_redis:} do that quoting.
780 JH/29 Move Events support from Experimental to mainline, enabled by default
781 and removable for a build by defining DISABLE_EVENT.
783 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
785 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
786 cached by the daemon.
788 JH/32 Move Redis support from Experimental to mainline, enabled for a build
789 by defining LOOKUP_REDIS. The libhiredis library is required.
791 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
792 keys are given for lookup.
794 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
795 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
796 only supported when built with TLS support. The PolarSSL SHA routines
797 are still used when the TLS library is too old for convenient support.
799 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
800 openssl_options), for security. OpenSSL forces this from version 1.1.0
801 server-side so match that on older versions.
803 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
804 allocation for $value could be released as the expansion processing
805 concluded, but leaving the global pointer active for it.
807 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
808 and to use the domains and local_parts ACL conditions.
810 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
811 incorrectly not doubled on cutthrough transmission, hence seen as a
812 body-termination at the receiving system - resulting in truncated mails.
813 Commonly the sender saw a TCP-level error, and retransmitted the message
814 via the normal store-and-forward channel. This could result in duplicates
815 received - but deduplicating mailstores were liable to retain only the
816 initial truncated version.
818 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
820 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
822 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
823 we're in there, support oversigning also; bug 1309.
825 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
827 HS/04 Add support for keep_environment and add_environment options.
829 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
830 either intentional arithmetic overflow during PRNG, or testing config-
833 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
834 delivery resulted in actual delivery. Cancel cutthrough before DATA
837 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
838 rejects a recipient: pass the reject to the originator.
840 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
841 Many were false-positives and ignorable, but it's worth fixing the
844 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
845 for the new environment-manipulation done at startup. Move the routines
846 from being local to tls.c to being global via the os.c file.
848 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
849 an extract embedded as result-arg for a map, the first arg for extract
850 is unavailable so we cannot tell if this is a numbered or keyed
851 extraction. Accept either.
857 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
860 JH/02 The smtp transport option "multi_domain" is now expanded.
862 JH/03 The smtp transport now requests PRDR by default, if the server offers
865 JH/04 Certificate name checking on server certificates, when exim is a client,
866 is now done by default. The transport option tls_verify_cert_hostnames
867 can be used to disable this per-host. The build option
868 EXPERIMENTAL_CERTNAMES is withdrawn.
870 JH/05 The value of the tls_verify_certificates smtp transport and main options
871 default to the word "system" to access the system default CA bundle.
872 For GnuTLS, only version 3.0.20 or later.
874 JH/06 Verification of the server certificate for a TLS connection is now tried
875 (but not required) by default. The verification status is now logged by
876 default, for both outbound TLS and client-certificate supplying inbound
879 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
882 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
883 Status Notification (bounce) messages are now MIME format per RFC 3464.
884 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
885 under the control of the dsn_advertise_hosts option, and routers may
886 have a dsn_lasthop option.
888 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
889 default, modifiable by a malware= option. The list separator for
890 the options can now be changed in the usual way. Bug 68.
892 JH/10 The smtp_receive_timeout main option is now expanded before use.
894 JH/11 The incoming_interface log option now also enables logging of the
895 local interface on delivery outgoing connections.
897 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
898 if the interface and destination host and port all match.
900 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
903 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
904 Patch from Andrew Lewis.
906 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
907 now supports optional time-restrictions, weighting, and priority
908 modifiers per server. Patch originally by <rommer@active.by>.
910 JH/16 The spamd_address main option now supports a mixed list of local
911 and remote servers. Remote servers can be IPv6 addresses, and
912 specify a port-range.
914 JH/17 Bug 68: The spamd_address main option now supports an optional
915 timeout value per server.
917 JH/18 Bug 1581: Router and transport options headers_add/remove can
918 now have the list separator specified.
920 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
923 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
926 JH/21 Support for the A6 type of dns record is withdrawn.
928 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
929 rather than the verbs used.
931 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
932 from 255 to 1024 chars.
934 JH/24 Verification callouts now attempt to use TLS by default.
936 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
937 are generic router options now. The defaults didn't change.
939 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
940 Original patch from Alexander Shikoff, worked over by JH.
942 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
943 files if ZCAT_COMMAND is not executable.
945 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
947 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
949 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
950 Normally benign, it bites when the pair was led to by a CNAME;
951 modern usage is to not canonicalize the domain to a CNAME target
952 (and we were inconsistent anyway for A-only vs AAAA+A).
954 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
956 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
957 when evaluating $sender_host_dnssec.
959 JH/31 Check the HELO verification lookup for DNSSEC, adding new
960 $sender_helo_dnssec variable.
962 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
964 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
966 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
968 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
969 documented as working, but never had. Support all but $spam_report.
971 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
972 added for tls authenticator.
974 HS/03 Add perl_taintmode main config option
980 TL/01 When running the test suite, the README says that variables such as
981 no_msglog_check are global and can be placed anywhere in a specific
982 test's script, however it was observed that placement needed to be near
983 the beginning for it to behave that way. Changed the runtest perl
984 script to read through the entire script once to detect and set these
985 variables, reset to the beginning of the script, and then run through
986 the script parsing/test process like normal.
988 TL/02 The BSD's have an arc4random API. One of the functions to induce
989 adding randomness was arc4random_stir(), but it has been removed in
990 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
991 function when detected.
993 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
994 cause callback expansion.
996 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
997 syntax errors in an expansion can be treated as a string instead of
998 logging or causing an error, due to the internal use of bool_lax
999 instead of bool when processing it.
1001 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
1002 server certificates when making smtp deliveries.
1004 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
1006 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
1008 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
1010 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
1011 Merged patch from Sebastian Wiedenroth.
1013 JH/05 Fix results-pipe from transport process. Several recipients, combined
1014 with certificate use, exposed issues where response data items split
1015 over buffer boundaries were not parsed properly. This eventually
1016 resulted in duplicates being sent. This issue only became common enough
1017 to notice due to the introduction of connection certificate information,
1018 the item size being so much larger. Found and fixed by Wolfgang Breyha.
1020 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
1021 size buffer was used, resulting in syntax errors when an expansion
1024 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1025 version 3.3.6 or later.
1027 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1028 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1029 both become "event_action", the variables become $event_name, $event_data
1030 and $event_defer_errno. There is a new variable $verify_mode, usable in
1031 routers, transports and related events. The tls:cert event is now also
1032 raised for inbound connections, if the main configuration event_action
1035 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1036 early OCSP support, but no stapling (appears to be less than 1.0.0).
1038 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1039 server certificate names available under the smtp transport option
1040 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1043 JH/10 Time-related extraction expansions from certificates now use the main
1044 option "timezone" setting for output formatting, and are consistent
1045 between OpenSSL and GnuTLS compilations. Bug 1541.
1047 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1048 encoded parameter in the incoming message. Bug 1558.
1050 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1051 include certificate info, eximon was claiming there were spoolfile
1054 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1056 JH/14 Log delivery-related information more consistently, using the sequence
1057 "H=<name> [<ip>]" wherever possible.
1059 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1060 are problematic for Debian distribution, omit them from the release
1063 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1065 JH/16 Fix string representation of time values on 64bit time_t architectures.
1068 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1069 output list separator was used.
1074 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1075 checkers that were complaining about end of non-void function with no
1078 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1079 This was a regression introduced in 4.83 by another bugfix.
1081 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1083 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1084 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1090 TF/01 Correctly close the server side of TLS when forking for delivery.
1092 When a message was received over SMTP with TLS, Exim failed to clear up
1093 the incoming connection properly after forking off the child process to
1094 deliver the message. In some situations the subsequent outgoing
1095 delivery connection happened to have the same fd number as the incoming
1096 connection previously had. Exim would try to use TLS and fail, logging
1097 a "Bad file descriptor" error.
1099 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1100 utilities have not been installed.
1102 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1103 temporary space as the ACL may create new global variables.
1105 TL/01 LDAP support uses per connection or global context settings, depending
1106 upon the detected version of the libraries at build time.
1108 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1109 to extract and use the src ip:port in logging and expansions as if it
1110 were a direct connection from the outside internet. PPv2 support was
1111 updated based on HAProxy spec change in May 2014.
1113 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1115 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1116 Properly escape header and check for NULL return.
1118 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1121 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1123 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1124 characters in header names, implemented as a verify condition.
1125 Contributed by Michael Fischer v. Mollard.
1127 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1128 results permerror and temperror. Previous values are deprecated but
1129 still accepted. In a future release, err_perm and err_temp will be
1130 completely removed, which will be a backward incompatibility if the
1131 ACL tests for either of these two old results. Patch contributed by
1132 user bes-internal on the mailing list.
1134 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1136 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1137 selectors, in both main and reject logs.
1139 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1142 JH/07 Add malware type "sock" for talking to simple daemon.
1144 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1146 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1147 routers/transports under cutthrough routing.
1149 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1150 numbers. Touch up "bool" conditional to keep the same definition.
1152 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1154 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1156 JH/12 Expand items in router/transport headers_add or headers_remove lists
1157 individually rather than the list as a whole. Bug 1452.
1159 Required for reasonable handling of multiple headers_ options when
1160 they may be empty; requires that headers_remove items with embedded
1161 colons must have them doubled (or the list-separator changed).
1163 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1164 view the policy declared in the DMARC record. Currently, $dmarc_status
1165 is a combined value of both the record presence and the result of the
1168 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1170 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1171 dnslookup router and the smtp transport (applying to the forward
1174 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1175 of ldap servers used for a specific lookup. Patch provided by Heiko
1178 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1179 New variable $lookup_dnssec_authenticated for observability.
1181 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1182 Patch submitted by Lars Timman.
1184 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1186 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1187 Requires trusted mode and valid format message id, aborts otherwise.
1188 Patch contributed by Heiko Schlichting.
1190 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1191 certextract with support for various fields. Bug 1358.
1193 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1194 is requested by default, modifiable by smtp transport option
1197 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1198 operate on certificate variables to give certificate fingerprints
1199 Also new ${sha256:cert_variable}.
1201 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1203 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1204 Christian Aistleitner.
1206 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1208 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1209 file. Patch from Wolfgang Breyha.
1211 JH/25 Expand the coverage of the delivery $host and $host_address to
1212 client authenticators run in verify callout. Bug 1476.
1214 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1215 align with daemon_smtp_ports. Bug 72.
1217 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1218 support and error reporting did not work properly.
1220 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1221 and is readable. Patch from Andrew Colin Kissa.
1223 TL/14 Enhance documentation of ${run expansion and how it parses the
1224 commandline after expansion, particularly in the case when an
1225 unquoted variable expansion results in an empty value.
1227 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1229 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1230 by renaming away from it. Addresses GH issue 15, reported by
1233 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1235 TL/15 SECURITY: prevent double expansion in math comparison functions
1236 (can expand unsanitized data). Not remotely exploitable.
1243 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1245 PP/02 Make -n do something, by making it not do something.
1246 When combined with -bP, the name of an option is not output.
1248 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1251 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1252 $sender_host_name and config options to manage this, and basic check
1255 PP/05 DSCP support for outbound connections and control modifier for inbound.
1257 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1258 (Only plugin which currently uses this is kerberos4, which nobody should
1259 be using, but we should make it available and other future plugins might
1260 conceivably use it, even though it would break NAT; stuff *should* be
1261 using channel bindings instead).
1263 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1264 name; added for Sendmail compatibility; requires admin caller.
1265 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1266 just ignore it); requires trusted caller.
1267 Also parse but ignore: -Ac -Am -X<logfile>
1270 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1272 TL/02 Add +smtp_confirmation as a default logging option.
1274 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1275 Patch by Magnus Holmgren from 2007-02-20.
1277 TL/04 Bugzilla 1281 - Spec typo.
1278 Bugzilla 1283 - Spec typo.
1279 Bugzilla 1290 - Spec grammar fixes.
1281 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1283 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1285 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1286 mailing list by by Dmitry Isaikin.
1288 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1290 JH/02 Support "G" suffix to numbers in ${if comparisons.
1292 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1294 NM/01 Bugzilla 1197 - Spec typo
1295 Bugzilla 1196 - Spec examples corrections
1297 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1299 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1300 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1303 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1304 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1306 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1307 "acl {{name}{arg}...}", and optional args on acl condition
1310 JH/05 Permit multiple router/transport headers_add/remove lines.
1312 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1314 JH/07 Avoid using a waiting database for a single-message-only transport.
1315 Performance patch from Paul Fisher. Bugzilla 1262.
1317 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1320 JH/09 Add $headers_added variable, with content from use of ACL modifier
1321 add_header (but not yet added to the message). Bugzilla 199.
1323 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1324 Pulled from Bugzilla 817 by Wolfgang Breyha.
1326 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1328 (nb: this is the same fix as in Exim 4.80.1)
1330 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1333 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1336 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1338 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1339 advertise SMTP AUTH mechanism to us, instead of a generic
1340 protocol violation error. Also, make Exim more robust to bad
1341 data from the Dovecot auth socket.
1343 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1345 When a queue runner is handling a message, Exim first routes the
1346 recipient addresses, during which it prunes them based on the retry
1347 hints database. After that it attempts to deliver the message to
1348 any remaining recipients. It then updates the hints database using
1351 So if a recipient address works intermittently, it can get repeatedly
1352 deferred at routing time. The retry hints record remains fresh so the
1353 address never reaches the final cutoff time.
1355 This is a fairly common occurrence when a user is bumping up against
1356 their storage quota. Exim had some logic in its local delivery code
1357 to deal with this. However it did not apply to per-recipient defers
1358 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1360 This change adds a proper retry rule check during routing so that the
1361 final cutoff time is checked against the message's age. We only do
1362 this check if there is an address retry record and there is not a
1363 domain retry record; this implies that previous attempts to handle
1364 the address had the retry_use_local_parts option turned on. We use
1365 this as an approximation for the destination being like a local
1366 delivery, as in LMTP.
1368 I suspect this new check makes the old local delivery cutoff check
1369 redundant, but I have not verified this so I left the code in place.
1371 TF/02 Correct gecos expansion when From: is a prefix of the username.
1373 Test 0254 submits a message to Exim with the header
1377 When I ran the test suite under the user fanf2, Exim expanded
1378 the header to contain my full name, whereas it should have added
1379 a Resent-Sender: header. It erroneously treats any prefix of the
1380 username as equal to the username.
1382 This change corrects that bug.
1384 GF/01 DCC debug and logging tidyup
1385 Error conditions log to paniclog rather than rejectlog.
1386 Debug lines prefixed by "DCC: " to remove any ambiguity.
1388 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1390 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1391 Bug spotted by Jeremy Harris; was flawed since initial commit.
1392 Would have resulted in OCSP responses post-SNI triggering an Exim
1393 NULL dereference and crash.
1395 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1397 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1398 Bug detection, analysis and fix by Samuel Thibault.
1399 Bugzilla 1331, Debian bug #698092.
1401 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1403 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1404 Server implementation by Todd Lyons, client by JH.
1405 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1406 config variable "prdr_enable" controls whether the server
1407 advertises the facility. If the client requests PRDR a new
1408 acl_data_smtp_prdr ACL is called once for each recipient, after
1409 the body content is received and before the acl_smtp_data ACL.
1410 The client is controlled by both of: a hosts_try_prdr option
1411 on the smtp transport, and the server advertisement.
1412 Default client logging of deliveries and rejections involving
1413 PRDR are flagged with the string "PRDR".
1415 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1416 fclose(). Diagnosis by Todd Lyons.
1418 PP/17 Update configure.default to handle IPv6 localhost better.
1419 Patch by Alain Williams (plus minor tweaks).
1422 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1423 This is now consistent with GnuTLS, and is now documented: the
1424 previous undocumented portable approach to treating the option as
1425 unset was to force an expansion failure. That still works, and
1426 an empty string is now equivalent.
1428 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1429 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1430 not performing validation itself.
1432 PP/20 Added force_command boolean option to pipe transport.
1433 Patch from Nick Koston, of cPanel Inc.
1435 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1438 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1440 PP/21 Fix eximon continuous updating with timestamped log-files.
1441 Broken in a format-string cleanup in 4.80, missed when I repaired the
1442 other false fix of the same issue.
1443 Report and fix from Heiko Schlichting.
1446 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1447 Report from Prashanth Katuri.
1449 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1450 It's SecureTransport, so affects any MacOS clients which use the
1451 system-integrated TLS libraries, including email clients.
1453 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1454 using a MIME ACL for non-SMTP local injection.
1455 Report and assistance in diagnosis by Warren Baker.
1457 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1459 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1461 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1462 last id that failed so it may be referenced in subsequent ACL's.
1464 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1467 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1468 ldap library initialization, allowing self-signed CA's to be
1469 used. Also properly sets require_cert option later in code by
1470 using NULL (global ldap config) instead of ldap handle (per
1471 session). Bug diagnosis and testing by alxgomz.
1473 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1474 the src/util/ subdirectory.
1476 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1477 renamed to Transport Post Delivery Action by Jeremy Harris, as
1480 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1481 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1482 redis_servers = needs to be configured which will be used by the redis
1483 lookup. Patch from Warren Baker, of The Packet Hub.
1485 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1487 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1488 hostname or reverse DNS when processing a host list. Used suggestions
1489 from multiple comments on this bug.
1491 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1493 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1494 Missed a few lines, added it to make the runtest require no keyboard
1497 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1498 contains upper case chars. Make router use caseful_local_part.
1500 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1501 support when GnuTLS has been built with p11-kit.
1507 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1509 This, or similar/improved, will also be change PP/11 of 4.82.
1515 PP/01 Handle short writes when writing local log-files.
1516 In practice, only affects FreeBSD (8 onwards).
1517 Bugzilla 1053, with thanks to Dmitry Isaikin.
1519 NM/01 Bugzilla 949 - Documentation tweak
1521 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1524 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1526 PP/02 Implemented gsasl authenticator.
1528 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1530 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1531 `pkg-config foo` for cflags/libs.
1533 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1534 with rest of GSASL and with heimdal_gssapi.
1536 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1537 `pkg-config foo` for cflags/libs for the TLS implementation.
1539 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1540 properties get this fed in as external SSF. A number of robustness
1541 and debugging improvements to the cyrus_sasl authenticator.
1543 PP/08 cyrus_sasl server now expands the server_realm option.
1545 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1546 Patch by Jeremy Harris.
1548 PP/10 Added dbmjz lookup type.
1550 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1552 PP/12 MAIL args handles TAB as well as SP, for better interop with
1553 non-compliant senders.
1554 Analysis and variant patch by Todd Lyons.
1556 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1557 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1558 Patch from Dirk Mueller <dmueller@suse.com>
1560 PP/13 tls_peerdn now print-escaped for spool files.
1561 Observed some $tls_peerdn in wild which contained \n, which resulted
1562 in spool file corruption.
1564 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1565 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1566 or write after TLS renegotiation, which otherwise led to messages
1569 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1570 as a tracking header (ie: a signed header comes before the signature).
1571 Patch from Wolfgang Breyha.
1573 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1574 comma-sep list; embedded commas doubled.
1576 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1578 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1580 Report and patch from Dmitry Banschikov.
1582 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1583 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1584 failures. We appear to now support TLS1.1+ with Exim.
1586 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1587 lets Exim select keys and certificates based upon TLS SNI from client.
1588 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1589 before an outbound SMTP session. New log_selector, +tls_sni.
1591 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1592 NULL dereference. Report and patch from Alun Jones.
1594 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1595 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1596 Not seeing resolver debug output on NetBSD, but suspect this is a
1597 resolver implementation change.
1599 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1600 Left warnings. Added "eximon gdb" invocation mode.
1602 PP/21 Defaulting "accept_8bitmime" to true, not false.
1604 PP/22 Added -bw for inetd wait mode support.
1606 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1607 locate the relevant includes and libraries. Made this the default.
1609 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1610 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1612 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1613 This may cause build issues on older platforms.
1615 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1616 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1617 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1618 Added SNI support via GnuTLS too.
1619 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1621 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1623 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1624 Applied second patch from Janne, implementing suggestion to default
1625 multiple-strings-in-record handling to match SPF spec.
1627 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1629 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1630 read-only, out of scope).
1631 Patch from Wolfgang Breyha, report from Stuart Northfield.
1633 PP/29 Fix three issues highlighted by clang analyser static analysis.
1634 Only crash-plausible issue would require the Cambridge-specific
1635 iplookup router and a misconfiguration.
1636 Report from Marcin Mirosław.
1638 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1640 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1641 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1642 As part of this, removing so much warning spew let me fix some minor
1643 real issues in debug logging.
1645 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1646 assignment on my part. Fixed.
1648 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1649 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1650 Janne Snabb (who went above and beyond: thank you).
1652 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1653 string otherwise requires a connection and a bunch more work and it's
1654 relatively easy to get wrong. Should also expose TLS library linkage
1657 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1658 64-bit ${eval} (JH/03).
1660 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1661 GNU libc to support some of the 64-bit stuff, should not lead to
1662 conflicts. Defined before os.h is pulled in, so if a given platform
1663 needs to override this, it can.
1665 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1666 protection layer was required, which is not implemented.
1667 Bugzilla 1254, patch from Wolfgang Breyha.
1669 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1670 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1671 tls_dhparam take prime identifiers. Also unbreak combination of
1672 OpenSSL+DH_params+TLSSNI.
1674 PP/39 Disable SSLv2 by default in OpenSSL support.
1680 PP/01 Solaris build fix for Oracle's LDAP libraries.
1681 Bugzilla 1109, patch from Stephen Usher.
1683 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1685 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1688 TF/02 Fix a couple more cases where we did not log the error message
1689 when unlink() failed. See also change 4.74-TF/03.
1691 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1692 lock up or crash if it happened to be inside a call to libc when it
1693 got a SIGUSR1 from exiwhat.
1695 The SIGUSR1 handler appends the current process status to the process
1696 log which is later printed by exiwhat. It used to use the general
1697 purpose logging code to do this, but several functions it calls are
1698 not safe for signals.
1700 The new output code in the SIGUSR1 handler is specific to the process
1701 log, and simple enough that it's easy to inspect for signal safety.
1702 Removing some special cases also simplifies the general logging code.
1703 Removing the spurious timestamps from the process log simplifies
1706 TF/04 Improved ratelimit ACL condition.
1708 The /noupdate option has been deprecated in favour of /readonly which
1709 has clearer semantics. The /leaky, /strict, and /readonly update modes
1710 are mutually exclusive. The update mode is no longer included in the
1711 database key; it just determines when the database is updated. (This
1712 means that when you upgrade Exim will forget old rate measurements.)
1714 Exim now checks that the per_* options are used with an update mode that
1715 makes sense for the current ACL. For example, when Exim is processing a
1716 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1717 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1718 must specify per_mail/readonly. If you omit the update mode it defaults to
1719 /leaky where that makes sense (as before) or /readonly where required.
1721 The /noupdate option is now undocumented but still supported for
1722 backwards compatibility. It is equivalent to /readonly except that in
1723 ACLs where /readonly is required you may specify /leaky/noupdate or
1724 /strict/noupdate which are treated the same as /readonly.
1726 A useful new feature is the /count= option. This is a generalization
1727 of the per_byte option, so that you can measure the throughput of other
1728 aggregate values. For example, the per_byte option is now equivalent
1729 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1731 The per_rcpt option has been generalized using the /count= mechanism
1732 (though it's more complicated than the per_byte equivalence). When it is
1733 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1734 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1735 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1736 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1737 Note that using per_rcpt with a non-readonly update mode in more than
1738 one ACL will cause the recipients to be double-counted. (The per_mail
1739 and per_byte options don't have this problem.)
1741 The handling of very low rates has changed slightly. If the computed rate
1742 is less than the event's count (usually one) then this event is the first
1743 after a long gap. In this case the rate is set to the same as this event's
1744 count, so that the first message of a spam run is counted properly.
1746 The major new feature is a mechanism for counting the rate of unique
1747 events. The new per_addr option counts the number of different
1748 recipients that someone has sent messages to in the last time period. It
1749 behaves like per_rcpt if all the recipient addresses are different, but
1750 duplicate recipient addresses do not increase the measured rate. Like
1751 the /count= option this is a general mechanism, so the per_addr option
1752 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1753 example, measure the rate that a client uses different sender addresses
1754 with the options per_mail/unique=$sender_address. There are further
1755 details in the main documentation.
1757 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1759 TF/06 Removed a few PCRE remnants.
1761 TF/07 Automatically extract Exim's version number from tags in the git
1762 repository when doing development or release builds.
1764 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1765 Bugzilla 879. Patch from Paul Fisher.
1767 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1768 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1771 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1773 PP/05 Variable $av_failed, true if the AV scanner deferred.
1774 Bugzilla 1078. Patch from John Horne.
1776 PP/06 Stop make process more reliably on build failure.
1777 Bugzilla 1087. Patch from Heiko Schlittermann.
1779 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1780 Bugzilla 1089. Patch from Heiko Schlittermann.
1782 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1783 Bugzilla 1131. Patch from Holger Weiß.
1785 PP/09 Handle IPv6 addresses with SPF.
1786 Bugzilla 860. Patch from Wolfgang Breyha.
1788 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1790 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1793 PP/11 match_* no longer expand right-hand-side by default.
1794 New compile-time build option, EXPAND_LISTMATCH_RHS.
1795 New expansion conditions, "inlist", "inlisti".
1797 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1799 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1801 PP/14 fix log_write() format string regression from TF/03.
1802 Bugzilla 1152. Patch from Dmitry Isaikin.
1808 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1810 PP/02 Harmonised TLS library version reporting; only show if debugging.
1811 Layout now matches that introduced for other libraries in 4.74 PP/03.
1813 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1815 PP/04 New "dns_use_edns0" global option.
1817 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1820 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1821 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1823 TK/01 Updated PolarSSL code to 0.14.2.
1824 Bugzilla 1097. Patch from Andreas Metzler.
1826 PP/07 Catch divide-by-zero in ${eval:...}.
1827 Fixes bugzilla 1102.
1829 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1832 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1833 format-string attack -- SECURITY: remote arbitrary code execution.
1835 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1836 time unintentionally subject to list matching rules, letting the header
1837 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1838 arbitrary string expansion). This allowed for information disclosure.
1840 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1841 INT_MIN/-1 -- value coerced to INT_MAX.
1847 NM/01 Workaround for PCRE version dependency in version reporting
1850 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1851 This fixes portability to compilers other than gcc, notably
1852 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1854 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1855 makefiles for portability to HP-UX and POSIX correctness.
1857 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1858 Also via indented variable definition in the Makefile.
1859 (Debugging by Oliver Heesakkers).
1861 PP/02 Restore caching of spamd results with expanded spamd_address.
1862 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1864 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1865 Improves build reliability. Fix from: Frank Elsner
1867 NM/02 Fix wide character breakage in the rfc2047 coding
1868 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1870 NM/03 Allow underscore in dnslist lookups
1871 Fixes bug 1026. Patch from Graeme Fowler
1873 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1874 Code patches from Adam Ciarcinski of NetBSD.
1876 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1879 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1880 is logged, to avoid truncation. Patch from John Horne.
1882 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1883 Patch from Jakob Hirsch.
1885 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1886 SQL string expansion failure details.
1887 Patch from Andrey Oktyabrski.
1889 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1890 Patch from Simon Arlott.
1892 PP/09 New lookups functionality failed to compile on old gcc which rejects
1893 extern declarations in function scope.
1894 Patch from Oliver Fleischmann
1896 PP/10 Use sig_atomic_t for flags set from signal handlers.
1897 Check getgroups() return and improve debugging.
1898 Fixed developed for diagnosis in bug 927 (which turned out to be
1901 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1902 Patch from Mark Zealey.
1904 PP/12 Bugzilla 1056: Improved spamd server selection.
1905 Patch from Mark Zealey.
1907 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1908 Based on patch from Heiko Schlittermann.
1910 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1911 Patch from Uwe Doering, sign-off by Michael Haardt.
1913 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1914 variable declaration deep within a block. Bug and patch from
1917 PP/15 lookups-Makefile IRIX compatibility coercion.
1919 PP/16 Make DISABLE_DKIM build knob functional.
1921 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1922 Patch by Simon Arlott
1924 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1925 variable argument macros. Our copy now differs from upstream.
1931 TF/01 Failure to get a lock on a hints database can have serious
1932 consequences so log it to the panic log.
1934 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1935 controlled using the smtp_confirmation log selector.
1937 TF/03 Include the error message when we fail to unlink a spool file.
1939 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1940 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1941 for maintaining out-of-tree patches for some time.
1943 PP/01 Bugzilla 139: Documentation and portability issues.
1944 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1945 Handle per-OS dynamic-module compilation flags.
1947 PP/02 Let /dev/null have normal permissions.
1948 The 4.73 fixes were a little too stringent and complained about the
1949 permissions on /dev/null. Exempt it from some checks.
1950 Reported by Andreas M. Kirchwitz.
1952 PP/03 Report version information for many libraries, including
1953 Exim version information for dynamically loaded libraries. Created
1954 version.h, now support a version extension string for distributors
1955 who patch heavily. Dynamic module ABI change.
1957 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1958 privilege escalation vulnerability whereby the Exim run-time user
1959 can cause root to append content of the attacker's choosing to
1962 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1965 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1966 If dropping privileges for untrusted macros, we disabled normal logging
1967 on the basis that it would fail; for the Exim run-time user, this is not
1968 the case, and it resulted in successful deliveries going unlogged.
1969 Fixed. Reported by Andreas Metzler.
1975 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1976 only prepend for the Resent-* case. Fixes regression introduced in
1977 Exim 4.70 by NM/22 for Bugzilla 607.
1979 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1980 increasing numbers of administrators be bitten by this.
1982 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1984 PP/03 Bugzilla 994: added openssl_options main configuration option.
1986 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1988 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1990 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1992 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1993 without a peer certificate, leading to a segfault because of an
1994 assumption that peers always have certificates. Be a little more
1995 paranoid. Problem reported by Martin Tscholak.
1997 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1998 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1999 NB: ClamAV planning to remove STREAM in "middle of 2010".
2000 CL also introduces -bmalware, various -d+acl logging additions and
2001 more caution in buffer sizes.
2003 PP/09 Implemented reverse_ip expansion operator.
2005 PP/10 Bugzilla 937: provide a "debug" ACL control.
2007 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
2009 PP/12 Bugzilla 973: Implement --version.
2011 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
2013 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
2015 PP/15 Bugzilla 816: support multiple condition rules on Routers.
2017 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
2018 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
2019 ignore trailing whitespace.
2021 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2023 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2026 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2027 Notification from Dr Andrew Aitchison.
2029 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2030 ExtendedDetectionInfo response format.
2031 Notification from John Horne.
2033 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2036 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2037 XSL and documented dependency on system catalogs, with examples of how
2040 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2043 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2044 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2045 configuration file which is writeable by the Exim user or group.
2047 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2048 of configuration files to cover files specified with the -C option if
2049 they are going to be used with root privileges, not just the default
2052 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2053 option (effectively making it always true).
2055 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2056 files to be used while preserving root privileges.
2058 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2059 that rogue child processes cannot use them.
2061 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2062 run-time user, instead of root.
2064 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2065 Exim run-time user without dropping privileges.
2067 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2068 result string, instead of calling string_vformat() twice with the same
2071 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2072 for other users. Others should always drop root privileges if they use
2073 -C on the command line, even for a whitelisted configure file.
2075 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2077 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2083 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2084 $data_path, and $header_path variables; fixed documentation bugs and
2087 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2088 exipick to access non-standard spools, including the "frozen" queue
2091 NM/01 Bugzilla 965: Support mysql stored procedures.
2092 Patch from Alain Williams
2094 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2096 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2097 Patch from Andreas Metzler
2099 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2100 Patch from Kirill Miazine
2102 NM/05 Bugzilla 671: Added umask to procmail example.
2104 JJ/03 installed exipick 20100323.0, fixing doc bug
2106 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2107 directory. Notification and patch from Dan Rosenberg.
2109 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2111 TK/02 Improve log output when DKIM signing operation fails.
2113 MH/01 Treat the transport option dkim_domain as a colon separated
2114 list, not as a single string, and sign the message with each element,
2115 omitting multiple occurences of the same signer.
2117 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2118 Bugzilla 985, 986. Patch by Simon Arlott
2120 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2121 Patch by Simon Arlott
2123 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2124 MBX locking. Notification from Dan Rosenberg.
2130 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2132 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2134 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2136 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2138 NM/04 Bugzilla 915: Flag broken perl installation during build.
2144 TK/01 Added patch by Johannes Berg that expands the main option
2145 "spamd_address" if it starts with a dollar sign.
2147 TK/02 Write list of recipients to X-Envelope-Sender header when building
2148 the mbox-format spool file for content scanning (suggested by Jakob
2151 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2152 (http://www.dcc-servers.net/) support via dccifd. Activated by
2153 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2155 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2156 by Mark Daniel Reidel <mr@df.eu>.
2158 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2159 When building exim an external PCRE library is now needed -
2160 PCRE is a system library on the majority of modern systems.
2161 See entry on PCRE_LIBS in EDITME file.
2163 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2164 conversation. Added nologin parameter to request.
2165 Patch contributed by Kirill Miazine.
2167 TF/01 Do not log submission mode rewrites if they do not change the address.
2169 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2171 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2172 log files in place. Contributed by Roberto Lima.
2174 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2176 TF/03 Bugzilla 615: When checking the local_parts router precondition
2177 after a local_part_suffix or local_part_prefix option, Exim now
2178 does not use the address's named list lookup cache, since this
2179 contains cached lookups for the whole local part.
2181 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2182 Robert Millan. Documentation is in experimental-spec.txt.
2184 TF/04 Bugzilla 668: Fix parallel build (make -j).
2186 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2188 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2189 Patch provided by Jan Srzednicki.
2191 TF/05 Leading white space used to be stripped from $spam_report which
2192 wrecked the formatting. Now it is preserved.
2194 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2195 that they are available at delivery time.
2197 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2199 TF/08 TLS error reporting now respects the incoming_interface and
2200 incoming_port log selectors.
2202 TF/09 Produce a more useful error message if an SMTP transport's hosts
2203 setting expands to an empty string.
2205 NM/06 Bugzilla 744: EXPN did not work under TLS.
2206 Patch provided by Phil Pennock.
2208 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2209 Patch provided by Richard Godbee.
2211 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2212 acl_smtp_notquit, added index entry.
2214 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2215 Patch provided by Eugene Bujak.
2217 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2218 accept(). Patch provided by Maxim Dounin.
2220 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2221 Patch provided by Phil Pennock.
2223 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2225 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2226 Patch provided by Brad "anomie" Jorsch.
2228 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2229 Patch provided by Dean Brooks.
2231 TK/05 Add native DKIM support (does not depend on external libraries).
2233 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2234 Patch provided by Graeme Fowler.
2236 NM/16 Bugzilla 851: Documentation example syntax fix.
2238 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2240 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2243 NM/19 Bugzilla 745: TLS version reporting.
2244 Patch provided by Phil Pennock.
2246 NM/20 Bugzilla 167: bool: condition support.
2247 Patch provided by Phil Pennock.
2249 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2250 clients. Patch provided by Phil Pennock.
2252 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2253 Patch provided by Brad "anomie" Jorsch.
2255 NM/23 Bugzilla 687: Fix misparses in eximstats.
2256 Patch provided by Heiko Schlittermann.
2258 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2259 Patch provided by Heiko Schlittermann.
2261 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2262 plus update to original patch.
2264 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2266 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2267 Patch provided by David Brownlee.
2269 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2271 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2273 NM/30 Bugzilla 888: TLS documentation bugfixes.
2275 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2277 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2278 Unlike the original bugzilla I have changed all shell scripts in src tree.
2280 NM/33 Bugzilla 898: Transport filter timeout fix.
2281 Patch by Todd Rinaldo.
2283 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2284 Patch by Serge Demonchaux.
2286 NM/35 Bugzilla 39: Base64 decode bug fixes.
2287 Patch by Jakob Hirsch.
2289 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2291 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2293 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2295 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2301 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2302 ALT-N's libdkim that I have put here:
2303 http://duncanthrax.net/exim-experimental/
2305 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2306 were called 'true' and it seems that C99 defines that as a reserved
2307 keyword to be used with 'bool' variable types. That means you could
2308 not include C99-style headers which use bools without triggering
2309 build errors in sieve.c.
2311 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2312 as mailq or other aliases. Changed the --help handling significantly
2313 to do whats expected. exim_usage() emits usage/help information.
2315 SC/01 Added the -bylocaldomain option to eximstats.
2317 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2319 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2321 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2327 PH/01 Another patch from the Sieve maintainer.
2329 PH/02 When an IPv6 address is converted to a string for single-key lookup
2330 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2331 dots are used instead of colons so that keys in lsearch files need not
2332 contain colons. This was done some time before quoting was made available
2333 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2334 (notated using the quote facility) so as to distinguish them from IPv4
2335 keys. This meant that lookups for IP addresses in host lists did not work
2336 for iplsearch lookups.
2338 This has been fixed by arranging for IPv6 addresses to be expressed with
2339 colons if the lookup type is iplsearch. This is not incompatible, because
2340 previously such lookups could never work.
2342 The situation is now rather anomalous, since one *can* have colons in
2343 ordinary lsearch keys. However, making the change in all cases is
2344 incompatible and would probably break a number of configurations.
2346 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2349 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2350 conversion specification without a maximum field width, thereby enabling
2351 a rogue spamd server to cause a buffer overflow. While nobody in their
2352 right mind would setup Exim to query an untrusted spamd server, an
2353 attacker that gains access to a server running spamd could potentially
2354 exploit this vulnerability to run arbitrary code as the Exim user.
2356 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2357 $primary_hostname instead of what libspf2 thinks the hosts name is.
2359 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2360 a directory entry by the name of the lookup key. Previously, if a
2361 symlink pointed to a non-existing file or a file in a directory that
2362 Exim lacked permissions to read, a lookup for a key matching that
2363 symlink would fail. Now it is enough that a matching directory entry
2364 exists, symlink or not. (Bugzilla 503.)
2366 PH/03 The body_linecount and body_zerocount variables are now exported in the
2369 PH/04 Added the $dnslist_matched variable.
2371 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2372 This means they are set thereafter only if the connection becomes
2375 PH/06 Added the client_condition to authenticators so that some can be skipped
2376 by clients under certain conditions.
2378 PH/07 The error message for a badly-placed control=no_multiline_responses left
2379 "_responses" off the end of the name.
2381 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2383 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2384 (without spaces) instead of just copying the configuration text.
2386 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2388 PH/11 Added $max_received_linelength.
2390 PH/12 Added +ignore_defer and +include_defer to host lists.
2392 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2393 way in which PCRE > 7.0 is built.
2395 PH/14 Implemented queue_only_load_latch.
2397 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2398 MAIL command. The effect was to mangle the value on 64-bit systems.
2400 PH/16 Another patch from the Sieve maintainer.
2402 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2404 PH/18 If a system quota error occurred while trying to create the file for
2405 a maildir delivery, the message "Mailbox is full" was not appended to the
2406 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2407 only to a quota excession during the actual writing of the file.
2409 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2410 characters?) which causes problems in log lines. The DN values are now
2411 passed through string_printing() before being added to log lines.
2413 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2414 and InterBase are left for another time.)
2416 PH/21 Added message_body_newlines option.
2418 PH/22 Guard against possible overflow in moan_check_errorcopy().
2420 PH/23 POSIX allows open() to be a macro; guard against that.
2422 PH/24 If the recipient of an error message contained an @ in the local part
2423 (suitably quoted, of course), incorrect values were put in $domain and
2424 $local_part during the evaluation of errors_copy.
2430 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2431 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2434 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2435 issue a MAIL command.
2437 PH/02 In an ACL statement such as
2439 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2441 if a client was not listed at all, or was listed with a value other than
2442 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2443 the condition was not true (as it should be), so access was not denied.
2444 The bug was that the ! inversion was incorrectly passed on to the second
2445 item. This has been fixed.
2447 PH/03 Added additional dnslists conditions == and =& which are different from
2448 = and & when the dns lookup returns more than one IP address.
2450 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2451 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2453 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2454 FSYNC, which compiles an option called disable_fsync that allows for
2455 bypassing fsync(). The documentation is heavily laced with warnings.
2457 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2459 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2460 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2461 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2462 including adding "make clean"; (3) Added -fPIC when compiling the test
2463 dynamically loaded module, to get rid of a warning.
2465 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2466 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2467 = 0s. The bug is otherwise harmless.
2469 PH/07 There was a bug in the dovecot authenticator such that the value of
2470 $auth1 could be overwritten, and so not correctly preserved, after a
2471 successful authentication. This usually meant that the value preserved by
2472 the server_setid option was incorrect.
2474 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2476 PH/09 Installed PCRE release 7.0.
2478 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2479 run for batched SMTP input. It is now run at the start of every message
2480 in the batch. While fixing this I discovered that the process information
2481 (output by running exiwhat) was not always getting set for -bs and -bS
2482 input. This is fixed, and it now also says "batched" for BSMTP.
2484 PH/11 Added control=no_pipelining.
2486 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2487 patch, slightly modified), and move the expansion of helo_data till after
2488 the connection is made in the smtp transport (so it can use these
2491 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2493 PH/14 Added log_selector = +pid.
2495 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2497 PH/16 Add ${if forany and ${if forall.
2499 PH/17 Added dsn_from option to vary the From: line in DSNs.
2501 PH/18 Flush SMTP output before performing a callout, unless control =
2502 no_callout_flush is set.
2504 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2505 was true (the default) a successful delivery failed to delete the retry
2506 item, thus causing premature timeout of the address. The bug is now
2509 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2511 PH/21 Long custom messages for fakedefer and fakereject are now split up
2512 into multiline responses in the same way that messages for "deny" and
2513 other ACL rejections are.
2515 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2516 with slight modification.
2518 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2519 draft, changing the syntax and factoring some duplicate code".
2521 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2522 for deliveries of the second and subsequent messages over the same SMTP
2525 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2526 ${reduce, with only minor "tidies".
2528 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2530 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2531 expansion side effects.
2533 PH/27 When a message times out after an over-quota error from an Exim-imposed
2534 quota, the bounce message says "mailbox is full". This message was not
2535 being given when it was a system quota that was exceeded. It now should
2538 MH/03 Made $recipients available in local_scan(). local_scan() already has
2539 better access to the recipient list through recipients_list[], but
2540 $recipients can be useful in postmaster-provided expansion strings.
2542 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2543 in the case of a MAIL command with additional options following the
2544 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2545 were accidentally chopped off.
2547 PH/29 SMTP synchronization checks are implemented when a command is read -
2548 there is a check that no more input is waiting when there shouldn't be
2549 any. However, for some commands, a delay in an ACL can mean that it is
2550 some time before the response is written. In this time, more input might
2551 arrive, invalidly. So now there are extra checks after an ACL has run for
2552 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2553 pipelining has not been advertised.
2555 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2557 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2558 returned in the SMTP response when smtp_return_error_details was false.
2559 This has been fixed.
2561 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2562 instead of the C I/O that was originally supplied, because problems were
2563 reported on Solaris.
2565 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2566 Exim which did not show up earlier: it was assuming that a call to
2567 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2568 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2569 was a macro that became an assignment, so it seemed to work. This has
2570 changed to a proper function call with a void return, hence the compile
2571 error. Exim's code has been fixed.
2573 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2576 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2578 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2580 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2581 because it turns out to be too big or there is a timeout) there is no
2582 "Completed" line in the log. When some messages of this type were
2583 selected by exigrep, they were listed as "not completed". Others were
2584 picked up by some special patterns. I have improved the selection
2585 criteria to be more general.
2587 PH/38 The host_find_failed option in the manualroute router can now be set
2588 to "ignore", to completely ignore a host whose IP address cannot be
2589 found. If all hosts are ignored, the behaviour is controlled by the new
2590 host_all_ignored option.
2592 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2593 homing or because of multiple MX records with /mx) generated more than
2594 one IP address, and the following item turned out to be the local host,
2595 all the secondary addresses of the first item were incorrectly removed
2596 from the list, along with the local host and any following hosts (which
2597 is what is supposed to happen).
2599 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2600 whoever called Exim into the -H file. In the case of the daemon it was
2601 behaving confusingly. When first started, it used values for whoever
2602 started the daemon, but after a SIGHUP it used the Exim user (because it
2603 calls itself on a restart). I have changed the code so that it now always
2606 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2607 message are rejected with the same error (e.g. no authentication or bad
2608 sender address), and a DATA command is nevertheless sent (as can happen
2609 with PIPELINING or a stupid MUA), the error message that was given to the
2610 RCPT commands is included in the rejection of the DATA command. This is
2611 intended to be helpful for MUAs that show only the final error to their
2614 PH/42 Another patch from the Sieve maintainer.
2616 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2617 Eximstats - Fixed some broken HTML links and added missing column headers
2619 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2620 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2622 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2624 PH/43 Yet another patch from the Sieve maintainer.
2626 PH/44 I found a way to check for a TCP/IP connection going away before sending
2627 the response to the final '.' that terminates a message, but only in the
2628 case where the client has not sent further data following the '.'
2629 (unfortunately, this is allowed). However, in many cases there won't be
2630 any further data because there won't be any more messages to send. A call
2631 to select() can be used: if it shows that the input is "ready", there is
2632 either input waiting, or the socket has been closed. An attempt to read
2633 the next input character can distinguish the two cases. Previously, Exim
2634 would have sent an OK response which the client would never have see.
2635 This could lead to message repetition. This fix should cure that, at
2636 least in a lot of common cases.
2638 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2639 advertised in response to EHLO.
2645 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2646 fixed by 4.65/MH/01 (is this a record?) are fixed:
2648 (i) An empty string was always treated as zero by the numeric comparison
2649 operators. This behaviour has been restored.
2651 (ii) It is documented that the numeric comparison operators always treat
2652 their arguments as decimal numbers. This was broken in that numbers
2653 starting with 0 were being interpreted as octal.
2655 While fixing these problems I realized that there was another issue that
2656 hadn't been noticed. Values of message_size_limit (both the global option
2657 and the transport option) were treated as octal if they started with 0.
2658 The documentation was vague. These values are now always treated as
2659 decimal, and I will make that clear in the documentation.
2665 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2666 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2669 MH/01 Don't check that the operands of numeric comparison operators are
2670 integers when their expansion is in "skipping" mode (fixes bug
2671 introduced by 4.64-PH/07).
2673 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2674 child addresses, Exim now panics and dies. Previously, because the count
2675 is held in a short int, deliveries were likely to be lost. As such a
2676 large number of recipients for a single message is ridiculous
2677 (performance will be very, very poor), I have chosen to impose a limit
2678 rather than extend the field.
2684 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2685 leftover -K file (the existence of which was triggered by #402).
2686 While we were at it, introduced process PID as part of the -K
2687 filename. This should rule out race conditions when creating
2690 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2691 processing considerably. Previous code took too long for large mails,
2692 triggering a timeout which in turn triggers #401.
2694 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2695 in the DK code in transports.c. sendfile() is not really portable,
2696 hence the _LINUX specificness.
2698 TF/01 In the add_headers option to the mail command in an Exim filter,
2699 there was a bug that Exim would claim a syntax error in any
2700 header after the first one which had an odd number of characters
2703 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2704 callout verification, Exim cached a "reject" for the entire domain. This
2705 is correct for most verifications, but it is not correct for a recipient
2706 verification with use_sender or use_postmaster set, because in that case
2707 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2708 case of MAIL FROM:<> rejection from other early rejections (e.g.
2709 rejection of HELO). When verifying a recipient using a non-null MAIL
2710 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2711 Whatever the result of the callout, the value of the domain cache is
2712 left unchanged (for any other kind of callout, getting as far as trying
2713 RCPT means that the domain itself is ok).
2715 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2718 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2719 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2720 session, the daemon ignores ECONNECT errors and logs others; it now
2721 ignores EPIPE as well.
2723 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2724 (quoted-printable decoding).
2726 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2727 later the small subsequent patch to fix an introduced bug.
2729 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2731 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2733 PH/08 An error is now given if message_size_limit is specified negative.
2735 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2736 to be given (somewhat) arbitrary names.
2738 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2741 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2742 miscellaneous code fixes
2744 PH/10 Added the log_reject_target ACL modifier to specify where to log
2747 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2748 hostname. This is wrong, because it relates to the incoming message (and
2749 probably the interface on which it is arriving) and not to the outgoing
2750 callout (which could be using a different interface). This has been
2751 changed to use the value of the helo_data option from the smtp transport
2752 instead - this is what is used when a message is actually being sent. If
2753 there is no remote transport (possible with a router that sets up host
2754 addresses), $smtp_active_hostname is used.
2756 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2757 tweaks were necessary in order to get it to work (see also 21 below):
2758 (a) The code assumed that strncpy() returns a negative number on buffer
2759 overflow, which isn't the case. Replaced with Exim's string_format()
2761 (b) There were several signed/unsigned issues. I just did the minimum
2762 hacking in of casts. There is scope for a larger refactoring.
2763 (c) The code used strcasecmp() which is not a standard C function.
2764 Replaced with Exim's strcmpic() function.
2765 (d) The code set only $1; it now sets $auth1 as well.
2766 (e) A simple test gave the error "authentication client didn't specify
2767 service in request". It would seem that Dovecot has changed its
2768 interface. Fortunately there's a specification; I followed it and
2769 changed what the client sends and it appears to be working now.
2771 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2774 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2775 address A is aliased to B and C, where B exists and C does not. Without
2776 -v the output is "A verified" because verification stops after a
2777 successful redirection if more than one address is generated. However,
2778 with -v the child addresses are also verified. Exim was outputting "A
2779 failed to verify" and then showing the successful verification for C,
2780 with its parentage. It now outputs "B failed to verify", showing B's
2781 parentage before showing the successful verification of C.
2783 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2784 look up a TXT record in a specific list after matching in a combined
2787 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2788 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2789 they consult the DNS. I had assumed they would set it the way they
2790 wanted; and indeed my experiments on Linux seem to show that in some
2791 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2792 To be on the safe side, however, I have now made the interface to
2793 host_find_byname() similar to host_find_bydns(), with an argument
2794 containing the DNS resolver options. The host_find_byname() function now
2795 sets these options at its start, just as host_find_bydns() does. The smtp
2796 transport options dns_qualify_single and dns_search_parents are passed to
2797 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2798 of host_find_byname() use the default settings of RES_DEFNAMES
2799 (qualify_single) but not RES_DNSRCH (search_parents).
2801 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2802 spool_read_header() do less string testing, by means of a preliminary
2803 switch on the second character of optional "-foo" lines. (This is
2804 overdue, caused by the large number of possibilities that now exist.
2805 Originally there were few.) While I was there, I also converted the
2806 str(n)cmp tests so they don't re-test the leading "-" and the first
2807 character, in the hope this might squeeze out yet more improvement.
2809 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2810 flag allowing group syntax was set by the header_syntax check but not
2811 turned off, possible causing trouble later; (2) The flag was not being
2812 set at all for the header_verify test, causing "group"-style headers to
2813 be rejected. I have now set it in this case, and also caused header_
2814 verify to ignore an empty address taken from a group. While doing this, I
2815 came across some other cases where the code for allowing group syntax
2816 while scanning a header line wasn't quite right (mostly, not resetting
2817 the flag correctly in the right place). These bugs could have caused
2818 trouble for malformed header lines. I hope it is now all correct.
2820 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2821 with the "reply" argument non-NULL. The code, however (which originally
2822 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2823 but it didn't always do it. This confused somebody who was copying the
2824 code for some other use. I have removed all the tests.
2826 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2827 feature that was used to support insecure browsers during the U.S. crypto
2828 embargo. It requires special client support, and Exim is probably the
2829 only MTA that supported it -- and would never use it because real RSA is
2830 always available. This code has been removed, because it had the bad
2831 effect of slowing Exim down by computing (never used) parameters for the
2832 RSA_EXPORT functionality.
2834 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2835 authenticator to fail if there's a tab character in the incoming data
2836 (there should never be unless someone is messing about, as it's supposed
2837 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2838 if the connection is using TLS or if the remote IP is the same as the
2839 local IP, and the "valid-client-cert option" if a client certificate has
2842 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2843 authenticators. This can be used for authorization after authentication
2844 succeeds. (In the case of plaintext, it servers for both authentication
2847 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2848 if any retry times were supplied.
2850 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2851 connection, where there is no client IP address to check. In this
2852 situation, the verify now always succeeds.
2854 PH/25 Applied John Jetmore's -Mset patch.
2856 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2858 PH/27 In a string expansion for a processed (not raw) header when multiple
2859 headers of the same name were present, leading whitespace was being
2860 removed from all of them, but trailing whitespace was being removed only
2861 from the last one. Now trailing whitespace is removed from each header
2862 before concatenation. Completely empty headers in a concatenation (as
2863 before) are ignored.
2865 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2866 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2868 PH/29 [Removed. This was a change that I later backed out, and forgot to
2869 correct the ChangeLog entry (that I had efficiently created) before
2870 committing the later change.]
2872 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2873 address errors (4xx response to RCPT) over the same connection as other
2874 messages routed to the same hosts. Such deliveries are always "forced",
2875 so retry times are not inspected. This resulted in far too many retries
2876 for the affected addresses. The effect occurred only when there were more
2877 hosts than the hosts_max_try setting in the smtp transport when it had
2878 the 4xx errors. Those hosts that it had tried were not added to the list
2879 of hosts for which the message was waiting, so if all were tried, there
2880 was no problem. Two fixes have been applied:
2882 (i) If there are any address or message errors in an SMTP delivery, none
2883 of the hosts (tried or untried) are now added to the list of hosts
2884 for which the message is waiting, so the message should not be a
2885 candidate for sending over the same connection that was used for a
2886 successful delivery of some other message. This seems entirely
2887 reasonable: after all the message is NOT "waiting for some host".
2888 This is so "obvious" that I'm not sure why it wasn't done
2889 previously. Hope I haven't missed anything, but it can't do any
2890 harm, as the worst effect is to miss an optimization.
2892 (ii) If, despite (i), such a delivery is accidentally attempted, the
2893 routing retry time is respected, so at least it doesn't keep
2894 hammering the server.
2896 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2897 in ${readsocket because some servers need this prod.
2899 PH/32 Added some extra debug output when updating a wait-xxx database.
2901 PH/33 The hint "could be header name not terminated by colon", which has been
2902 given for certain expansion errors for a long time, was not being given
2903 for the ${if def:h_colon_omitted{... case.
2905 PH/34 The spec says: "With one important exception, whenever a domain list is
2906 being scanned, $domain contains the subject domain." There was at least
2907 one case where this was not true.
2909 PH/35 The error "getsockname() failed: connection reset by peer" was being
2910 written to the panic log as well as the main log, but it isn't really
2911 panic-worthy as it just means the connection died rather early on. I have
2912 removed the panic log writing for the ECONNRESET error when getsockname()
2915 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2916 runs only) independently of the message's sender address. This meant
2917 that, if the 4xx error was in fact related to the sender, a different
2918 message to the same recipient with a different sender could confuse
2919 things. In particular, this can happen when sending to a greylisting
2920 server, but other circumstances could also provoke similar problems.
2921 I have changed the default so that the retry time for these errors is now
2922 based a combination of the sender and recipient addresses. This change
2923 can be overridden by setting address_retry_include_sender=false in the
2926 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2927 remote server are returned as part of bounce messages. This was not
2928 happening for LMTP over a pipe (the lmtp transport), but now it is the
2929 same for both kinds of LMTP.
2931 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2932 in header lines that were in fact CNAMEs. This is no longer the case.
2934 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2935 and queue runs started by the daemon processed all messages. This has
2936 been fixed so that -R and -S can now usefully be given with -q<time>.
2938 PH/40 Import PCRE release 6.7 (fixes some bugs).
2940 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2942 PH/42 Give an error if -q is specified more than once.
2944 PH/43 Renamed the variables $interface_address and $interface_port as
2945 $received_ip_address and $received_port, to make it clear that these
2946 values apply to message reception, and not to the outgoing interface when
2947 a message is delivered. (The old names remain recognized, of course.)
2949 PH/44 There was no timeout on the connect() call when using a Unix domain
2950 socket in the ${readsocket expansion. There now is.
2952 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2953 be meaningful with "accept".
2955 SC/01 Eximstats V1.43
2956 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2958 SC/02 Eximstats V1.44
2959 Use a glob alias rather than an array ref in the generated
2960 parser. This improves both readability and performance.
2962 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2963 Collect SpamAssassin and rejection statistics.
2964 Don't display local sender or destination tables unless
2965 there is data to show.
2966 Added average volumes into the top table text output.
2968 SC/04 Eximstats V1.46
2969 Collect data on the number of addresses (recipients)
2970 as well as the number of messages.
2972 SC/05 Eximstats V1.47
2973 Added 'Message too big' to the list of mail rejection
2974 reasons (thanks to Marco Gaiarin).
2976 SC/06 Eximstats V1.48
2977 Mainlog lines which have GMT offsets and are too short to
2978 have a flag are now skipped.
2980 SC/07 Eximstats V1.49 (Alain Williams)
2981 Added the -emptyok flag.
2983 SC/08 Eximstats V1.50
2984 Fixes for obtaining the IP address from reject messages.
2986 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2987 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2988 whitespace changes from 4.64-PH/27
2990 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2993 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2994 are found, allow negative numbers in numeric criteria)
2996 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2998 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2999 to match changes made in 4.64-PH/43
3001 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
3003 PH/47 Put in an explicit test for a DNS lookup of an address record where the
3004 "domain" is actually an IP address, and force a failure. This locks out
3005 those revolvers/nameservers that support "A-for-A" lookups, in
3006 contravention of the specifications.
3008 PH/48 When a host name was looked up from an IP address, and the subsequent
3009 forward lookup of the name timed out, the host name was left in
3010 $sender_host_name, contrary to the specification.
3012 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
3013 restricted to single-key lookups, Exim was not diagnosing an error if
3014 * or *@ was used with a query-style lookup.
3016 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
3018 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
3019 long ago, but noone interested enough thought of it. Let's just say that
3020 the "1.1" means that there are some new functions that weren't there at
3021 some point in the past.
3023 PH/51 Error processing for expansion failure of helo_data from an smtp
3024 transport during callout processing was broken.
3026 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3027 tested/used via the -bh/-bhc/-bs options.
3029 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3030 bug, fixed in subsequent PCRE releases).
3032 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3033 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3035 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3041 SC/01 Use a glob alias rather than an array ref in eximstats generated
3042 parser. This improves both readability and performance.
3044 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3045 Don't display local sender or destination tables in eximstats unless
3046 there is data to show.
3047 Added average volumes into the eximstats top table text output.
3049 SC/03 Collect data on the number of addresses (recipients) as well
3050 as the number of messages in eximstats.
3052 TF/01 Correct an error in the documentation for the redirect router. Exim
3053 does (usually) call initgroups() when daemonizing.
3055 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3056 with consistent privilege compared to when running as a daemon.
3058 TF/03 Note in the spec that $authenticated_id is not set for local
3059 submissions from trusted users.
3061 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3062 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3064 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3065 by adding some example configuration directives to the default
3066 configuration file. A little bit of work is required to uncomment the
3067 directives and define how usernames and passwords are checked, but
3068 there is now a framework to start from.
3070 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3071 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3072 without this. I don't know how relevant this is to other LDAP libraries.
3074 PH/02 Add the verb name to the "unknown ACL verb" error.
3076 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3078 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3080 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3081 directory not expanded when it should be if an expanded home directory
3082 was set for the address (which is overridden by the transport).
3084 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3087 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3088 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3089 because it is too late at that time, and has no effect.
3091 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3092 security issue with \' (bugzilla #107). I could not use the
3093 PQescapeStringConn() function, because it needs a PGconn value as one of
3096 PH/08 When testing addresses using -bt, indicate those final addresses that
3097 are duplicates that would not cause an additional delivery. At least one
3098 person was confused, thinking that -bt output corresponded to deliveries.
3099 (Suppressing duplicates isn't a good idea as you lose the information
3100 about possibly different redirections that led to the duplicates.)
3102 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3103 systems where poll() doesn't work, in particular OS X.
3105 PH/10 Added more information to debugging output for retry time not reached.
3107 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3108 operations in malware.c.
3110 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3113 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3114 syslog_duplication set false, log lines that would normally be written
3115 both the the main log and to the reject log were not written to syslog at
3118 PH/14 In the default configuration, change the use of "message" in ACL warn
3119 statements to "add_header".
3121 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3122 not followed by a command (e.g. "seen endif").
3124 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3125 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3128 PH/17 Added extra conditions to the default value of delay_warning_condition
3132 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3133 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3134 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3137 The Auto-Submitted: and various List- headers are standardised, whereas I
3138 don't think Precedence: ever was.
3140 PH/18 Refactored debugging code in route_finduser() to show more information,
3141 in particular, the error code if getpwnam() issues one.
3143 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3144 This is apparently needed in addition to the PH/07 change above to avoid
3145 any possible encoding problems.
3147 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3148 but not after initializing Perl.
3150 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3151 output them only if debugging. By default they are written stderr,
3152 apparently, which is not desirable.
3154 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3157 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3160 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3162 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3163 authenticated or an ident call has been made. Suppress the default
3164 values for $authenticated_id and $authenticated_sender (but permit -oMai
3165 and -oMas) when testing with -bh.
3167 PH/24 Re-jigged the order of the tests in the default configuration so that the
3168 tests for valid domains and recipients precede the DNS black list and CSA
3169 tests, on the grounds that those ones are more expensive.
3171 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3172 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3173 This bug exists in every version of Exim that I still have, right back to
3176 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3177 However, an attempt to turn on case-sensitivity in a regex key by
3178 including (?-i) didn't work because the subject string was already
3179 lowercased, and the effects were non-intuitive. It turns out that a
3180 one-line patch can be used to allow (?-i) to work as expected.
3186 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3187 other effects) broke the use of negated acl sub-conditions.
3189 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3192 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3193 "Deny" causes Exim to reject the incoming connection with a 554 error.
3194 Unfortunately, if there is a major crisis, such as a disk failure,
3195 tcp-wrappers gives "deny", whereas what one would like would be some
3196 kind of temporary error. A kludge has been added to help with this.
3197 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3198 554 error is used if errno is still zero or contains ENOENT (which occurs
3199 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3202 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3204 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3205 errors. Otherwise a message that provokes a temporary error (when other
3206 messages do not) can cause a whole host to time out.
3208 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3209 addresses were routed directly to files or pipes from a redirect router.
3210 File deliveries just didn't batch; pipe deliveries might have suffered
3213 PH/06 A failure to get a lock for a hints database would erroneously always say
3214 "Failed to get write lock", even when it was really a read lock.
3216 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3217 of 0600. This has been changed to use the value of the lockfile_mode
3218 option (which defaults to 0600).
3220 PH/08 Applied small patch from the Sieve maintainer.
3222 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3223 folder from quota calculations, a direct delivery into this folder messed
3224 up the contents of the maildirsize file. This was because the regex was
3225 used only to exclude .Trash (or whatever) when the size of the mailbox
3226 was calculated. There was no check that a delivery was happening into an
3227 excluded directory. This bug has been fixed by ignoring all quota
3228 processing for deliveries into excluded directories.
3230 PH/10 Added the maildirfolder_create_regex option to appendfile.
3236 PH/01 The code for finding all the local interface addresses on a FreeBSD
3237 system running IPv6 was broken. This may well have applied to all BSD
3238 systems, as well as to others that have similar system calls. The broken
3239 code found IPv4 interfaces correctly, but gave incorrect values for the
3240 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3241 that it would not match correctly against @[] and not recognize the IPv6
3244 PH/02 The ipliteral router was not recognizing addresses of the form user@
3245 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3247 PH/03 Added disable_ipv6.
3249 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3250 decoded form, because it is most often used to construct To: headers
3251 lines in autoreplies, and the decoded form may well be syntactically
3252 invalid. However, $reply_address has leading white space removed, and all
3253 newlines turned into spaces so that the autoreply transport does not
3256 PH/05 If group was specified without a user on a router, and no group or user
3257 was specified on a transport, the group from the router was ignored.
3259 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3260 for visible compile-time settings that can be used to change these
3261 numbers, for those that want even more. Backwards compatibility with old
3262 spool files has been maintained. However, going back to a previous Exim
3263 release will lost any variables that are in spool files.
3265 PH/07 Two small changes when running in the test harness: increase delay when
3266 passing a TCP/IP connection to a new process, in case the original
3267 process has to generate a bounce, and remove special handling of
3268 127.0.0.2 (sic), which is no longer necessary.
3270 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3271 be the same on different OS.
3273 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3276 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3277 whether --show-vars was specified or not
3279 JJ/02 exipick: Added support for new ACL variable spool format introduced
3282 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3283 syntactically invalid From: or Reply-to: line, and a filter used this to
3284 generate an autoreply, and therefore failed to obtain an address for the
3285 autoreply, Exim could try to deliver to a non-existent relative file
3286 name, causing unrelated and misleading errors. What now happens is that
3287 it logs this as a hard delivery error, but does not attempt to create a
3290 PH/11 The exinext utility has a -C option for testing purposes, but although
3291 the given file was scanned by exinext itself; it wasn't being passed on
3292 when Exim was called.
3294 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3295 an end-of-file indication when reading a command response.
3297 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3298 compiled. In many other places in Exim, IPv6 addresses are always
3299 recognized, so I have changed this. It also means that IPv4 domain
3300 literals of the form [IPV4:n.n.n.n] are now always recognized.
3302 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3303 used if the router is not running as root, for example, when verifying at
3304 ACL time, or when using -bh. The debugging output from this situation was
3305 non-existent - all you got was a failure to exec. I have made two
3308 (a) Failures to set uid/gid, the current directory, or a process leader
3309 in a subprocess such as that created by queryprogram now generate
3310 suitable debugging output when -d is set.
3312 (b) The queryprogram router detects when it is not running as root,
3313 outputs suitable debugging information if -d is set, and then runs
3314 the subprocess without attempting to change uid/gid.
3316 PH/15 Minor change to Makefile for building test_host (undocumented testing
3319 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3320 additional section of a DNS packet that returns MX or SRV records.
3321 Instead, it always explicitly searches for A/AAAA records. This avoids
3322 major problems that occur when a DNS server includes only records of one
3323 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3324 fixed another bug: if SRV records were looked up and the corresponding
3325 address records were *not* found in the additional section, the port
3326 values from the SRV records were lost.
3328 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3329 using the correct key (the original address) when searching the retry
3330 rules in order to find which one to use for generating the retry hint.
3332 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3333 adding the default one. Similarly, if it contains a Reply-To: header, the
3334 errors_reply_to option, if set, is not used.
3336 PH/19 When calculating a retry time, Exim used to measure the "time since
3337 failure" by looking at the "first failed" field in the retry record. Now
3338 it does not use this if it is later than than the arrival time of the
3339 message. Instead it uses the arrival time. This makes for better
3340 behaviour in cases where some deliveries succeed, thus re-setting the
3341 "first failed" field. An example is a quota failure for a huge message
3342 when small messages continue to be delivered. Without this change, the
3343 "time since failure" will always be short, possible causing more frequent
3344 delivery attempts for the huge message than are intended.
3345 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3347 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3348 $1, $2, $3) because the numerical variables can be reset during some
3349 expansion items (e.g. "match"), thereby losing the authentication data.
3351 PH/21 Make -bV show the size of off_t variables so that the test suite can
3352 decide whether to run tests for quotas > 2G.
3354 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3355 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3356 filecount value is greater than 2G or if a quota value is greater than 2G
3357 on a system where the size of off_t is not greater than 4, a panic error
3360 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3361 never match. The debug and -bh output now contains an explicit error
3362 message indicating a malformed IPv4 address or mask.
3364 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3365 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3366 PH/23 above applies.
3368 PH/25 Do not write to syslog when running in the test harness. The only
3369 occasion when this arises is a failure to open the main or panic logs
3370 (for which there is an explicit test).
3372 PH/26 Added the /no_tell option to "control=freeze".
3374 PH/27 If a host name lookup failed very early in a connection, for example, if
3375 the IP address matched host_lookup and the reverse lookup yielded a name
3376 that did not have a forward lookup, an error message of the form "no IP
3377 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3378 could be logged. Now it outputs the IP address instead of "NULL".
3380 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3381 allows the sender and the authenticated sender to be set when
3382 submitting a message from within Exim. Since child_open_exim() is
3383 documented for local_scan(), the new function should be too.
3385 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3386 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3387 results in an empty string is now treated as unset.
3389 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3391 PH/31 Added sender_verify_fail logging option.
3393 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3394 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3395 tidied the source and removed it altogether.
3397 PH/33 When a queue run was abandoned because the load average was too high, a
3398 log line was always written; now it is written only if the queue_run log
3399 selector is set. In addition, the log line for abandonment now contains
3400 information about the queue run such as the pid. This is always present
3401 in "start" and "stop" lines but was omitted from the "abandon" line.
3403 PH/34 Omit spaces between a header name and the colon in the error message that
3404 is given when verify = headers_syntax fails (if there are lots of them,
3405 the message gets confusing).
3407 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3408 names, as there are now some PTR records that contain slashes. This check
3409 is only to protect against broken name servers that fall over on strange
3410 characters, so the fact that it applies to all lookups doesn't matter.
3412 PH/36 Now that the new test suite is complete, we can remove some of the
3413 special code in Exim that was needed for the old test suite. For example,
3414 sorting DNS records because real resolvers return them in an arbitrary
3415 order. The new test suite's fake resolver always returns records in the
3418 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3419 bounces) except when queue_only is set, to avoid logging races between
3420 the different processes.
3422 PH/38 Panic-die if .include specifies a non-absolute path.
3424 PH/39 A tweak to the "H" retry rule from its user.
3426 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3427 a label. They prevented compilation on older perls.
3429 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3430 a warning to be raised on newish perls.
3432 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3433 on queue. Changes to match documented behaviour of showing count of
3434 messages matching specified criteria.
3436 PH/40 Changed the default ident timeout from 30s to 5s.
3438 PH/41 Added support for the use of login_cap features, on those BSD systems
3439 that have them, for controlling the resources used by pipe deliveries.
3441 PH/42 The content-scanning code uses fopen() to create files in which to put
3442 message data. Previously it was not paying any attention to the mode of
3443 the files. Exim runs with umask(0) because the rest of the code creates
3444 files with open(), and sets the required mode explicitly. Thus, these
3445 files were ending up world-writeable. This was not a big issue, because,
3446 being within the spool directory, they were not world-accessible. I have
3447 created a function called modefopen, which takes an additional mode
3448 argument. It sets umask(777), creates the file, chmods it to the required
3449 mode, then resets the umask. All the relevant calls to fopen() in the
3450 content scanning code have been changed to use this function.
3452 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3453 to 24 hours. This avoids potential overflow problems when processing G
3454 and H retry rules. I suspect nobody ever tinkers with this value.
3456 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3458 PH/45 When the plaintext authenticator is running as a client, the server's
3459 challenges are checked to ensure they are valid base64 strings. By
3460 default, the authentication attempt is cancelled if an invalid string is
3461 received. Setting client_ignore_invalid_base64 true ignores these errors.
3462 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3463 they are received. Thus, the responses can be made to depend on the
3464 challenges. If an invalid string is ignored, an empty string is placed in
3467 PH/46 Messages that are created by the autoreply transport now contains a
3468 References: header, in accordance with RFCs 2822 and 3834.
3470 PH/47 Added authenticated_sender_force to the smtp transport.
3472 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3474 PH/49 Installed latest patch from the Sieve maintainer.
3476 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3477 was also set, the appendfile transport was unnecessarily scanning a
3478 directory of message files (e.g. for maildir delivery) to find the count
3479 of files (along with the size), even though it did not need this
3480 information. It now does the scan only if it needs to find either the
3481 size of the count of files.
3483 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3485 PH/52 Two bugs concerned with error handling when the smtp transport is
3488 (i) Exim was not creating retry information for temporary errors given
3489 for individual recipients after the DATA command when the smtp transport
3490 was used in LMTP mode. This meant that they could be retried too
3491 frequently, and not timed out correctly.
3493 (ii) Exim was setting the flag that allows error details to be returned
3494 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3495 recipients that were returned after the DATA command.
3497 PH/53 This is related to PH/52, but is more general: for any failing address,
3498 when detailed error information was permitted to be returned to the
3499 sender, but the error was temporary, then after the final timeout, only
3500 "retry timeout exceeded" was returned. Now it returns the full error as
3501 well as "retry timeout exceeded".
3503 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3504 do this, and (what is worse) MTAs that accept it.
3506 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3507 will now be deprecated.
3509 PH/56 New os.c-cygwin from the Cygwin maintainer.
3511 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3512 formats (previously only available in exim formats via -bpr, -bpru,
3513 and -bpra. Now also available in native and exiqgrep formats)
3515 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3516 with very large, slow to parse queues
3518 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3520 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3522 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3523 responses to authentication challenges, though it was showing the
3524 challenges; (ii) I've removed the CR characters from the debug output for
3527 PH/58 Allow for the insertion of a newline as well as a space when a string
3528 is turned into more than one encoded-word during RFC 2047 encoding. The
3529 Sieve code now uses this.
3531 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3532 data_4xx, lost_connection, tls_required.
3534 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3535 message was being sent as an SMTP response.
3537 PH/61 Add -l and -k options to exicyclog.
3539 PH/62 When verifying, if an address was redirected to one new address, so that
3540 verification continued, and the new address failed or deferred after
3541 having set something in $address_data, the value of $address_data was not
3542 passed back to the ACL. This was different to the case when no
3543 redirection occurred. The value is now passed back in both cases.
3545 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3546 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3547 use login_cap.h, so on its own it isn't the distinguishing feature. The
3548 new name refers directly to the setclassresources() function.
3550 PH/65 Added configuration files for NetBSD3.
3552 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3554 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3555 is preferred over IPv4.
3557 PH/68 The bounce_return_message and bounce_return_body options were not being
3558 honoured for bounces generated during the reception of non-SMTP messages.
3559 In particular, this applied to messages rejected by the ACL. This bug has
3560 been fixed. However, if bounce_return_message is true and bounce_return_
3561 body is false, the headers that are returned for a non-SMTP message
3562 include only those that have been read before the error was detected.
3563 (In the case of an ACL rejection, they have all been read.)
3565 PH/69 The HTML version of the specification is now built in a directory called
3566 spec_html instead of spec.html, because the latter looks like a path with
3567 a MIME-type, and this confuses some software.
3569 PH/70 Catch two compiler warnings in sieve.c.
3571 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3572 function verify_get_ident() calls ip_connect() to connect a socket, but
3573 if the "connect()" function timed out, ip_connect() used to close the
3574 socket. However, verify_get_ident() also closes the socket later, and in
3575 between Exim writes to the log, which may get opened at this point. When
3576 the socket was closed in ip_connect(), the log could get the same file
3577 descriptor number as the socket. This naturally causes chaos. The fix is
3578 not to close the socket in ip_connect(); the socket should be closed by
3579 the function that creates it. There was only one place in the code where
3580 this was missing, in the iplookup router, which I don't think anybody now
3581 uses, but I've fixed it anyway.
3583 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3584 well as to direct DNS lookups. Otherwise the handling of names in host
3585 lists is inconsistent and therefore confusing.
3591 PH/01 Two changes to the default runtime configuration:
3593 (1) Move the checks for relay_from_hosts and authenticated clients from
3594 after to before the (commented out) DNS black list checks.
3596 (2) Add control=submission to the relay_from_hosts and authenticated
3597 clients checks, on the grounds that messages accepted by these
3598 statements are most likely to be submissions.
3600 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3602 (1) Generate an error if the third argument for the ${prvs expansion is
3605 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3608 (3) Reset the variables that are obtained from the first argument of
3609 ${prvscheck and used in the second argument before leaving the code,
3610 because their memory is reclaimed, so using them afterwards may do
3613 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3614 one (it's much easier than Tom thought :-).
3616 (5) Because of (4), we can now allow for the use of $prvscheck_result
3617 inside the third argument.
3619 PH/03 For some reason, the default setting of PATH when running a command from
3620 a pipe transport was just "/usr/bin". I have changed it to
3623 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3624 anything to be listed in the output from -bV.
3626 PH/05 When a filter generated an autoreply, the entire To: header line was
3627 quoted in the delivery log line, like this:
3629 => >A.N.Other <ano@some.domain> <original@ddress> ...
3631 This has been changed so that it extracts the operative address. There
3632 may be more than one such address. If so, they are comma-separated, like
3635 => >ano@some.domain,ona@other.domain <original@ddress> ...
3637 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3638 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3639 not being looked up in the rDNS to get a host name, Exim was showing the
3640 IP address twice in Received: lines, even though the IP addresses were
3641 identical. For example:
3643 Received: from [1.2.3.4] (helo=[1.2.3.4])
3645 However, if the real host name was known, it was omitting the HELO data
3646 if it matched the actual IP address. This has been tidied up so that it
3647 doesn't show the same IP address twice.
3649 PH/07 When both +timestamp and +memory debugging was on, the value given by
3650 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3651 called by the debug printing, thereby overwriting the timestamp buffer.
3652 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3654 PH/08 When the original message was included in an autoreply transport, it
3655 always said "this is a copy of the message, including all the headers",
3656 even if body_only or headers_only was set. It now gives an appropriate
3659 PH/09 Applied a patch from the Sieve maintainer which:
3661 o fixes some comments
3662 o adds the (disabled) notify extension core
3663 o adds some debug output for the result of if/elsif tests
3664 o points to the current vacation draft in the documentation
3665 and documents the missing references header update
3669 o fixes a bug in processing the envelope test (when testing
3670 multiple envelope elements, the last element determined the
3673 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3674 Electronic Mail") by including:
3676 Auto-submitted: auto-generated
3678 in the messages that it generates (bounce messages and others, such as
3679 warnings). In the case of bounce messages for non-SMTP messages, there was
3680 also a typo: it was using "Auto_submitted" (underscore instead of
3681 hyphen). Since every message generated by Exim is necessarily in response
3682 to another message, thes have all been changed to:
3684 Auto-Submitted: auto-replied
3686 in accordance with these statements in the RFC:
3688 The auto-replied keyword:
3690 - SHOULD be used on messages sent in direct response to another
3691 message by an automatic process,
3693 - MUST NOT be used on manually-generated messages,
3695 - MAY be used on Delivery Status Notifications (DSNs) and Message
3696 Disposition Notifications (MDNs),
3698 - MUST NOT be used on messages generated by automatic or periodic
3699 processes, except for messages which are automatic responses to
3702 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3703 to the default Received: header definition.
3705 PH/12 Added log selector acl_warn_skipped (default on).
3707 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3708 variables because (a) they are in the wrong storage pool and (b) even if
3709 they were copied, it wouldn't work properly because of the caching.
3711 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3712 checking when decoding. Apparently there are clients that generate
3713 overlong encoded strings. Why am I not surprised?
3715 PH/15 If the first argument of "${if match_address" was not empty, but did not
3716 contain an "@" character, Exim crashed. Now it writes a panic log message
3717 and treats the condition as false.
3719 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3721 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3722 extension "envelope-auth". The code is finished and in agreement with
3723 other implementations, but there is no documentation so far and in fact,
3724 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3725 not changing the active code.
3727 Print executed "if" and "elsif" statements when debugging is used. This
3728 helps a great deal to understand what a filter does.
3730 Document more things not specified clearly in RFC3028. I had all this
3731 sorted out, when out of a sudden new issues came to my mind. Oops."
3733 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3736 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3737 canonical form (as documented). However, after a host name lookup from
3738 the IP address, check_host() was doing a simple string comparison with
3739 addresses acquired from the DNS when checking that the found name did
3740 have the original IP as one of its addresses. Since any found IPv6
3741 addresses are likely to be in abbreviated form, the comparison could
3742 fail. Luckily, there already exists a function for doing the comparison
3743 by converting both addresses to binary, so now that is used instead of
3744 the text comparison.
3746 PH/20 There was another similar case to PH/19, when a complete host name was
3747 given in a host list; looking up its IP address could give an abbreviated
3748 form, whereas the current host's name might or might not be abbreviated.
3749 The same fix has been applied.
3755 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3756 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3759 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3760 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3762 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3764 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3765 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3766 Also, for "Auto-Submitted", treat anything other than "no" as
3767 non-personal, in accordance with RFC 3834. (Previously it treated
3768 anything starting "auto-" as non-personal.)
3770 TF/01 The control=submission/name=... option had a problem with syntax
3771 errors if the name included a slash character. The /name= option
3772 now slurps the rest of the string, so it can include any characters
3773 but it must come last in the list of options (after /sender_retain
3776 PH/05 Some modifications to the interface to the fake nameserver for the new
3784 TK/01 Added the "success_on_redirect" address verification option. See
3785 NewStuff for rationale and an example.
3787 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3789 PH/02 Patch to exigrep to allow it to work on syslog lines.
3791 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3792 fread() to skip over the body file's header line, because in Cygwin the
3793 header line is locked and is inaccessible.
3795 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3796 co-exist for some time) to make it clear that it is the Exim ID that is
3797 referenced, not the Message-ID: header line.
3799 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3800 string_format() function, because snprintf() does not exist on all
3803 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3804 ${stat: expansion item.
3806 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3807 protocol synchronization error", to keep the pedants happy.
3809 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3810 well as for IRIX systems, when gcc is being used. See the host.c source
3813 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3815 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3818 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3819 transport and to the smtp transport in LMTP mode.
3821 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3823 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3824 run a filter in a subprocess. This could lead to confusion in subsequent
3825 lookups in the parent process. There should also be a search_tidyup() at
3826 the end of the subprocess.
3828 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3829 only if the host matched helo_try_verify_hosts, which caused the
3830 verification to occur when the EHLO/HELO command was issued. The ACL just
3831 tested the remembered result. Now, if a previous verification attempt has
3832 not happened, "verify = helo" does it there and then.
3834 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3836 TK/03 Fix log output including CR from clamd.
3838 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3839 exist provoked a memory error which could cause a segfault.
3841 PH/15 Installed PCRE 6.2
3843 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3845 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3846 of the problem. Specifically, suggested +O2 rather than +O1 for the
3849 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3851 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3852 record, the port was not correctly logged when the outgoing_port log
3853 selector was set (it logged the transort's default port).
3855 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3856 fallback_hosts, and "hosts" in the smtp transport.
3858 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3859 host errors such as "Connection refused".
3861 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3862 authentication with radiusclient 0.4.9:
3864 - Error returned from rc_read_config was caught wrongly
3865 - Username/password not passed on to radius server due to wrong length.
3867 The presumption is that some radiusclient API changes for 4.51/PH/17
3868 were not taken care of correctly. The code is still untested by me (my
3869 Linux distribution still has 0.3.2 of radiusclient), but it was
3870 contributed by a Radius user.
3872 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3873 expanding the "port" option of the smtp transport.
3875 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3876 while reading a MIME header. Thanks to Tom Hughes for a patch.
3878 PH/24 Include config.h inside local_scan.h so that configuration settings are
3881 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3882 that in an ACL for RCPT (for example), you can examine exactly what was
3885 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3886 commands, but it was not correctly comparing the address with the actual
3887 client host address. Thus, it would show the EHLO address in Received:
3888 header lines when this was not necessary.
3890 PH/27 Added the % operator to ${eval:}.
3892 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3893 it should be ignoring failures (because with -C, for example, it has lost
3894 privilege). It wasn't ignoring creation failures other than "already
3897 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3900 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3901 because an input line was too long, either on its own, or by virtue of
3902 too many continuations, the temporary file was not being removed, and the
3903 return code was incorrect.
3905 PH/31 Missing "BOOL" in function definition in filtertest.c.
3907 PH/32 Applied Sieve patches from the maintainer.
3909 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3911 PH/33 Added "verify = not_blind".
3913 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3914 Local/Makefile (with some defaults set). These are used in built scripts
3915 such as exicyclog, but they have never been used in the exim_install
3916 script (though there are many overriding facilities there). I have
3917 arranged that the exim_install script now takes note of these two
3920 PH/35 Installed configuration files for Dragonfly.
3922 PH/36 When a locally submitted message by a trusted user did not contain a
3923 From: header, and the sender address was obtained from -f or from an SMTP
3924 MAIL command, and the trusted user did not use -F to supply a sender
3925 name, $originator_name was incorrectly used when constructing a From:
3926 header. Furthermore, $originator_name was used for submission mode
3927 messages from external hosts without From: headers in a similar way,
3928 which is clearly wrong.
3930 PH/37 Added control=suppress_local_fixups.
3932 PH/38 When log_selector = +received_sender was set, and the addition of the
3933 sender made the log line's construction buffer exactly full, or one byte
3934 less than full, an overflow happened when the terminating "\n" was
3937 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3938 when the result of a list match is failure because a DNS lookup failed.
3940 PH/40 RM_COMMAND is now used in the building process.
3942 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3943 the "build-* directories that it finds.
3945 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3946 address in a domain literal was a prefix of an interface address.
3948 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3949 when verifying a sender address, unless rewrite_headers is false.
3951 PH/44 Wrote a long comment about why errors_to addresses are verified as
3952 recipients, not senders.
3954 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3955 the ratelimit ACL was added.
3957 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3959 PH/46 Added extra information about PostgreSQL errors to the error string.
3961 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3962 avoiding the need to install special zones in a real server. This is
3963 backwards compatible; if it can't find the fake resolver, it drops back.
3964 Thus, both old and new test suites can be run.
3966 TF/02 Added util/ratelimit.pl
3968 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3969 clock is set back in time.
3971 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3972 Candler <B.Candler@pobox.com>.
3974 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3975 OK for addresses that are the result of redirections.
3977 PH/48 A number of further additions for the benefit of the new test suite,
3978 including a fake gethostbyname() that interfaces to the fake DNS resolver
3981 TF/06 The fix for widen_domains has also been applied to qualify_single and
3982 search_parents which are the other dnslookup options that can cause
3985 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3988 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3990 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3991 service" if it encountered an MX record with an empty target hostname.
3992 The message is now "an MX or SRV record indicated no SMTP service".
3994 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3995 if the DNS of the sending site is misconfigured. This is quite a
3996 common situation. This change restores the behaviour of treating a
3997 helo verification defer as a failure.
3999 PH/51 If self=fail was set on a router, the bounce message did not include the
4000 actual error message.
4006 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
4008 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
4009 command itself ended in error, the underlying message about the transport
4010 filter timeout was being overwritten with the pipe command error. Now the
4011 underlying error message should be appended to the second error message.
4013 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
4015 PH/02 Reduce the amount of output that "make" produces by default. Full output
4016 can still be requested.
4018 PH/03 The warning log line about a condition test deferring for a "warn" verb
4019 was being output only once per connection, rather than after each
4020 occurrence (because it was using the same function as for successful
4021 "warn" verbs). This seems wrong, so I have changed it.
4023 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4024 it should not have, which might have caused a crash in the right
4025 circumstances, but probably never did.
4027 PH/04 Installed a modified version of Tony Finch's patch to make submission
4028 mode fix the return path as well as the Sender: header line, and to
4029 add a /name= option so that you can make the user's friendly name appear
4032 TF/03 Added the control = fakedefer ACL modifier.
4034 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4035 Mark Lowes for thorough testing.
4037 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4039 TK/03 Merged latest SRS patch from Miles Wilton.
4041 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4042 with the definition in sysexits.h (which is #included earlier).
4043 Fortunately, Exim does not actually use EX_OK. The code used to try to
4044 preserve the sysexits.h value, by assuming that macro definitions were
4045 scanned for macro replacements. I have been disabused of this notion,
4046 so now the code just undefines EX_OK before #including unistd.h.
4048 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4049 in the smtp transport. When a block could not be written in a single
4050 write() function, the timeout was being re-applied to each part-write.
4051 This seems wrong - if the receiver was accepting one byte at a time it
4052 would take for ever. The timeout is now adjusted when this happens. It
4053 doesn't have to be particularly precise.
4055 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4056 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4058 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4059 without a domain if the check to <postmaster@domain> fails.
4061 SC/01 Eximstats: added -xls and the ability to specify output files
4062 (patch written by Frank Heydlauf).
4064 SC/02 Eximstats: use FileHandles for outputting results.
4066 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4068 SC/04 Eximstats: fixed display of large numbers with -nvr option
4070 SC/05 Eximstats: fixed merging of reports with empty tables.
4072 SC/06 Eximstats: added the -include_original_destination flag
4074 SC/07 Eximstats: removed tabs and trailing whitespace.
4076 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4078 TK/06 MBOX spool code: Add real "From " MBOX separator line
4079 so the .eml file is really in mbox format (even though
4080 most programs do not really care). Patch from Alex Miller.
4082 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4083 The latter is generated from $received_to and is only set if the
4084 message has one envelope recipient. SA can use these headers,
4085 obviously out-of-the-box. Patch from Alex Miller.
4087 PH/08 The ${def test on a variable was returning false if the variable's
4088 value was "0", contrary to what the specification has always said!
4089 The result should be true unless the variable is empty.
4091 PH/09 The syntax error of a character other than { following "${if
4092 def:variable_name" (after optional whitespace) was not being diagnosed.
4093 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4094 accidental colon was present, for example, could give incorrect results.
4096 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4097 result is used (not including appendfile, where other changes are about
4100 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4101 This involved changing a lot of size variables from int to off_t. It
4102 should work with maildirs and everything.
4104 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4105 spamd dying while we are connected to it.
4107 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4110 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4111 function for BDB 4.3.
4113 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4115 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4116 the sources for PCRE have been re-arranged and more files are now
4119 PH/15 The code I had for printing potentially long long variables in PH/11
4120 above was not the best (it lost precision). The length of off_t variables
4121 is now inspected at build time, and an appropriate printing format (%ld
4122 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4123 to be "long long int" or "long int". This is needed for the internal
4124 formatting function string_vformat().
4126 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4127 the configuration file to be ":syslog", then the script "guesses" where
4128 the logs files are, rather than using the compiled in default. In our
4129 case the guess is not the same as the compiled default, so the script
4130 suddenly stopped working when I started to use syslog. The patch checks
4131 to see if log_file_path is "". If so, it attempts to read it from exim
4132 with no configuration file to get the compiled in version, before it
4133 falls back to the previous guessing code."
4135 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4136 implementing BATV in an Exim configuration. See NewStuff for the gory
4139 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4140 Makefile that are specific to HP-UX.
4142 PH/18 If the "use_postmaster" option was set for a recipient callout together
4143 with the "random" option, the postmaster address was used as the MAIL
4144 FROM address for the random test, but not for the subsequent recipient
4145 test. It is now used for both.
4147 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4148 patch removes a few documentation additions to RFC 3028, because the
4149 latest draft now contains them. It adds the new en;ascii-case comparator
4150 and a new error check for 8bit text in MIME parts. Comparator and
4151 require names are now matched exactly. I enabled the subaddress
4152 extension, but it is not well tested yet (read: it works for me)."
4154 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4155 rework some of the code of TK/09 above to avoid the hardwired use of
4156 "%lld" and "long long". Replaced the call to snprintf() with a call to
4159 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4160 records point to non-existent hosts", "retry timeout exceeded", and
4161 "retry time not reached for any host after a long failure period".
4163 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4164 experimental DomainKeys support:
4166 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4167 (2) On an error such as an illegally used "control", the wrong name for
4168 the control was given.
4170 These problems did NOT occur unless DomainKeys support was compiled.
4172 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4174 PH/24 Added ${if match_ip condition.
4176 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4177 databases so that it will be absolutely obvious if a crash occurs in the
4178 DB library. This is a regular occurrence (often caused by mis-matched
4181 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4182 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4183 picked up on a user's system that detects such things. There doesn't seem
4184 to be a gcc warning option for this - only an attribute that has to be
4185 put on the function's prototype. It seems that in Fedora Core 4 they have
4186 set this on a number of new functions. No doubt there will be more in due
4189 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4190 specify a transport. However, if an address that was verified by such a
4191 router was the subject of a callout, Exim crashed because it tried to
4192 read the rcpt_include_affixes from the non-existent transport. Now it
4193 just assumes that the setting of that option is false. This bug was
4194 introduced by 4.51/PH/31.
4196 PH/28 Changed -d+all to exclude +memory, because that information is very
4197 rarely of interest, but it makes the output a lot bigger. People tend to
4198 do -d+all out of habit.
4200 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4201 code in os-type was giving problems when libc.so lives in lib64, like on
4204 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4205 aren't the modern standard, and it seems that some systems' include files
4206 don't always have them. Exim was already checking for some of the newer
4207 ones like T_AAAA, and defining it itself. I've added checks for all the
4208 record types that Exim uses.
4210 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4211 not automatically generating a new one, as it is supposed to. This
4212 prevented TLS from working. If the file did exist, but contained invalid
4213 data, a new version was generated, as expected. It was only the case of a
4214 non-existent file that was broken.
4216 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4217 with a change in libdomainkeys > 0.64.
4219 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4220 from DNS. If the selector record carries the flag, it now has
4221 precedence over the domain-wide flag.
4223 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4225 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4226 the use of a port name that isn't defined in /etc/services) occurred, the
4227 message was deferred as in a normal delivery, and thus remained on the
4228 spool, instead of being failed because of the mua_wrapper setting. This
4229 is now fixed, and I tidied up some of the mua_wrapper messages at the
4232 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4233 the messages in a hash of arrays rather than using individual hashes.
4234 This is a bit cleaner and results in dramatic memory savings, albeit
4235 at a slight CPU cost.
4237 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4238 as requested by Marc Sherman.
4240 SC/10 Eximstats: added histograms for user specified patterns as requested
4243 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4245 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4246 fopen() in the content-scanning modules that did not already have it.
4252 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4253 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4255 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4257 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4259 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4260 to test Sieve filters that use "vacation".
4262 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4263 that changes the way the GnuTLS parameters are stored in the cache file.
4264 The new format can be generated externally. For backward compatibility,
4265 if the data in the cache doesn't make sense, Exim assumes it has read an
4266 old-format file, and it generates new data and writes a new file. This
4267 means that you can't go back to an older release without removing the
4270 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4271 work if there are any delivery delays because "one_time" forces the
4272 parent to be marked "delivered", so its unseen clone is never tried
4273 again. For this reason, Exim now forbids the simultaneous setting of
4276 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4277 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4278 entry if you want to know the details. The fix, however, neglected to
4279 consider the case when local delivery batching is involved. The test for
4280 "previously delivered" was not happening when checking to see if an
4281 address could be batched with a previous (undelivered) one; under
4282 certain circumstances this could lead to multiple deliveries to the same
4285 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4286 in its include files, and this causes problems building Exim.
4288 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4289 header_syntax) but Exim was just ignoring anything given after a slash.
4290 In particular, this caused confusion with an attempt to use "verify =
4291 reverse_host_lookup/defer_ok". An error is now given when options are
4292 supplied for verify items that do not have them. (Maybe reverse_host_
4293 lookup should have a defer_ok option, but that's a different point.)
4295 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4296 defined by RFC 821) to 2048, because there were problems with some AUTH
4297 commands, and RFC 1869 says the size should be increased for extended
4298 SMTP commands that take arguments.
4300 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4303 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4304 "unknown" error; now it says that the functionality isn't in the binary.
4306 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4307 an address' error message when a string expansion fails (syntax or
4308 whatever). Otherwise the password may appear in the log. Following change
4309 PH/42 below, there is no longer a chance of it appearing in a bounce
4312 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4314 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4315 of its addresses was ever tried. (Bugzilla bug #2.)
4317 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4318 the result incorrectly in the debug output. (It correctly added a newline
4319 to what was transported.)
4321 TF/01 Added $received_time.
4323 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4324 commented out examples of how to interface to a virus scanner and to
4325 SpamAssassin. Also added commented examples of av_scanner and
4326 spamd_address settings.
4328 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4329 and controls are allowed in which ACLs. There were a couple of minor
4330 errors. Some of the entries in the conditions table (which is a table of
4331 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4332 negation of where the condition IS allowed.
4334 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4336 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4337 header file does not have a version number, so I've had to invent a new
4338 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4339 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4340 radiusclient), but it was contributed by a Radius user.
4342 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4343 files or directories, for OpenSSL.
4345 PH/19 When an Exim process that is running as root has to create an Exim log
4346 file, it does so in a subprocess that runs as exim:exim so as to get the
4347 ownership right at creation (otherwise, other Exim processes might see
4348 the file with the wrong ownership). There was no test for failure of this
4349 fork() call, which would lead to the process getting stuck as it waited
4350 for a non-existent subprocess. Forks do occasionally fail when resources
4351 run out. I reviewed all the other calls to fork(); they all seem to check
4354 PH/20 When checking for unexpected SMTP input at connect time (before writing
4355 the banner), Exim was not dealing correctly with a non-positive return
4356 from the read() function. If the client had disconnected by this time,
4357 the result was a log entry for a synchronization error with an empty
4358 string after "input=" when read() returned zero. If read() returned -1
4359 (an event I could not check), uninitialized data bytes were printed.
4360 There were reports of junk text (parts of files, etc) appearing after
4363 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4365 PH/22 Added support for macro redefinition, and (re)definition in between
4366 driver and ACL definitions.
4368 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4369 forgetting to use the resulting value; it was using the unexpanded value.
4371 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4372 hadn't been configured. The fix is from Juergen Kreileder, who
4373 understands it better than I do:
4375 "Here's what I see happening with three configured cyrus_sasl
4376 authenticators configured (plain, login, cram-md5):
4378 On startup auth_cyrus_sasl_init() gets called for each of these.
4379 This means three calls to sasl_listmech() without a specified mech_list.
4380 => SASL tests which mechs of all available mechs actually work
4381 => three warnings about OTP not working
4382 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4384 With the patch, sasl_listmech() also gets called three times. But now
4385 SASL's mech_list option is set to the server_mech specified in the the
4386 authenticator. Or in other words, the answer from sasl_listmech()
4387 gets limited to just the mech you're testing for (which is different
4389 => the return list contains just 'plain' or 'login', 'cram-md5' or
4390 nothing depending on the value of ob->server_mech.
4392 I've just tested the patch: Authentication still works fine,
4393 unavailable mechs specified in the exim configuration are still
4394 caught, and the auth.log warnings about OTP are gone."
4396 PH/25 When debugging is enabled, the contents of the command line are added
4397 to the debugging output, even when log_selector=+arguments is not
4400 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4401 answer is "GNU", and only if the return is "GNU/something" is the answer
4404 PH/27 $acl_verify_message is now set immediately after the failure of a
4405 verification in an ACL, and so is available in subsequent modifiers. In
4406 particular, the message can be preserved by coding like this:
4408 warn !verify = sender
4409 set acl_m0 = $acl_verify_message
4411 Previously, $acl_verify_message was set only while expanding "message"
4412 and "log_message" when a very denied access.
4414 PH/28 Modified OS/os.c-Linux with
4416 -#ifndef OS_LOAD_AVERAGE
4417 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4419 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4420 nomenclature these days.)
4422 PH/29 Installed patch from the Sieve maintainer that adds the options
4423 sieve_useraddress and sieve_subaddress to the redirect router.
4425 PH/30 In these circumstances:
4426 . Two addresses routed to the same list of hosts;
4427 . First host does not offer TLS;
4428 . First host accepts first address;
4429 . First host gives temporary error to second address;
4430 . Second host offers TLS and a TLS session is established;
4431 . Second host accepts second address.
4432 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4433 and peerdn, if requested) that were in fact used only for the second
4436 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4437 attention to any local part prefix or suffix that was matched by the
4438 router that accepted the address. It now behaves in the same way as it
4439 does for delivery: the affixes are removed from the local part unless
4440 rcpt_include_affixes is set on the transport.
4442 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4443 timeout during the DATA phase of an incoming message.
4445 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4446 applied a patch sanctioned by the Sieve maintainer.
4448 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4449 the uid or gid is negative. A case of a negative gid caused this to be
4450 noticed. The fix allows for either to be negative.
4452 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4453 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4456 PH/36 The change PH/12 above was broken. Fixed it.
4458 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4459 the grounds that routing the same address twice would always produce the
4460 same answer. This might have been true once, but it is certainly no
4461 longer true now. Routing a child address may depend on the previous
4462 routing that produced that child. Some complicated redirection strategies
4463 went wrong when messages had multiple recipients, and made Exim's
4464 behaviour dependent on the order in which the addresses were given.
4466 I have moved the duplicate checking until after the routing is complete.
4467 Exim scans the addresses that are assigned to local and remote
4468 transports, and removes any duplicates. This means that more work will be
4469 done, as duplicates will always all be routed, but duplicates are
4470 presumably rare, so I don't expect this is of any significance.
4472 For deliveries to pipes, files, and autoreplies, the duplicate checking
4473 still happens during the routing process, since they are not going to be
4476 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4477 It corrects a timeout issue with spamd. This is Ian's comment: "The
4478 background is that sometimes spamd either never reads data from a
4479 connection it has accepted, or it never writes response data. The exiscan
4480 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4481 blindly assumes that writes won't block so it may never time out."
4483 PH/39 Allow G after quota size as well as K and M.
4485 PH/40 The value set for $authenticated_id in an authenticator may not contain
4486 binary zeroes or newlines because the value is written to log lines and
4487 to spool files. There was no check on this. Now the value is run through
4488 the string_printing() function so that such characters are converted to
4489 printable escape sequences.
4491 PH/41 $message_linecount is a new variable that contains the total number of
4492 lines in the message. Compare $body_linecount, which is the count for the
4495 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4496 bounce and delay warning messages, except in certain special cases, which
4499 (a) An SMTP error message from a remote host;
4500 (b) A message specified in a :fail: redirection;
4501 (c) A message specified in a "fail" command in a system filter;
4502 (d) A message specified in a FAIL return from the queryprogram router;
4503 (e) A message specified by the cannot_route_message router option.
4505 In these cases only, Exim does include the error details in bounce and
4506 warning messages. There are also a few cases where bland messages such
4507 as "unrouteable address" or "local delivery error" are given.
4509 PH/43 $value is now also set for the "else" part of a ${run expansion.
4511 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4512 being worked on, but at least Exim now implements the latest version to
4515 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4516 process to complete was treated as a delivery failure, a timeout while
4517 writing the message to the pipe was logged, but erroneously treated as a
4518 successful delivery. Such timeouts include transport filter timeouts. For
4519 consistency with the overall process timeout, these timeouts are now
4520 treated as errors, giving rise to delivery failures by default. However,
4521 there is now a new Boolean option for the pipe transport called
4522 timeout_defer, which, if set TRUE, converts the failures into defers for
4523 both kinds of timeout. A transport filter timeout is now identified in
4526 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4527 systems where "make" and "gmake" are different, calling "gmake" at top
4528 level broke things. I've arranged for the value of $(MAKE) to be passed
4529 from the Makefile to this script so that it can call the same version of
4533 A note about Exim versions 4.44 and 4.50
4534 ----------------------------------------
4536 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4537 changes of various kinds. As a consequence, a big documentation update was
4538 needed. This delayed the release for rather longer than seemed good, especially
4539 in the light of a couple of (minor) security issues. Therefore, the changes
4540 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4541 release. So 4.44 and 4.50 are in effect two different branches that both start
4544 I have left the 4.50 change log unchanged; it contains all the changes since
4545 4.43. The change log for 4.44 is below; many of its items are identical to
4546 those for 4.50. This seems to be the most sensible way to preserve the
4547 historical information.
4553 1. Minor wording change to the doc/README.SIEVE file.
4555 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4556 computation of the current number of files was incorrect.
4558 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4559 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4562 4. Give more explanation in the error message when the command for a transport
4563 filter fails to execute.
4565 5. There are several places where Exim runs a non-Exim command in a
4566 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4567 was being done only for the command run by the queryprogram router. It is
4568 now done for all such subprocesses. The other cases are: ${run, transport
4569 filters, and the commands run by the lmtp and pipe transports.
4571 6. Added CONFIGURE_GROUP build-time option.
4573 7. Some older OS have a limit of 256 on the maximum number of file
4574 descriptors. Exim was using setrlimit() to set 1000 as a large value
4575 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4576 systems. I've change it so that if it can't get 1000, it tries for 256.
4578 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4579 was an oversight, and furthermore, ever since the addition of extra
4580 controls (e.g. 4.43/32), the checks on when to allow different forms of
4581 "control" were broken. There should now be diagnostics for all cases when a
4582 control that does not make sense is encountered.
4584 9. Added the /retain_sender option to "control=submission".
4586 10. $recipients is now available in the predata ACL (oversight).
4588 11. Tidy the search cache before the fork to do a delivery from a message
4589 received from the command line. Otherwise the child will trigger a lookup
4590 failure and thereby defer the delivery if it tries to use (for example) a
4591 cached ldap connection that the parent has called unbind on.
4593 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4594 of $address_data from the recipient verification was clobbered by the
4595 sender verification.
4597 13. The value of address_data from a sender verification is now available in
4598 $sender_address_data in subsequent conditions in the ACL statement.
4600 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4602 15. Added a new option "connect=<time>" to callout options, to set a different
4605 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4606 was its contents. (It was OK if the option was not defined at all.)
4608 17. A "Completed" log line is now written for messages that are removed from
4609 the spool by the -Mrm option.
4611 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4612 information about exactly what failed.
4614 19. Added -dd to debug only the daemon process.
4616 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4617 handles timeouts, both on the server side and network timeouts. Renamed the
4618 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4620 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4621 It is now set to "smtps".
4623 22. $host_address is now set to the target address during the checking of
4624 ignore_target_hosts.
4626 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4627 being passed; this would have caused $sender_host_name to have been used if
4628 matching the list had actually called for a host name (not very likely,
4629 since this list is usually IP addresses). A host name is now passed as
4632 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4633 code that specifies a non-restarting handler (typically sigaction() in
4634 modern systems) in an attempt to fix a rare and obscure crash bug.
4636 25. Narrowed the window for a race in the daemon that could cause it to ignore
4637 SIGCHLD signals. This is not a major problem, because they are used only to
4638 wake it up if nothing else does.
4640 26. A malformed maildirsize file could cause Exim to calculate negative values
4641 for the mailbox size or file count. Odd effects could occur as a result.
4642 The maildirsize information is now recalculated if the size or filecount
4645 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4646 support for a long time. Removed HAVE_SYS_VFS_H.
4648 28. Installed the latest version of exipick from John Jetmore.
4650 29. In an address list, if the pattern was not a regular expression, an empty
4651 subject address (from a bounce message) matched only if the pattern was an
4652 empty string. Non-empty patterns were not even tested. This was the wrong
4653 because it is perfectly reasonable to use an empty address as part of a
4654 database query. An empty address is now tested by patterns that are
4655 lookups. However, all the other forms of pattern expect the subject to
4656 contain a local part and a domain, and therefore, for them, an empty
4657 address still always fails if the pattern is not itself empty.
4659 30. Exim went into a mad DNS loop when attempting to do a callout where the
4660 host was specified on an smtp transport, and looking it up yielded more
4661 than one IP address.
4663 31. Re-factored the code for checking spool and log partition space into a
4664 function that finds that data and another that does the check. The former
4665 is then used to implement four new variables: $spool_space, $log_space,
4666 $spool_inodes, and $log_inodes.
4668 32. The RFC2047 encoding function was originally intended for short strings
4669 such as real names; it was not keeping to the 75-character limit for
4670 encoded words that the RFC imposes. It now respects the limit, and
4671 generates multiple encoded words if necessary. To be on the safe side, I
4672 have increased the buffer size for the ${rfc2047: expansion operator from
4675 33. It is now permitted to omit both strings after an "if" condition; if the
4676 condition is true, the result is "true". As before, when the second string
4677 is omitted, a false condition yields an empty string. This makes it less
4678 cumbersome to write custom ACL and router conditions.
4680 34. Failure to deliver a bounce message always caused it to be frozen, even if
4681 there was an errors_to setting on the router. The errors_to setting is now
4684 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4685 canonical form (fully expanded) before being placed in
4686 $sender_host_address.
4688 36. The table in the code that translates DNS record types into text (T_A to
4689 "A" for instance) was missing entries for NS and CNAME. It is just possible
4690 that this could have caused confusion if both these types were looked up
4691 for the same domain, because the text type is used as part of Exim's
4692 per-process caching. But the chance of anyone hitting this buglet seems
4695 37. The dnsdb lookup has been extended in a number of ways.
4697 (1) There is a new type, "zns", which walks up the domain tree until it
4698 finds some nameserver records. It should be used with care.
4700 (2) There is a new type, "mxh", which is like "mx" except that it returns
4701 just the host names, not the priorities.
4703 (3) It is now possible to give a list of domains (or IP addresses) to be
4704 looked up. The behaviour when one of the lookups defers can be
4705 controlled by a keyword.
4707 (4) It is now possible to specify the separator character for use when
4708 multiple records are returned.
4710 38. The dnslists ACL condition has been extended: it is now possible to supply
4711 a list of IP addresses and/or domains to be looked up in a particular DNS
4714 39. Added log_selector=+queue_time_overall.
4716 40. When running the queue in the test harness, wait just a tad after forking a
4717 delivery process, to get repeatability of debugging output.
4719 41. Include certificate and key file names in error message when GnuTLS fails
4720 to set them up, because the GnuTLS error message doesn't include the name
4721 of the failing file when there is a problem reading it.
4723 42. Allow both -bf and -bF in the same test run.
4725 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4727 44. The "Exiscan patch" is now merged into the mainline Exim source.
4729 45. Sometimes the final signoff response after QUIT could fail to get
4730 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4731 < 0 before doing a fflush(). This bug looks as though it goes back to the
4732 introduction of TLS in release 3.20, but "sometimes" must have been rare
4733 because the tests only now provoked it.
4735 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4736 (this can affect the format of dates).
4738 47. exim_tidydb, when checking for the continued existence of a message for
4739 which it has found a message-specific retry record, was not finding
4740 messages that were in split spool directories. Consequently, it was
4741 deleting retry records that should have stayed in existence.
4743 48. Steve fixed some bugs in eximstats.
4745 49. The SPA authentication driver was not abandoning authentication and moving
4746 on to the next authenticator when an expansion was forced to fail,
4747 contradicting the general specification for all authenticators. Instead it
4748 was generating a temporary error. It now behaves as specified.
4750 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4751 (the order specifies the preference for clients). The order is now AES256,
4752 AES128, 3DES, ARCFOUR128.
4754 51. Small patch to Sieve code - explicitly set From: when generating an
4757 52. Exim crashed if a remote delivery caused a very long error message to be
4758 recorded - for instance if somebody sent an entire SpamAssassin report back
4759 as a large number of 550 error lines. This bug was coincidentally fixed by
4760 increasing the size of one of Exim's internal buffers (big_buffer) that
4761 happened as part of the Exiscan merge. However, to be on the safe side, I
4762 have made the code more robust (and fixed the comments that describe what
4765 53. Now that there can be additional text after "Completed" in log lines (if
4766 the queue_time_overall log selector is set), a one-byte patch to exigrep
4767 was needed to allow it to recognize "Completed" as not the last thing in
4770 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4771 patch that reportedly fixes this has been added. I am not expert enough to
4772 create a test for it. This is what the patch creator wrote:
4774 "I found a little strange behaviour of ldap code when working with
4775 Windows 2003 AD Domain, where users was placed in more than one
4776 Organization Units. When I tried to give exim partial DN, the exit code
4777 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4778 But simultaneously result of request was absolutely normal ldap result,
4779 so I produce this patch..."
4781 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4782 REFERENCE, so I have modified the code to exclude the patch when that macro
4785 55. Some experimental protocols are using DNS PTR records for new purposes. The
4786 keys for these records are domain names, not reversed IP addresses. The
4787 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4788 leaves it alone. Component reversal etc. now happens only for IP addresses.
4791 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4793 57. Double the size of the debug message buffer (to 2048) so that more of very
4794 long debug lines gets shown.
4796 58. The exicyclog utility now does better if the number of log files to keep
4797 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4799 59. Two changes related to the smtp_active_hostname option:
4801 (1) $smtp_active_hostname is now available as a variable.
4802 (2) The default for smtp_banner uses $smtp_active_hostname instead
4803 of $primary_hostname.
4805 60. The host_aton() function is supposed to be passed a string that is known
4806 to be a valid IP address. However, in the case of IPv6 addresses, it was
4807 not checking this. This is a hostage to fortune. Exim now panics and dies
4808 if the condition is not met. A case was found where this could be provoked
4809 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4810 components; fortuitously, this particular loophole had already been fixed
4811 by change 4.50/55 above.
4813 If there are any other similar loopholes, the new check in host_aton()
4814 itself should stop them being exploited. The report I received stated that
4815 data on the command line could provoke the exploit when Exim was running as
4816 exim, but did not say which command line option was involved. All I could
4817 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4818 running as the user.
4821 61. There was a buffer overflow vulnerability in the SPA authentication code
4822 (which came originally from the Samba project). I have added a test to the
4823 spa_base64_to_bits() function which I hope fixes it.
4826 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4827 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4829 63. The daemon start-up calls getloadavg() while still root for those OS that
4830 need the first call to be done as root, but it missed one case: when
4831 deliver_queue_load_max is set with deliver_drop_privilege. This is
4832 necessary for the benefit of the queue runner, because there is no re-exec
4833 when deliver_drop_privilege is set.
4835 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4836 This has been fixed.
4838 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4839 was in use, was not putting the data itself into the right store pool;
4840 consequently, it could be overwritten for a subsequent message in the same
4841 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4844 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4846 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4847 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4848 the function were treating the return as a boolean value, which happened to
4849 work because 0=false and not-0=true, but is not correct code.
4851 68. The host_aton() function was not handling scoped IPv6 addresses (those
4852 with, for example, "%eth0" on the end) correctly.
4854 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4855 negated items (that is, ~something) in unsigned ints. Some compilers
4856 apparently mutter when there is no cast.
4858 70. If an address verification called from an ACL failed, and did not produce a
4859 user-specific message (i.e. there was only a "system" message), nothing was
4860 put in $acl_verify_message. In this situation, it now puts the system
4863 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4864 session; change 4.31/43 added the unwanted input to the log line - except
4865 that it did not do this in the start of session case. It now does.
4867 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4868 This is wrong and can cause the other end to generate a synchronization
4869 error if it is another Exim or anything else that does the synchronization
4870 check. A QUIT command is no longer sent after a timeout.
4872 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4873 during host lookups.
4875 74. The defer_ok option of callout verification was not working if it was used
4876 when verifying addresses in header lines, that is, for this case:
4878 verify = header_sender/callout=defer_ok
4880 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4881 those file descriptors could be used for SMTP connections. If anything
4882 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4883 could be sent to the SMTP client, causing chaos. The daemon now opens
4884 stdin, stdout, and stderr to /dev/null when it puts itself into the
4887 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4888 log by default. The user can override this with suitable Perl magic.
4890 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4891 the log message when discard triggers, was not working for the DATA ACL or
4892 for the non-SMTP ACL.
4894 78. Error message wording change in sieve.c.
4896 79. If smtp_accept_max_per_host was set, the number of connections could be
4897 restricted to fewer than expected, because the daemon was trying to set up
4898 a new connection before checking whether the processes handling previous
4899 connections had finished. The check for completed processes is now done
4900 earlier. On busy systems, this bug wouldn't be noticed because something
4901 else would have woken the daemon, and it would have reaped the completed
4904 80. If a message was submitted locally by a user whose login name contained one
4905 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4906 It caused a spool format error. I have fixed the spool reading code. A
4907 related problem was that the "from" clause in the Received: line became
4908 illegal because of the space(s). It is now covered by ${quote_local_part.
4910 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4913 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4915 83. Patch from Sieve maintainer to fix unterminated string problem in
4916 "vacation" handling.
4918 84. Some minor changes to the Linux configuration files to help with other
4919 OS variants using glibc.
4921 85. One more patch for Sieve to update vacation handling to latest spec.
4924 ----------------------------------------------------
4925 See the note above about the 4.44 and 4.50 releases.
4926 ----------------------------------------------------
4932 1. Change 4.43/35 introduced a bug that caused file counts to be
4933 incorrectly computed when quota_filecount was set in an appendfile
4936 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4937 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4940 3. Give more explanation in the error message when the command for a transport
4941 filter fails to execute.
4943 4. There are several places where Exim runs a non-Exim command in a
4944 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4945 was being done only for the command run by the queryprogram router. It is
4946 now done for all such subprocesses. The other cases are: ${run, transport
4947 filters, and the commands run by the lmtp and pipe transports.
4949 5. Some older OS have a limit of 256 on the maximum number of file
4950 descriptors. Exim was using setrlimit() to set 1000 as a large value
4951 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4952 systems. I've change it so that if it can't get 1000, it tries for 256.
4954 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4955 was an oversight, and furthermore, ever since the addition of extra
4956 controls (e.g. 4.43/32), the checks on when to allow different forms of
4957 "control" were broken. There should now be diagnostics for all cases when a
4958 control that does not make sense is encountered.
4960 7. $recipients is now available in the predata ACL (oversight).
4962 8. Tidy the search cache before the fork to do a delivery from a message
4963 received from the command line. Otherwise the child will trigger a lookup
4964 failure and thereby defer the delivery if it tries to use (for example) a
4965 cached ldap connection that the parent has called unbind on.
4967 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4968 of $address_data from the recipient verification was clobbered by the
4969 sender verification.
4971 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4972 was its contents. (It was OK if the option was not defined at all.)
4974 11. A "Completed" log line is now written for messages that are removed from
4975 the spool by the -Mrm option.
4977 12. $host_address is now set to the target address during the checking of
4978 ignore_target_hosts.
4980 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4981 being passed; this would have caused $sender_host_name to have been used if
4982 matching the list had actually called for a host name (not very likely,
4983 since this list is usually IP addresses). A host name is now passed as
4986 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4987 code that specifies a non-restarting handler (typically sigaction() in
4988 modern systems) in an attempt to fix a rare and obscure crash bug.
4990 15. Narrowed the window for a race in the daemon that could cause it to ignore
4991 SIGCHLD signals. This is not a major problem, because they are used only to
4992 wake it up if nothing else does.
4994 16. A malformed maildirsize file could cause Exim to calculate negative values
4995 for the mailbox size or file count. Odd effects could occur as a result.
4996 The maildirsize information is now recalculated if the size or filecount
4999 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5000 support for a long time. Removed HAVE_SYS_VFS_H.
5002 18. Updated exipick to current release from John Jetmore.
5004 19. Allow an empty sender to be matched against a lookup in an address list.
5005 Previously the only cases considered were a regular expression, or an
5008 20. Exim went into a mad DNS lookup loop when doing a callout where the
5009 host was specified on the transport, if the DNS lookup yielded more than
5012 21. The RFC2047 encoding function was originally intended for short strings
5013 such as real names; it was not keeping to the 75-character limit for
5014 encoded words that the RFC imposes. It now respects the limit, and
5015 generates multiple encoded words if necessary. To be on the safe side, I
5016 have increased the buffer size for the ${rfc2047: expansion operator from
5019 22. Failure to deliver a bounce message always caused it to be frozen, even if
5020 there was an errors_to setting on the router. The errors_to setting is now
5023 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5024 canonical form (fully expanded) before being placed in
5025 $sender_host_address.
5027 24. Updated eximstats to version 1.33
5029 25. Include certificate and key file names in error message when GnuTLS fails
5030 to set them up, because the GnuTLS error message doesn't include the name
5031 of the failing file when there is a problem reading it.
5033 26. Expand error message when OpenSSL has problems setting up cert/key files.
5036 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5037 (this can affect the format of dates).
5039 28. exim_tidydb, when checking for the continued existence of a message for
5040 which it has found a message-specific retry record, was not finding
5041 messages that were in split spool directories. Consequently, it was
5042 deleting retry records that should have stayed in existence.
5044 29. eximstats updated to version 1.35
5045 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5046 1.35 - bugfix such that pie charts by volume are generated correctly
5048 30. The SPA authentication driver was not abandoning authentication and moving
5049 on to the next authenticator when an expansion was forced to fail,
5050 contradicting the general specification for all authenticators. Instead it
5051 was generating a temporary error. It now behaves as specified.
5053 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5054 (the order specifies the preference for clients). The order is now AES256,
5055 AES128, 3DES, ARCFOUR128.
5057 31. Small patch to Sieve code - explicitly set From: when generating an
5060 32. Exim crashed if a remote delivery caused a very long error message to be
5061 recorded - for instance if somebody sent an entire SpamAssassin report back
5062 as a large number of 550 error lines. This bug was coincidentally fixed by
5063 increasing the size of one of Exim's internal buffers (big_buffer) that
5064 happened as part of the Exiscan merge. However, to be on the safe side, I
5065 have made the code more robust (and fixed the comments that describe what
5068 33. Some experimental protocols are using DNS PTR records for new purposes. The
5069 keys for these records are domain names, not reversed IP addresses. The
5070 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5071 leaves it alone. Component reversal etc. now happens only for IP addresses.
5074 34. The host_aton() function is supposed to be passed a string that is known
5075 to be a valid IP address. However, in the case of IPv6 addresses, it was
5076 not checking this. This is a hostage to fortune. Exim now panics and dies
5077 if the condition is not met. A case was found where this could be provoked
5078 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5079 components; fortuitously, this particular loophole had already been fixed
5080 by change 4.50/55 or 4.44/33 above.
5082 If there are any other similar loopholes, the new check in host_aton()
5083 itself should stop them being exploited. The report I received stated that
5084 data on the command line could provoke the exploit when Exim was running as
5085 exim, but did not say which command line option was involved. All I could
5086 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5087 running as the user.
5090 35. There was a buffer overflow vulnerability in the SPA authentication code
5091 (which came originally from the Samba project). I have added a test to the
5092 spa_base64_to_bits() function which I hope fixes it.
5095 36. The daemon start-up calls getloadavg() while still root for those OS that
5096 need the first call to be done as root, but it missed one case: when
5097 deliver_queue_load_max is set with deliver_drop_privilege. This is
5098 necessary for the benefit of the queue runner, because there is no re-exec
5099 when deliver_drop_privilege is set.
5101 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5102 was in use, was not putting the data itself into the right store pool;
5103 consequently, it could be overwritten for a subsequent message in the same
5104 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5107 38. Sometimes the final signoff response after QUIT could fail to get
5108 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5109 < 0 before doing a fflush(). This bug looks as though it goes back to the
5110 introduction of TLS in release 3.20, but "sometimes" must have been rare
5111 because the tests only now provoked it.
5117 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5118 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5119 Now it returns an integer. A number of calls were still expecting a T/F
5120 return. Fortuitously, in all cases, the tests worked in OK situations,
5121 which is the norm. However, things would have gone wrong on any write
5122 failures on the smtp file descriptor. This function is used when sending
5123 messages over SMTP and also when doing verify callouts.
5125 2. When Exim is called to do synchronous delivery of a locally submitted
5126 message (the -odf or -odi options), it no longer closes stderr before doing
5129 3. Implemented the mua_wrapper option.
5131 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5133 5. Implemented the functions header_remove(), header_testname(),
5134 header_add_at_position(), and receive_remove_recipient(), and exported them
5137 6. If an ACL "warn" statement specified the addition of headers, Exim already
5138 inserted X-ACL-Warn: at the start if there was no header name. However, it
5139 was not making this test for the second and subsequent header lines if
5140 there were newlines in the string. This meant that an invalid header could
5141 be inserted if Exim was badly configured.
5143 7. Allow an ACL "warn" statement to add header lines at the start or after all
5144 the Received: headers, as well as at the end.
5146 8. Added the rcpt_4xx retry error code.
5148 9. Added postmaster_mailfrom=xxx to callout verification option.
5150 10. Added mailfrom=xxxx to the callout verification option, for verify=
5153 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5154 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5156 12. Inserted some casts to stop certain compilers warning when using pointer
5157 differences as field lengths or precisions in printf-type calls (mostly
5158 affecting debugging statements).
5160 13. Added optional readline() support for -be (dynamically loaded).
5162 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5163 same clock tick as a message's arrival, so that its received time was the
5164 same as the "first fail" time on the retry record, and that message
5165 remained on the queue past the ultimate address timeout, every queue runner
5166 would try a delivery (because it was past the ultimate address timeout) but
5167 after another failure, the ultimate address timeout, which should have then
5168 bounced the address, did not kick in. This was a "< instead of <=" error;
5169 in most cases the first failure would have been in the next clock tick
5170 after the received time, and all would be well.
5172 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5173 being recognized when the domain list was tested by the match_domain
5174 condition in an expansion string.
5176 16. Added the ${str2b64: operator.
5178 17. Exim was always calling setrlimit() to set a large limit for the number of
5179 processes, without checking whether the existing limit was already
5180 adequate. (It did check for the limit on file descriptors.) Furthermore,
5181 errors from getrlimit() and setrlimit() were being ignored. Now they are
5182 logged to the main and panic logs, but Exim does carry on, to try to do its
5183 job under whatever limits there are.
5185 18. Imported PCRE 5.0.
5187 19. Trivial typo in log message " temporarily refused connection" (the leading
5190 20. If the log selector return_path_on_delivery was set and an address was
5191 redirected to /dev/null, the delivery process crashed because it assumed
5192 that a return path would always be set for a "successful" delivery. In this
5193 case, the whole delivery is bypassed as an optimization, and therefore no
5196 21. Internal re-arrangement: the function for sending a challenge and reading
5197 a response while authentication was assuming a zero-terminated challenge
5198 string. It's now changed to take a pointer and a length, to allow for
5199 binary data in such strings.
5201 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5203 23. Exim was not respecting finduser_retries when seeking the login of the
5204 uid under which it was called; it was always trying 10 times. (The default
5205 setting of finduser_retries is zero.) Also, it was sleeping after the final
5206 failure, which is pointless.
5208 24. Implemented tls_on_connect_ports.
5210 25. Implemented acl_smtp_predata.
5212 26. If the domain in control=submission is set empty, Exim assumes that the
5213 authenticated id is a complete email address when it generates From: or
5214 Sender: header lines.
5216 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5217 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5218 chown and chgrp in /bin and hostname in /usr/bin.
5220 28. Exim was keeping the "process log" file open after each use, just as it
5221 does for the main log. This opens the possibility of it remaining open for
5222 long periods when the USR1 signal hits a daemon. Occasional processlog
5223 errors were reported, that could have been caused by this. Anyway, it seems
5224 much more sensible not to leave this file open at all, so that is what now
5227 29. The long-running daemon process does not normally write to the log once it
5228 has entered its main loop, and it closes the log before doing so. This is
5229 so that log files can straightforwardly be renamed and moved. However,
5230 there are a couple of unusual error situations where the daemon does write
5231 log entries, and I had neglected to close the log afterwards.
5233 30. The text of an SMTP error response that was received during a remote
5234 delivery was being truncated at 512 bytes. This is too short for some of
5235 the long messages that one sometimes sees. I've increased the limit to
5238 31. It is now possible to make retry rules that apply only when a message has a
5239 specific sender, in particular, an empty sender.
5241 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5242 it possible to be selective about when SMTP synchronization is enforced.
5244 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5246 32. Implemented hosts_connection_nolog.
5248 33. Added an ACL for QUIT.
5250 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5253 35. Added mailbox_size and mailbox_filecount to appendfile.
5255 36. Added control = no_multiline_responses to ACLs.
5257 37. There was a bug in the logic of the code that waits for the clock to tick
5258 in the case where the clock went backwards by a substantial amount such
5259 that the microsecond fraction of "now" was more than the microsecond
5260 fraction of "then" (but the whole seconds number was less).
5262 38. Added support for the libradius Radius client library this is found on
5263 FreeBSD (previously only the radiusclient library was supported).
5269 1. When certain lookups returned multiple values in the form name=value, the
5270 quoting of the values was not always being done properly. Specifically:
5271 (a) If the value started with a double quote, but contained no whitespace,
5273 (b) If the value contained whitespace other than a space character (i.e.
5274 tabs or newlines or carriage returns) it was not quoted.
5275 This fix has been applied to the mysql and pgsql lookups by writing a
5276 separate quoting function and calling it from the lookup code. The fix
5277 should probably also be applied to nisplus, ibase and oracle lookups, but
5278 since I cannot test any of those, I have not disturbed their existing code.
5280 2. A hit in the callout cache for a specific address caused a log line with no
5281 reason for rejecting RCPT. Now it says "Previous (cached) callout
5282 verification failure".
5284 3. There was an off-by-one bug in the queryprogram router. An over-long
5285 return line was truncated at 256 instead of 255 characters, thereby
5286 overflowing its buffer with the terminating zero. As well as fixing this, I
5287 have increased the buffer size to 1024 (and made a note to document this).
5289 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5290 when Exim is waiting for an SMTP response from a remote server, Exim
5291 restarts its select() call on the socket, thereby resetting its timeout.
5292 This is not a problem when such interrupts are rare. Somebody set up a cron
5293 job to run exiwhat every 2 minutes, which is less than the normal select()
5294 timeout (5 or 10 minutes). This meant that the select() timeout never
5295 kicked in because it was always reset. I have fixed this by comparing the
5296 time when an interrupt arrives with the time at the start of the first call
5297 to select(). If more time than the timeout has elapsed, the interrupt is
5298 treated as a timeout.
5300 5. Some internal re-factoring in preparation for the addition of Sieve
5301 extensions (by MH). In particular, the "personal" test is moved to a
5302 separate function, and given an option for scanning Cc: and Bcc: (which is
5303 not set for Exim filters).
5305 6. When Exim created an email address using the login of the caller as the
5306 local part (e.g. when creating a From: or Sender: header line), it was not
5307 quoting the local part when it contained special characters such as @.
5309 7. Installed new OpenBSD configuration files.
5311 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5312 try to make them clearer.
5314 9. Callout options, other than the timeout value, were being ignored when
5315 verifying sender addresses in header lines. For example, when using
5317 verify = header_sender/callout=no_cache
5319 the cache was (incorrectly) being used.
5321 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5322 only the Cygwin environment.
5324 11. When return_path_on_delivery was set as a log selector, if different remote
5325 addresses in the same message used different return paths and parallel
5326 remote delivery occurred, the wrong values would sometimes be logged.
5327 (Whenever a remote delivery process finished, the return path value from
5328 the most recently started remote delivery process was logged.)
5330 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5331 lines when AUTH and/or TLS are in use. This is the "received protocol"
5332 field. Exim used to use "asmtp" for authenticated SMTP, without any
5333 indication (in the protocol name) for TLS use. Now it follows the RFC and
5334 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5335 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5336 names appear in log lines as well as in Received: header lines.
5338 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5339 extensions, and comparison tests, and to fix some bugs.
5341 14. Changes to the "personal" filter test:
5343 (1) The test was buggy in that it was just doing the equivalent of
5344 "contains" tests on header lines. For example, if a user's address was
5345 anne@some.where, the "personal" test would incorrectly be true for
5347 To: susanne@some.where
5349 This test is now done by extracting each address from the header in turn,
5350 and checking the entire address. Other tests that are part of "personal"
5351 are now done using regular expressions (for example, to check local parts
5352 of addresses in From: header lines).
5354 (2) The list of non-personal local parts in From: addresses has been
5355 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5356 taken from the Sieve specification recommendations.
5358 (3) If the message contains any header line starting with "List-" it is
5359 treated as non-personal.
5361 (4) The test for "circular" in the Subject: header line has been removed
5362 because it now seems ill-conceived.
5364 15. Minor typos in src/EDITME comments corrected.
5366 16. Installed latest exipick from John Jetmore.
5368 17. If headers_add on a router specified a text string that was too long for
5369 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5370 of string_sprintf() is now avoided.
5372 18. $message_body_size was not set (it was always zero) when running the DATA
5373 ACL and the local_scan() function.
5375 19. For the "mail" command in an Exim filter, no default was being set for
5376 the once_repeat time, causing a random time value to be used if "once" was
5377 specified. (If the value happened to be <= 0, no repeat happened.) The
5378 default is now 0s, meaning "never repeat". The "vacation" command was OK
5379 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5380 (I found it when inspecting the code).
5382 20. There is now an overall timeout for performing a callout verification. It
5383 defaults to 4 times the callout timeout, which applies to individual SMTP
5384 commands during the callout. The overall timeout applies when there is more
5385 than one host that can be tried. The timeout is checked before trying the
5386 next host. This prevents very long delays if there are a large number of
5387 hosts and all are timing out (e.g. when the network connections are timing
5388 out). The value of the overall timeout can be changed by specifying an
5389 additional sub-option for "callout", called "maxwait". For example:
5391 verify = sender/callout=5s,maxwait=20s
5393 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5394 to the end before writing, but this should make it even safer).
5396 22. Exim was forgetting that it had advertised PIPELINING for the second and
5397 subsequent messages on an SMTP connection. It was also not resetting its
5398 memory on STARTTLS and an internal HELO.
5400 23. When Exim logs an SMTP synchronization error within a session, it now
5401 records whether PIPELINING has been advertised or not.
5403 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5404 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5405 rather than long int.
5407 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5409 26. Added the never_mail option to autoreply.
5415 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5416 crash if the getsockname() call failed; this can happen if a connection is
5417 closed very soon after it is established. The problem was simply in the
5418 order in which certain operations were done, causing Exim to try to write
5419 to the SMTP stream before it had set up the file descriptor. The bug has
5420 been fixed by making things happen in the correct order.
5426 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5427 before the connection was closed, thus losing the rejection response.
5429 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5430 some early Solaris releases, but causes trouble in current releases where
5431 socklen_t is defined.
5433 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5436 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5439 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5440 that releasing the top of it at the end releases what was used for sub-
5441 expansions (unless the block got too big). However, discard this block if
5442 the first thing is a variable or header, so that we can use its block when
5443 it is dynamic (useful for very large $message_headers, for example).
5445 6. Lookups now cache *every* query, not just the most recent. A new, separate
5446 store pool is used for this. It can be recovered when all lookup caches are
5447 flushed. Lookups now release memory at the end of their result strings.
5448 This has involved some general refactoring of the lookup sources.
5450 7. Some code has been added to the store_xxx() functions to reduce the amount
5451 of flapping under certain conditions.
5453 8. log_incoming_interface used to affect only the <= reception log lines. Now
5454 it causes the local interface and port to be added to several more SMTP log
5455 lines, for example "SMTP connection from", and rejection lines.
5457 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5459 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5461 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5462 could be overwritten at the end of the current message (or the start of a
5463 new message if it was set in a HELO ACL). The value is now preserved for
5464 the duration of the SMTP connection.
5466 12. If a transport had a headers_rewrite setting, and a matching header line
5467 contained an unqualified address, that address was qualified, even if it
5468 did not match any rewriting rules. The underlying bug was that the values
5469 of the flags that permit the existence of unqualified sender and recipient
5470 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5471 non-local messages, and by -bnq for local messages) were not being
5472 preserved with the message after it was received.
5474 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5475 "next input=" as part of the text comprising the host identity instead of
5476 the correct text. The code was using the same buffer for two different
5477 strings. However, depending on which order the printing function evaluated
5478 its arguments, the bug did not always show up. Under Linux, for example, my
5479 test suite worked just fine.
5481 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5482 doesn't work with some older versions of Perl. It has been changed to "my",
5483 which in any case is probably the better facility to use.
5485 15. A really picky compiler found some instances of statements for creating
5486 error messages that either had too many or two few arguments for the format
5489 16. The size of the buffer for calls to the DNS resolver has been increased
5490 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5491 for addresses that have a lot of PTR records. This alleviates a problem; it
5492 does not fully solve it.
5494 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5495 buffer now truncates the list and logs the incident, which is the same
5496 action as happens when Exim is looking up a host name and its aliases.
5497 Previously in this situation something unpredictable would happen;
5498 sometimes it was "internal error: store_reset failed".
5500 18. If a server dropped the connection unexpectedly when an Exim client was
5501 using GnuTLS and trying to read a response, the client delivery process
5502 crashed while trying to generate an error log message.
5504 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5505 string, for example:
5507 warn message = H1: something\nH2: something
5509 the text was added as a single header line from Exim's point of view
5510 though it ended up OK in the delivered message. However, searching for the
5511 second and subsequent header lines using $h_h2: did not work. This has been
5512 fixed. Similarly, if a system filter added multiple headers in this way,
5513 the routers could not see them.
5515 20. Expanded the error message when iplsearch is called with an invalid key to
5516 suggest using net-iplsearch in a host list.
5518 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5519 ACLs are no longer actually imposed (and a message to that effect is
5522 22. If a "gecos" field in a passwd entry contained escaped characters, in
5523 particular, if it contained a \" sequence, Exim got it wrong when building
5524 a From: or a Sender: header from that name. A second bug also caused
5525 incorrect handling when an unquoted " was present following a character
5526 that needed quoting.
5528 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5529 was not being matched caselessly.
5531 24. Arranged for all hyphens in the exim.8 source to be escaped with
5534 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5535 properly. Recipient callout cache records were still being keyed to include
5536 the sender, even when use_sender was set false. This led to far more
5537 callouts that were necessary. The sender is no longer included in the key
5538 when use_sender is false.
5540 26. Added "control = submission" modifier to ACLs.
5542 27. Added the ${base62d: operator to decode base 62 numbers.
5544 28. dnsdb lookups can now access SRV records.
5546 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5547 the configuration file.
5549 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5550 (-v) mode. This makes the output for a verbose queue run more intelligible.
5552 31. Added a use_postmaster feature to recipient callouts.
5554 32. Added the $body_zerocount variable, containing the number of binary zero
5555 bytes in the message body.
5557 33. The time of last modification of the "new" subdirectory is now used as the
5558 "mailbox time last read" when there is a quota error for a maildir
5561 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5563 35. Added +ignore_unknown as a special item in host lists.
5565 36. Code for decoding IPv6 addresses in host lists is now included, even if
5566 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5567 address was recognized as an IP address, but was then not correctly decoded
5568 into binary, causing unexpected and incorrect effects when compared with
5575 1. Very minor rewording of debugging text in manualroute to say "list of
5576 hosts" instead of "hostlist".
5578 2. If verify=header_syntax was set, and a header line with an unqualified
5579 address (no domain) and a large number of spaces between the end of the
5580 name and the colon was received, the reception process suffered a buffer
5581 overflow, and (when I tested it) crashed. This was caused by some obsolete
5582 code that should have been removed. The fix is to remove it!
5584 3. When running in the test harness, delay a bit after writing a bounce
5585 message to get a bit more predictability in the log output.
5587 4. Added a call to search_tidyup() just before forking a reception process. In
5588 theory, someone could use a lookup in the expansion of smtp_accept_max_
5589 per_host which, without the tidyup, could leave open a database connection.
5591 5. Added the variables $recipient_data and $sender_data which get set from a
5592 lookup success in an ACL "recipients" or "senders" condition, or a router
5593 "senders" option, similar to $domain_data and $local_part_data.
5595 6. Moved the writing of debug_print from before to after the "senders" test
5598 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5599 problems for message scanning, either using a data ACL, or using
5600 local_scan() because the Received: header was not generated till after they
5601 were called (in order to set the time as the time of reception completion).
5602 I have revised the way this works. The header is now generated after the
5603 body is received, but before the ACL or local_scan() are called. After they
5604 are run, the timestamp in the header is updated.
5610 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5611 before starting a queue runner without re-exec. This happened only when
5612 deliver_drop_privilege was set or when the Exim user was set to root. The
5613 effect of the bug was that timeouts during subsequent deliveries caused
5614 crashes instead of being properly handled. The handler is now left at its
5615 default (and expected) setting.
5617 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5618 message, again when deliver_drop_privilege is set or Exim is run as root.
5619 The bug described in (1) was not present in this case, but the tidying up
5620 of the other signals was missing. I have made the two cases consistent.
5622 3. The ignore_target_hosts setting on a manualroute router was being ignored
5623 for hosts that were looked up using the /MX notation.
5625 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5628 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5629 operated on the sender address. After changing the $sender_address to <>
5630 for the sender address verify, Exim was re-instated it as the original
5631 (before rewriting) address, but remembering that it had rewritten it, so it
5632 wasn't rewriting it again. This bug also had the effect of breaking the
5633 sender address verification caching when the sender address was rewritten.
5635 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5636 This has been changed so that if the ip literal address matches
5637 ignore_target_hosts, the router declines.
5639 7. Added expansion conditions match_domain, match_address, and match_local_
5640 part (NOT match_host).
5642 8. The placeholder for the Received: header didn't have a length field set.
5644 9. Added code to Exim itself and to exim_lock to test for a specific race
5645 condition that could lead to file corruption when using MBX delivery. The
5646 issue is with the lockfile that is created in /tmp. If this file is removed
5647 after a process has opened it but before that process has acquired a lock,
5648 there is the potential for a second process to recreate the file and also
5649 acquire a lock. This could lead to two Exim processes writing to the file
5650 at the same time. The added code performs the same test as UW imapd; it
5651 checks after acquiring the lock that its file descriptor still refers to
5652 the same named file.
5654 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5655 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5658 11. Added the smtp_active_hostname option. If used, this will typically be made
5659 to depend on the incoming interface address. Because $interface_address is
5660 not set up until the daemon has forked a reception process, error responses
5661 that can happen earlier (such as "too many connections") no longer contain
5664 12. If an expansion in a condition on a "warn" statement fails because a lookup
5665 defers, the "warn" statement is abandoned, and the next ACL statement is
5666 processed. Previously this caused the whole ACL to be aborted.
5668 13. Added the iplsearch lookup type.
5670 14. Added ident_timeout as a log selector.
5672 15. Added tls_certificate_verified as a log selector.
5674 16. Added a global option tls_require_ciphers (compare the smtp transport
5675 option of the same name). This controls incoming TLS connections.
5677 17. I finally figured out how to make tls_require_ciphers do a similar thing
5678 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5679 before starting the TLS session.
5681 18. Tabs are now shown as \t in -bP output.
5683 19. If the log selector return_path_on_delivery was set, Exim crashed when
5684 bouncing a message because it had too many Received: header lines.
5686 20. If two routers both had headers_remove settings, and the first one included
5687 a superfluous trailing colon, the final name in the first list and the
5688 first name in the second list were incorrectly joined into one item (with a
5689 colon in the middle).
5695 1. Added -C and -D options to the exinext utility, mainly to make it easier
5696 to include in the automated testing, but these could be helpful when
5697 multiple configurations are in use.
5699 2. The exinext utility was not formatting the output nicely when there was
5700 an alternate port involved in the retry record key, nor when there was a
5701 message id as well (for retries that were specific to a specific message
5702 and a specific host). It was also confused by IPv6 addresses, because of
5703 the additional colons they contain. I have fixed the IPv4 problem, and
5704 patched it up to do a reasonable job for IPv6.
5706 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5707 delivery, the log line now contains "pipelined" if PIPELINING was used.
5709 4. An SMTP transport process used to panic and die if the bind() call to set
5710 an explicit outgoing interface failed. This has been changed; it is now
5711 treated in the same way as a connect() failure.
5713 5. A reference to $sender_host_name in the part of a conditional expansion
5714 that was being skipped was still causing a DNS lookup. This no longer
5717 6. The def: expansion condition was not recognizing references to header lines
5718 that used bh_ and bheader_.
5720 7. Added the _cache feature to named lists.
5722 8. The code for checking quota_filecount in the appendfile transport was
5723 allowing one more file than it should have been.
5725 9. For compatibility with Sendmail, the command line option
5733 and sets the incoming protocol and host name (for trusted callers). The
5734 host name and its colon can be omitted when only the protocol is to be set.
5735 Note the Exim already has two private options, -pd and -ps, that refer to
5736 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5737 "s", but I don't think that's a major issue.
5739 10. A number of refactoring changes to the code, none of which should affect
5742 (a) The number of logging options was getting close to filling up the
5743 32-bit word that was used as a bit map. I have split them into two classes:
5744 those that are passed in the argument to log_write(), and those that are
5745 only ever tested independently outside of that function. These are now in
5746 separate 32-bit words, so there is plenty of room for expansion again.
5747 There is no change in the user interface or the logging behaviour.
5749 (b) When building, for example, log lines, the code previously used a
5750 macro that called string_cat() twice, in order to add two strings. This is
5751 not really sufficiently general. Furthermore, there was one instance where
5752 it was actually wrong because one of the argument was used twice, and in
5753 one call a function was used. (As it happened, calling the function twice
5754 did not affect the overall behaviour.) The macro has been replaced by a
5755 function that can join an arbitrary number of extra strings onto a growing
5758 (c) The code for expansion conditions now uses a table and a binary chop
5759 instead of a serial search (which was left over from when there were very
5760 few conditions). Also, it now recognizes conditions like "pam" even when
5761 the relevant support is not compiled in: a suitably worded error message is
5762 given if an attempt is made to use such a condition.
5764 11. Added ${time_interval:xxxxx}.
5766 12. A bug was causing one of the ddress fields not to be passed back correctly
5767 from remote delivery subprocesses. The field in question was not being
5768 subsequently used, so this caused to problems in practice.
5770 13. Added new log selectors queue_time and deliver_time.
5772 14. Might have fixed a bug in maildirsizefile handling that threw up
5773 "unexpected character" debug warnings, and recalculated the data
5774 unnecessarily. In any case, I expanded the warning message to give more
5777 15. Added the message "Restricted characters in address" to the statements in
5778 the default ACL that block characters like @ and % in local parts.
5780 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5781 Three changes have been made:
5783 (a) There was a serious bug; a negative response to MAIL caused the whole
5784 recipient domain to be cached as invalid, thereby blocking all messages
5785 to all local parts at the same domain, from all senders. This bug has
5786 been fixed. The domain is no longer cached after a negative response to
5787 MAIL if the sender used is not empty.
5789 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5792 (c) A new callout option, "use_sender" has been added for people who want
5793 the modified behaviour.
5799 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5802 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5803 indeed breaks things for older releases.
5805 3. Added additional logging to the case where there is a problem reading data
5806 from a filter that is running in a subprocess using a pipe, in order to
5807 try to track down a specific problem.
5809 4. Testing facility fudge: when running in the test harness and attempting
5810 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5811 getting "No route to host". Convert this to a timeout.
5813 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5816 6. Some OS don't have socklen_t but use size_t instead. This affects the
5817 fifth argument of getsockopt() amongst other things. This is now
5818 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5819 can be set for individual OS. I have set it for SunOS5, OSF1, and
5820 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5821 some earlier ones do not.
5823 7. Change 4.30/15 was not doing the test caselessly.
5825 8. The standard form for an IPv6 address literal was being rejected by address
5826 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5827 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5828 this, as well as the form without the "IPv6" on the front (but only when
5829 address literals are enabled, of course).
5831 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5833 10. Exim crashed if a message with an empty sender address specified by -f
5834 encountered a router with an errors_to setting. This could be provoked only
5835 by a command such as
5839 where an empty string was supplied; "<>" did not hit this bug.
5841 11. Installed PCRE release 4.5.
5843 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5844 remained set. It is now erased.
5846 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5847 times from message ids (which are base 36 rather than the normal 62).
5849 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5850 were being counted as actual protocol errors, and logged if the log
5851 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5852 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5853 and DATA following a set of rejected RCPTs do not count as protocol errors.
5854 In other words, Exim assumes they were pipelined, though this may not
5855 actually be the case. Of course, in all cases the client gets an
5856 appropriate error code.
5858 15. If a lookup fails in an ACL condition, a message about the failure may
5859 be available; it is used if testing the ACL cannot continue, because most
5860 such messages specify what the cause of the deferral is. However, some
5861 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5862 that caused an old message to be retained and used if a later statement
5863 caused a defer, replacing the real cause of the deferral.
5865 16. If an IP address had so many PTR records that the DNS lookup buffer
5866 was not large enough to hold them, Exim could crash while trying to process
5867 the truncated data. It now detects and logs this case.
5869 17. Further to 4.21/58, another change has been made: if (and only if) the
5870 first line of a message (the first header line) ends with CRLF, a bare LF
5871 in a subsequent header line has a space inserted after it, so as not to
5872 terminate the header.
5874 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5875 unnecessarily, used atoi() instead of strtol(), and didn't check the
5876 termination when getting file sizes from file names by regex.
5878 19. Completely re-implemented the support for maildirsize files, in the light
5879 of a number of problems with the previous contributed implementation
5880 (4.30/29). In particular:
5882 . If the quota is zero, the maildirsize file is maintained, but no quota is
5885 . If the maildir directory does not exist, it is created before any attempt
5886 to write a maildirsize file.
5888 . The quota value in the file is just a cache; if the quota is changed in
5889 the transport, the new value overrides.
5891 . A regular expression is available for excluding directories from the
5894 20. The autoreply transport checks the characters in options that define the
5895 message's headers; it allows continued headers, but it was checking with
5896 isspace() after an embedded newline instead of explicitly looking for a
5899 21. If all the "regular" hosts to which an address was routed had passed their
5900 expiry times, and had not reached their retry times, the address was
5901 bounced, even if fallback hosts were defined. Now Exim should go on to try
5904 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5905 equivalent code in the SMTP transport. Some hosts send humungous responses
5906 to HELO/EHLO, more than 1024 it seems.
5908 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5909 gives compiler warnings in some environments. I've now done it "properly",
5912 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5913 (because of problems with the built-in one) was declared to return uschar *
5914 instead of char *, causing compiler failure.
5916 25. Fixed a file descriptor leak when processing alias/forward files.
5918 26. Fixed a minor format string issue in dbfn.c.
5920 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5922 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5923 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5926 29. When a sender address is verified, it is cached, to save repeating the test
5927 when there is more than one recipient in a message. However, when the
5928 verification involves a callout, it is possible for different callout
5929 options to be set for different recipients. It is too complicated to keep
5930 track of this in the cache, so now Exim always runs a verification when a
5931 callout is required, relying on the callout cache for the optimization.
5932 The overhead is duplication of the address routing, but this should not be
5935 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5936 to be verified with callout=postmaster, and the main callout worked but the
5937 postmaster check failed, the verification correctly failed. However, if a
5938 subsequent RCPT command asked for sender verification *without* the
5939 postmaster check, incorrect caching caused this verification also to fail,
5942 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5943 it was not caching the DNS options (qualify_single, search_parents) that
5944 were used when the lookup failed. A subsequent lookup with different
5945 options therefore always gave the same answer, though there were cases
5946 where it should not have. (Example: a "domains = !$mx_any" option on a
5947 dnslookup router: the "domains" option is always processed without any
5948 widening, but the router might have qualify_single set.) Now Exim uses the
5949 cached value only when the same options are set.
5951 32. Added John Jetmore's "exipick" utility to the distribution.
5953 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5954 than a timeout (e.g. a certificate is required, and is not provided), an
5955 Exim server now closes the connection immediately. Previously it waited for
5956 the client to close - but if the client is SSL, it seems that they each
5957 wait for each other, leading to a delay before one of them times out.
5959 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5960 maintained 0.8.x compatibility because I don't think many are using it, and
5961 it is clearly obsolete.
5963 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5966 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5967 client certificate was expired. A simple patch fixes this, though I don't
5968 understand the full logic of why the verify callback is called multiple
5971 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5972 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5973 which causes problems with some clients (such as the Certicom SSL Plus
5974 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5975 disables the coutermeasure allowing Eudora to connect."
5977 38. Exim was not checking that a write() to a log file succeeded. This could
5978 lead to Bad Things if a log got too big, in particular if it hit a file
5979 size limit. Exim now panics and dies if it cannot write to a log file, just
5980 as it does if it cannot open a log file.
5982 39. Modified OS/Makefile-Linux so that it now contains
5984 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5986 The two -D definitions ensure that Exim is compiled with large file
5987 support, which makes it possible to handle log files that are bigger than
5990 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5991 instance) a domain was checked against a named list that involved a lookup,
5992 causing $domain_data to be set, then another domain was checked against the
5993 same list, then the first domain was re-checked, the value of $domain_data
5994 after the final check could be wrong. In particular, if the second check
5995 failed, it could be set empty. This bug probably also applied to
5998 41. The strip_trailing_dot option was not being applied to the address given
5999 with the -f command-line option.
6001 42. The code for reading a message's header from the spool was incrementing
6002 $received_count, but never initializing it. This meant that the value was
6003 incorrect (doubled) while delivering a message in the same process in which
6004 it was received. In the most common configuration of Exim, this never
6005 happens - a fresh exec is done - but it can happen when
6006 deliver_drop_privilege is set.
6008 43. When Exim logs an SMTP synchronization error - client data sent too soon -
6009 it now includes up to 150 characters of the unexpected data in the log
6012 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
6013 and building data strings. The size of both of these buffers was 10 000
6014 bytes - far larger than anybody would *ever* want, thought I. Needless to
6015 say, somebody hit the limit. I have increased the maximum line length to
6016 20 000 and the maximum data length of concatenated lines to 100 000. I have
6017 also fixed two bugs, because there was no checking on these buffers. Tsk,
6018 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
6019 buffer is too small.
6021 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6022 lsearch lookups. Now it does.
6024 46. When parsing a route_list item in a manualroute router, a fixed-length
6025 buffer was used for the list of hosts. I made this 1024 bytes long,
6026 thinking that nobody would ever have a list of hosts that long. Wrong.
6027 Somebody had a whole pile of complicated expansion conditions, and the
6028 string was silently truncated, leading to an expansion error. It turns out
6029 that it is easier to change to an unlimited length (owing to other changes
6030 that have happened since this code was originally written) than to build
6031 structure for giving a limitation error. The length of the item that
6032 expands into the list of hosts is now unlimited.
6034 47. The lsearch lookup could not handle data where the length of text line was
6035 more than 4095 characters. Such lines were truncated, leading to shortened
6036 data being returned. It should now handle lines of any length.
6038 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6039 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6042 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6043 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6044 no longer happens in comment lines. A list of replacements is now placed
6045 at the head of all of the source files, except those whose only change is
6046 to replace PERL_COMMAND in the very first #! line.
6048 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6049 messages on the queue, with a bottom-up merge sort, using code contributed
6050 by Michael Haardt. This should make operations like -bp somewhat faster on
6051 large queues. It won't affect queue runners, except when queue_run_in_order
6054 51. Installed eximstats 1.31 in the distribution.
6056 52. Added support for SRV lookups to the dnslookup router.
6058 53. If an ACL referred to $message_body or $message_body_end, the value was not
6059 reset for any messages that followed in the same SMTP session.
6061 54. The store-handling optimization for building very long strings was not
6062 differentiating between the different store pools. I don't think this
6063 actually made any difference in practice, but I've tidied it.
6065 55. While running the routers to verify a sender address, $sender_address
6066 was still set to the sender address. This is wrong, because when routing to
6067 send a bounce to the sender, it would be empty. Therefore, I have changed
6068 it so that, while verifying a sender address, $sender_address is set to <>.
6069 (There is no change to what happens when verifying a recipient address.)
6071 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6072 target A or AAAA records (if not already returned) without resetting the
6073 qualify_single or search_parents options of the DNS resolver. These are
6074 inappropriate in this case because the targets of MX and SRV records must
6075 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6076 target that, when qualified, matched something in the local domain. These
6077 two options are now turned off when doing these lookups.
6079 57. It seems that at least some releases of Reiserfs (which does not have the
6080 concept of a fixed number of inodes) returns zero and not -1 for the
6081 number of available inodes. This interacted badly with check_spool_inodes,
6082 which assumed that -1 was the "no such thing" setting. What I have done is
6083 to check that the total number of inodes is greater than zero before doing
6084 the test of how many are available.
6086 58. When a "warn" ACL statement has a log_message modifier, the message is
6087 remembered, and not repeated. This is to avoid a lot of repetition when a
6088 message has many recipients that cause the same warning to be written.
6089 However, Exim was preserving the list of already written lines for an
6090 entire SMTP session, which doesn't seem right. The memory is now reset if a
6091 new message is started.
6093 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6094 debugging output unless log_rewrite was also set.
6096 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6097 of (int)(handle) when we know that handle contains (void *)(-1).
6099 61. The Exim daemon panic-logs an error return when it closes the incoming
6100 connection. However "connection reset by peer" seems to be common, and
6101 isn't really an error worthy of noting specially, so that particular error
6104 62. When Exim is trying to find all the local interfaces, it used to panic and
6105 die if the ioctl to get the interface flags failed. However, it seems that
6106 on at least one OS (Solaris 9) it is possible to have an interface that is
6107 included in the list of interfaces, but for which you get a failure error
6108 for this call. This happens when the interface is not "plumbed" into a
6109 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6110 failure of the "get flags" call assumes that the interface is down.
6112 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6113 makes life easier for people who are doing arithmetic on fields extracted
6114 from dates, where you often get leading zeros that should not be
6115 interpreted as octal.
6117 64. Added qualify_domain to the redirect router, to override the global
6120 65. If a pathologically long header line contained very many addresses (the
6121 report of this problem mentioned 10 000) and each of them was rewritten,
6122 Exim could use up a very large amount of memory. (It kept on making new
6123 copies of the header line as it rewrote, and never released the old ones.)
6124 At the expense of a bit more processing, the header rewriting function has
6125 been changed so that it no longer eats memory in this way.
6127 66. The generation of the Received: header has been moved from the time that a
6128 message starts to be received, to the time that it finishes. The timestamp
6129 in the Received: header should now be very close to that of the <= log
6130 line. There are two side-effects of this change:
6132 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6133 logged header lines no longer include the local Received: line, because
6134 it has not yet been created. The same applies to a copy of the message
6135 that is returned to a non-SMTP sender when a message is rejected.
6137 (b) When a filter file is tested using -bf, no additional Received: header
6138 is added to the test message. After some thought, I decided that this
6141 This change does not affect the value of $received_for. It is still set
6142 after address rewriting, but before local_scan() is called.
6144 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6146 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6147 gave an unhelpful panic error message, and a defer error. I have managed to
6148 change this behaviour so that it now rejects any supplied certificate,
6149 which seems right, as the list of acceptable certificates is empty.
6151 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6152 gave an unhelpful defer error. I have not managed to make this reject any
6153 supplied certificates, but the error message it gives is "no certificate
6154 supplied", which is not helpful.
6156 70. exigrep's output now also includes lines that are not associated with any
6157 message, but which match the given pattern. Implemented by a patch from
6158 Martin Sluka, which also tidied up the Perl a bit.
6160 71. Recipient callout verification, like sender verification, was using <> in
6161 the MAIL FROM command. This isn't really the right thing, since the actual
6162 sender may affect whether the remote host accepts the recipient or not. I
6163 have changed it to use the actual sender in the callout; this means that
6164 the cache record is now keyed on a recipient/sender pair, not just the
6165 recipient address. There doesn't seem to be a real danger of callout loops,
6166 since a callout by the remote host to check the sender would use <>.
6167 [SEE ABOVE: changed after hitting problems.]
6169 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6170 temporary errors. However, in the case of such a code being given after
6171 the end of a data transmission (i.e. after ".") Exim was failing to write
6172 a retry record for the message. (Yes, there was some broken host that was
6173 actually sending 8xx at this point.)
6175 73. An unknown lookup type in a host list could cause Exim to panic-die when
6176 the list was checked. (An example that provoked this was putting <; in the
6177 middle of a list instead of at the start.) If this happened during a DATA
6178 ACL check, a -D file could be left lying around. This kind of configuration
6179 error no longer causes Exim to die; instead it causes a defer error. The
6180 incident is still logged to the main and panic logs.
6182 74. Buglet left over from Exim 3 conversion. The message "too many messages
6183 in one connection" was written to the rejectlog but not the mainlog, except
6184 when address rewriting (yes!) was being logged.
6186 75. Added write_rejectlog option.
6188 76. When a system filter was run not as root (that is, when system_filter_user
6189 was set), the values of the $n variables were not being returned to the
6190 main process; thus, they were not subsequently available in the $sn
6193 77. Added +return_path_on_delivery log selector.
6195 78. A connection timeout was being treated differently from recipients deferred
6196 when testing hosts_max_try with a message that was older than the host's
6197 retry timeout. (The host should not be counted, thus allowing all hosts to
6198 be tried at least once before bouncing.) This may have been the cause of an
6199 occasionally reported bug whereby a message would remain on the queue
6200 longer than the retry timeout, but would be bounced if a delivery was
6201 forced. I say "may" because I never totally pinned down the problem;
6202 setting up timeout/retry tests is difficult. See also the next item.
6204 79. The ultimate address timeout was not being applied to errors that involved
6205 a combination of host plus message (for example, a timeout on a MAIL
6206 command). When an address resolved to a number of possible hosts, and they
6207 were not all tried for each delivery (e.g. because of hosts_max_try), a
6208 message could remain on the queue longer than the retry timeout.
6210 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6213 81. Fixed an obscure SMTP outgoing bug which required at least the following
6214 conditions: (a) there was another message waiting for the same server;
6215 (b) the server returned 5xx to all RCPT commands in the first message so
6216 that the message was not completed; (c) the server dropped the connection
6217 or gave a negative response to the RSET that Exim sends to abort the
6218 transaction. The observed case was a dropped connection after DATA that had
6219 been sent in pipelining mode. That is, the server had advertised PIPELINING
6220 but was not implementing it correctly. The effect of the bug was incorrect
6221 behaviour, such as trying another host, and this could lead to a crash.
6227 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6228 and daemon.c were passed as pointers to ints; they should have been
6229 pointers to socklen_t variables (which are typically unsigned ints).
6231 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6234 3. Fixed a really odd bug that affected only the testing scheme; patching a
6235 certain fixed string in the binary changed the value of another string that
6236 happened to be identical to the end of the original first string.
6238 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6239 name", it returns that address as the IP address. On some operating
6240 systems (e.g. Solaris), it also passes back the IP address string as the
6241 "host name". However, on others (e.g. Linux), it passes back an empty
6242 string. Exim wasn't checking for this, and was changing the host name to an
6243 empty string, assuming it had been canonicalized.
6245 5. Although rare, it is permitted to have more than one PTR record for a given
6246 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6247 all the names associated with an address, because they do in Solaris.
6248 However, it seems that they do not in Linux for data that comes from the
6249 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6250 I found this out when I moved to a new Linux workstation and tried to run
6251 the Exim test suite.
6253 To get round this problem I have changed the code so that it now does its
6254 own call to the DNS to look up PTR records when searching for a host name.
6255 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6256 addresses that are only in /etc/hosts are still found.
6258 This behaviour is, however, controlled by an option called host_lookup_
6259 order, which defaults to "bydns:byaddr". If people want to use the other
6260 order, or indeed, just use one or the other means of lookup, they can
6261 specify it in this variable.
6263 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6264 some operating systems, this comes back from gethostbyaddr() as an empty
6265 string, and this is what Exim used to test for. However, it seems that in
6266 other systems, "." is yielded. Exim now tests for this case too.
6268 7. The values of check_spool_space and check_log_space are now held internally
6269 as a number of kilobytes instead of an absolute number of bytes. If a
6270 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6271 kilobyte. This means that much larger values can be stored.
6273 8. Exim monitor: an attempt to get the action menu when not actually pointing
6274 at a message produces an empty menu entitled "No message selected". This
6275 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6276 no entries in it ("Shell widget menu has zero width and/or height"). So I
6277 have added a single, blank menu entry in this case.
6279 9. Added ${quote_local_part.
6281 10. MIME decoding is now applied to the contents of Subject: header lines when
6284 11. Now that a reference to $sender_host_address automatically causes a reverse
6285 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6286 host lookup before query-style lookups in lists that might use this
6287 variable. This has therefore been abolished, and the "net-" prefix is no
6288 longer necessary for query-style lookups.
6290 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6291 have been a typo for LFLAGS, so it has been changed.
6293 13. The install script calls Exim with "-C /dev/null" in order to find the
6294 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6295 to be output. However, since Exim outputs its version number before the
6296 error, it didn't break the script. It just looked ugly. I fixed this by
6297 always allowing "-C /dev/null" if the caller is root.
6299 14. Ignore overlarge ACL variable number when reading spool file - insurance
6300 against a later release with more variables having written the file.
6302 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6303 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6304 this, as well as the form without the "IPv6" on the front.
6306 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6307 OS/Makefile-Darwin file.
6309 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6310 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6312 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6313 variable. However, some OS use ints for inodes. I've added cast to long int
6314 to get rid of the compiler warning.
6316 19. I had forgotten to lock out "/../" in configuration file names when
6317 ALT_CONFIG_PREFIX was set.
6319 20. Routers used for verification do not need to specify transports. However,
6320 if such a router generated a host list, and callout was configured, Exim
6321 crashed, because it could not find a port number from the (non-existent)
6322 transport. It now assumes port 25 in this circumstance.
6324 21. Added the -t option to exigrep.
6326 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6327 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6328 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6329 them is set, LOOKUP_LSEARCH is forced.
6331 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6333 24. Added sender and host information to the "rejected by local_scan()" log
6334 line; previously there was no indication of these.
6336 25. Added .include_if_exists.
6338 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6339 Linux. It turns out that not all file systems support this. Apparently some
6340 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6341 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6342 not supported on the file descriptor, is now ignored when Exim is trying to
6343 sync a directory. This applies only to Linux.
6345 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6347 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6348 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6351 29. Added support for maildirsize files from supplied patch (modified a bit).
6353 30. The use of :fail: followed by an empty string could lead Exim to respond to
6354 sender verification failures with (e.g.):
6356 550 Verification failed for <xxx>
6357 550 Sender verify failed
6359 where the first response line was missing the '-' that indicates it is not
6360 the final line of the response.
6362 31. The loop for finding the name of the user that called Exim had a hardwired
6363 limit of 10; it now uses the value of finduser_retries, which is used for
6364 all other user lookups.
6366 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6369 33. Exim was neglecting to zero errno before one call of strtol() when
6370 expanding a string and expecting an integer value. On some systems this
6371 resulted in spurious "integer overflow" errors. Also, it was casting the
6372 result into an int without checking.
6374 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6375 did not work. The code looks as if it has *never* worked, though it appears
6376 to have been documented since at least release 1.62. I have made it work.
6378 35. The "timeout_DNS" error in retry rules, also documented since at least
6379 1.62, also never worked. As it isn't clear exactly what this means, and
6380 clearly it isn't a major issue, I have abolished the feature by treating it
6381 as "timeout", and writing a warning to the main and panic logs.
6383 36. The display of retry rules for -brt wasn't always showing the error code
6386 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6387 timeout_connect_A, timeout_connect_MX.
6389 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6390 to the empty sender.
6392 39. The daemon was not analysing the content of -oX till after it had closed
6393 stderr and disconnected from the controlling terminal. This meant that any
6394 syntax errors were only noted on the panic log, and the return code from
6395 the command was 0. By re-arranging the code a little, I've made the
6396 decoding happen first, so such errors now appear on stderr, and the return
6397 code is 1. However, the actual setting up of the sockets still happens in
6398 the disconnected process, so errors there are still only recorded on the
6401 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6402 connections (as happens on some IP stacks) was logged at start up time as
6403 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6404 it from "IPv6 and IPv4", which means that two separate sockets are being
6407 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6408 says whether AF_INET or AF_INET6 was passed as an argument.
6410 42. Exiwhat output was messed up when time zones were included in log
6413 43. Exiwhat now gives more information about the daemon's listening ports,
6414 and whether -tls-on-connect was used.
6416 44. The "port" option of the smtp transport is now expanded.
6418 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6419 silently ignored. Now an error message is written to the main and panic
6422 46. There's a new ACL modifier called "logwrite" which writes to a log file
6423 as soon as it is encountered.
6425 47. Added $local_user_uid and $local_user_gid at routing time.
6427 48. Exim crashed when trying to verify a sender address that was being
6430 49. Exim was recognizing only a space character after ".include". It now also
6431 recognizes a tab character.
6433 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6434 extracting the relevant information from the specification. The man page no
6435 longer contains scrambled data for the -d option, and I've added a section
6436 at the front about calling Exim under different names.
6438 51. Added "extra_headers" argument to the "mail" command in filter files.
6440 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6443 53. Installed eximstats 1.29.
6445 54. Added transport_filter_timeout as a generic transport option.
6447 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6448 Cc: header lines. This was required by RFC 822, but it not required by RFC
6451 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6452 incoming messages that did not have them. Now it does so only if the
6453 message originates locally, that is, if there is no associated remote host
6454 address. When Resent- header lines are present, this applies to the Resent-
6455 lines rather than the non-Resent- lines.
6457 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6458 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6460 58. Messages for configuration errors now include the name of the main
6461 configuration file - useful now that there may be more than one file in a
6462 list (.included file names were always shown).
6464 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6465 for those rare installations that do not start the daemon as root or run it
6466 setuid root. I've cut out the call to initgroups() if the daemon is not
6469 60. The Exim user and group can now be bound into the binary as text strings
6470 that are looked up at the start of Exim's processing.
6472 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6474 62. Added $mailstore_basename variable.
6476 63. Installed patch to sieve.c from Michael Haardt.
6478 64. When Exim failed to open the panic log after failing to open the main log,
6479 the original message it was trying to log was written to stderr and debug
6480 output, but if they were not available (the usual case in production), it
6481 was lost. Now it is written to syslog before the two lines that record the
6482 failures to open the logs.
6484 65. Users' Exim filters run in subprocesses under the user's uid. It is
6485 possible for a "deliver" command or an alias in a "personal" command to
6486 provoke an address rewrite. If logging of address rewriting is configured,
6487 this fails because the process is not running as root or exim. There may be
6488 a better way of dealing with this, but for the moment (because 4.30 needs
6489 to be released), I have disabled address rewrite logging when running a
6490 filter in a non-root, non-exim process.
6496 1. The buildconfig auxiliary program wasn't quoting the value set for
6497 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6498 not defined. This bug was masked in 4.22 by the effect that was fixed in
6501 2. Some messages that were rejected after a message id was allocated were
6502 shown as "incomplete" by exigrep. It no longer does this for messages that
6503 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6505 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6506 have allow_domain_literals set, the ID did not get logged in the <= line.
6507 Domain literals are now always recognized in Message-ID: header lines.
6509 4. The first argument for a ${extract expansion item is the key name or field
6510 number. Leading and trailing spaces in this item were not being ignored,
6511 causing some misleading effects.
6513 5. When deliver_drop_privilege was set, single queue runner processes started
6514 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6515 same command in the process it spins off) were not dropping privilege.
6517 6. When the daemon running as "exim" started a queue runner, it always
6518 re-executed Exim in the spun-off process. This is a waste of effort when
6519 deliver_drop_privilege is set. The new process now just calls the
6520 queue-runner function directly.
6526 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6529 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6530 ignored. Though the use of -oP was forcing the writing of a pid file, it
6531 was always written to the default place.
6533 3. If the message "no IP address found for host xxxx" is generated during
6534 incoming verification, it is now followed by identification of the incoming
6535 connection (so you can more easily find what provoked it).
6537 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6539 5. Added some features to "harden" Exim a bit more against certain attacks:
6541 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6542 be put in Local/Makefile. This is like the never_users runtime option,
6543 but it cannot be overridden. The default setting is "root".
6545 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6546 prefix string with which any file named in a -C command line option
6549 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6550 is retained for -C and -D only if the caller of Exim is root. Without
6551 it, the exim user may also use -C and -D and retain privilege.
6553 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6554 command line option is disabled.
6556 6. Macro names set by the -D option must start with an upper case letter, just
6557 like macro names defined in the configuration file.
6559 7. Added "dereference=" facility to LDAP.
6561 8. Two instances of the typo "uknown" in the source files are fixed.
6563 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6564 the Configure-Makefile script screwed up while processing it.
6566 10. Incorporated PCRE 4.4.
6568 11. The SMTP synchronization check was not operating right at the start of an
6569 SMTP session. For example, it could not catch a HELO sent before the client
6570 waited for the greeting. There is now a check for outstanding input at the
6571 point when the greeting is written. Because of the duplex, asynchronous
6572 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6573 way, but not yet received, when the check is performed.
6575 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6576 on TCP/IP sockets, because this apparently causes some broken clients to
6579 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6580 unchanged) from the Cygwin maintainer.
6582 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6583 was supported instead of testing for mbx. Effectively a typo.
6585 15. The spa authenticator server code was not checking that the input it
6586 received was valid base64.
6588 16. The debug output line for the "set" modifier in ACLs was not showing the
6589 name of the variable that was being set.
6591 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6593 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6594 lookup on its own. Something else was needed to trigger the lookup. For
6595 example, a match in host_lookup or the need for a host name in a host list.
6596 Now, if $sender_host_name is referenced and the host name has not yet been
6597 looked up, a lookup is performed. If the lookup fails, the variable remains
6598 empty, and $host_lookup_failed is set to "1".
6600 19. Added "eqi" as a case-independent comparison operator.
6602 20. The saslauthd authentication condition could segfault if neither service
6603 nor realm was specified.
6605 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6606 error message that was logged was misleading, and incoming SMTP
6607 connections were dropped. The message is now more accurate, and temporary
6608 errors are given to SMTP connections.
6610 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6611 (except RSET) with a 421 error, until QUIT is received. However, it was
6612 failing to send a response to QUIT.
6614 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6615 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6616 ACL is now run after the helo_try_verify_hosts code.
6618 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6619 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6620 case-independent, so other case variants are also recognized). Apparently
6621 some systems use these upper case variants.
6623 25. If more than two messages were waiting for the same host, and a transport
6624 filter was specified for the transport, Exim sent two messages over the
6625 same TCP/IP connection, and then failed with "socket operation on non-
6626 socket" when it tried to send the third.
6628 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6630 27. The extern definition of crypt16() in expand.c was not being excluded when
6631 the OS had its own crypt16() function.
6633 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6634 as a preferred synonym for return_size_limit, both as an option and as an
6637 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6639 30. Changed the default configuration ACL to relax the local part checking rule
6640 for addresses that are not in any local domains. For these addresses,
6641 slashes and pipe symbols are allowed within local parts, but the sequence
6642 /../ is explicitly forbidden.
6644 31. SPA server authentication was not clearing the challenge buffer before
6647 32. log_message in a "warn" ACL statement was writing to the reject log as
6648 well as to the main log, which contradicts the documentation and doesn't
6649 seem right (because no rejection is happening). So I have stopped it.
6651 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6652 However, I am unable to do any testing of this.
6654 34. Fixed an infelicity in the appendfile transport. When checking directories
6655 for a mailbox, to see if any needed to be created, it was accidentally
6656 using path names with one or more superfluous leading slashes; tracing
6657 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6659 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6660 contents are added to the log line that is written for every discarded
6661 recipient. (Previously a log_message setting was ignored.)
6663 36. The ${quote: operator now quotes the string if it is empty.
6665 37. The install script runs exim in order to find its version number. If for
6666 some reason other than non-existence or emptiness, which it checks, it
6667 could not run './exim', it was installing it with an empty version number,
6668 i.e. as "exim-". This error state is now caught, and the installation is
6671 38. An argument was missing from the function that creates an error message
6672 when Exim fails to connect to the socket for saslauthd authentication.
6673 This could cause Exim to crash, or give a corrupted message.
6675 39. Added isip, isip4, and isip6 to ${if conditions.
6677 40. The ACL variables $acl_xx are now saved with the message, and can be
6678 accessed later in routers, transports, and filters.
6680 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6681 strings in the file are not string-expanded.
6683 42. If a MAIL command specified a SIZE value that was too large to fit into an
6684 int variable, the check against message_size_limit failed. Such values are
6685 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6686 day this will have to be increased, but I don't think I want to be around
6687 when emails are that large.
6694 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6695 iconv() is not standard in FreeBSD.
6697 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6698 IPv6 enabled. The observed symptom was a segmentation fault on return from
6699 the function os_common_find_running_interfaces() in src/os.c.
6701 3. In the check_special_case() function in daemon.c I had used "errno" as an
6702 argument name, which causes warnings on some systems. This was basically a
6703 typo, since it was named "eno" in the comments!
6705 4. The code that waits for the clock to tick (at a resolution of some fraction
6706 of a second) so as to ensure message-id uniqueness was always waiting for
6707 at least one whole tick, when it could have waited for less. [This is
6708 almost certainly not relevant at current processor speeds, where it is
6709 unlikely to ever wait at all. But we try to future-proof.]
6711 5. The function that sleeps for a time interval that includes fractions of a
6712 second contained a race. It did not block SIGALRM between setting the
6713 timer, and suspending (a couple of lines later). If the interval was short
6714 and the sigsuspend() was delayed until after it had expired, the suspension
6715 never ended. On busy systems this could lead to processes getting stuck for
6718 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6719 process, before it forks any delivery processes. The open lookup caching
6720 mechanism meant that the open file or database connection was passed into
6721 the delivery process. The problem was that delivery processes always tidy
6722 up cached lookup data. This could cause a problem for the next delivery
6723 process started by the queue runner, because the external queue runner
6724 process does not know about the closure. So the next delivery process
6725 still has data in the lookup cache. In the case of a file lookup, there was
6726 no problem because closing a file descriptor in a subprocess doesn't affect
6727 the parent. However, if the lookup was caching a connection to a database,
6728 the connection was closed, and the second delivery process was likely to
6729 see errors such as "PGSQL: query failed: server closed the connection
6730 unexpectedly". The problem has been fixed by closing all cached lookups
6731 in a queue runner before running a delivery process.
6733 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6734 seem to have the "const" qualifier which it has on other OS. I've
6737 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6738 *of the same type* (client or server) with the same public name that an
6739 error should be diagnosed.
6741 9. When Exim looked up a host name for an IP address, but failed to find the
6742 original IP address when looking up the host name (a safety check), it
6743 output the message "<ip address> does not match any IP for NULL", which was
6744 confusing, to say the least. The bug was that the host name should have
6745 appeared instead of "NULL".
6747 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6748 user that is built into the binary, and the -C or -D options is used, root
6749 privilege is dropped before the configuration file is read. In addition,
6750 logging is switched to stderr instead of the normal log files. If the
6751 configuration then re-defines the Exim user, the unprivileged environment
6752 is probably not what is expected, so Exim logs a panic warning message (but
6755 However, if deliver_drop_privilege is set, the unprivileged state may well
6756 be exactly what is intended, so the warning has been cut out in that case,
6757 and Exim is allowed to try to write to its normal log files.
6763 1. smtp_return_error_details was not giving details for temporary sender
6764 or receiver verification errors.
6766 2. Diagnose a configuration error if two authenticators have the same public
6769 3. Exim used not to create the message log file for a message until the first
6770 delivery attempt. This could be confusing when incoming messages were held
6771 for policy or load reasons. The message log file is now created at the time
6772 the message is received, and an initial "Received" line is written to it.
6774 4. The automatically generated man page for command line options had a minor
6775 bug that caused no ill effects; however, a more serious problem was that
6776 the procedure for building the man page automatically didn't always
6777 operate. Consequently, release 4.20 contains an out-of-date version. This
6778 shouldn't happen again.
6780 5. When building Exim with embedded Perl support, the script that builds the
6781 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6782 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6784 6. The freeze_tell option was not being used for messages that were frozen on
6785 arrival, either by an ACL or by local_scan().
6787 7. Added the smtp_incomplete_transaction log selector.
6789 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6790 was accepting AUTH without a new EHLO.
6792 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6793 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6796 10. Logging of TCP/IP connections (when configured) now happens in the main
6797 daemon process instead of the child process, so that the TCP/IP connection
6798 count is more accurate (but it can never be perfect).
6800 11. The use of "drop" in a nested ACL was not being handled correctly in the
6801 outer ACL. Now, if condition failure induced by the nested "drop" causes
6802 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6803 or "require"), the connection is dropped.
6805 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6806 that yield "discard" can now be used with an "accept" or a "discard" verb,
6807 but an error is generated for any others (because I can't see a useful way
6808 to define what should happen).
6810 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6811 are now processed in the same way as lines in the Exim configuration file.
6812 In particular, continuation lines are supported.
6814 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6816 15. Added -ti meaning -t -i.
6818 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6819 domains, and warn by logging if others are found.
6821 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6822 returned from GIFCONF when Exim is trying to find the list of interfaces on
6823 a host. The code in os.c has been modified to copy each ifreq to an aligned
6824 structure in all cases.
6826 Also, in some cases, the returned ifreq's were being copied to a 'struct
6827 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6828 means the last couple of bytes of an IPv6 address could be chopped if the
6829 ifreq contained only a normal sockaddr (14 bytes storage).
6831 18. Named domain lists were not supported in the hosts_treat_as_local option.
6832 An entry such as +xxxx was not recognized, and was treated as a literal
6835 19. Ensure that header lines added by a DATA ACL are included in the reject log
6836 if the ACL subsequently rejects the message.
6838 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6839 MD5 (which is deprecated).
6841 21. When testing a filter file using -bf, Exim was writing a message when it
6842 took the sender from a "From " line in the message, but it was not doing so
6843 when it took $return_path from a Return-Path: header line. It now does.
6845 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6846 with a valid header line field name (a series of printing characters
6847 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6849 23. Changed "disc" in the source to "disk" to conform to the documentation and
6850 the book and for uniformity.
6852 24. Ignore Sendmail's -Ooption=value command line item.
6854 25. When execve() failed while trying to run a command in a pipe transport,
6855 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6856 could be confused with a return value of 69 from the command itself. This
6857 has been changed to 127, the value the shell returns if it is asked to run
6858 a non-existent command. The wording for the related log line suggests a
6859 non-existent command as the problem.
6861 26. If received_header_text expands to an empty string, do not add a Received:
6862 header line to the message. (Well, it adds a token one on the spool, but
6863 marks it "old" so that it doesn't get used or transmitted.)
6865 27. Installed eximstats 1.28 (addition of -nt option).
6867 28. There was no check for failure on the call to getsockname() in the daemon
6868 code. This can fail if there is a shortage of resources on the system, with
6869 ENOMEM, for example. A temporary error is now given on failure.
6871 29. Contrary to the C standard, it seems that in some environments, the
6872 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6873 program. Exim now does this explicitly; it affects the formatting of
6874 timestamps using strftime().
6876 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6877 complaints. I've now initialized all the variables, to avoid this.
6879 32. Header lines added by a system filter were not being "seen" during
6880 transport-time rewrites.
6882 33. The info_callback() function passed to OpenSSL is set up with type void
6883 (*)(SSL *, int, int), as described somewhere. However, when calling the
6884 function (actually a macro) that sets it up, the type void(*)() is
6885 expected. I've put in a cast to prevent warnings from picky compilers.
6887 34. If a DNS black list lookup found a CNAME record, but there were no A
6888 records associated with the domain it pointed at, Exim crashed.
6890 35. If a DNS black list lookup returned more than one A record, Exim ignored
6891 all but the first. It now scans all returned addresses if a particular IP
6892 value is being sought. In this situation, the contents of the
6893 $dnslist_value variable are a list of all the addresses, separated by a
6896 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6897 to accept a host name and all its aliases if the forward lookup for any of
6898 them yielded the IP address of the incoming connection. Now it accepts only
6899 those names whose forward lookup yields the correct IP address. Any other
6900 names are discarded. This closes a loophole whereby a rogue DNS
6901 administrator could create reverse DNS records to break through a
6902 wildcarded host restriction in an ACL.
6904 37. If a user filter or a system filter that ran in a subprocess used any of
6905 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6906 the wrong values were passed to the pipe command ($thisaddress had the
6907 value of $0, $0 had the value of $1, etc). This bug was introduced by
6908 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6910 38. Improved the line breaking for long SMTP error messages from ACLs.
6911 Previously, if there was no break point between 40 and 75 characters, Exim
6912 left the rest of the message alone. Two changes have been made: (a) I've
6913 reduced the minimum length to 35 characters; (b) if it can't find a break
6914 point between 35 and 75 characters, it looks ahead and uses the first one
6915 that it finds. This may give the occasional overlong line, but at least the
6916 remaining text gets split now.
6918 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6919 file descriptors might be low, and that setting 1000 would always raise it.
6920 It turns out that in some environments, the limit is already over 1000 and
6921 that lowering it causes trouble. So now Exim takes care not to decrease it.
6923 40. When delivering a message, the value of $return_path is set to $sender_
6924 address at the start of routing (routers may change the value). By an
6925 oversight, this default was not being set up when an address was tested by
6926 -bt or -bv, which affected the outcome if any router or filter referred to
6929 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6930 log only when "message" or "log_message" are set. However, if one of the
6931 conditions was an address verification, or a call to a nested ACL, the
6932 messages generated by the underlying test were being passed through. This
6933 no longer happens. The underlying message is available in $acl_verify_
6934 message for both "message" and "log_message" expansions, so it can be
6935 passed through if needed.
6937 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6938 new expansion $bh_ to give the encoded byte string without charset
6939 translation. Translation happens only if iconv() is available; HAVE_ICONV
6940 indicates this at build time. HEADERS_CHARSET gives the charset to
6941 translate to; headers_charset can change it in the configuration, and
6942 "headers charset" can change it in an individual filter file.
6944 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6945 that creates RFC 2047 encoded "words" labels them as that charset instead
6946 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6947 expansion operator; (ii) when Exim creates a From: line for a local
6948 message; (iii) when a header line is rewritten to include a "phrase" part.
6950 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6951 buggy, causing it to skip the first lines of messages whose message ID
6952 ended in 'D'. This would not have bitten before Exim release 4.14, because
6953 message IDs were unlikely to end in 'D' before then. The effect was to have
6954 incorrect size information for certain domains.
6956 45. #include "config.h" was missing at the start of the crypt16.c module. This
6957 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6960 46. If there was a timeout during a "random" callout check, Exim treated it as
6961 a failure of the random address, and carried on sending RSET and the real
6962 address. If the delay was just some slowness somewhere, the response to the
6963 original RCPT would be taken as a response to RSET and so on, causing
6964 mayhem of various kinds.
6966 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6967 when I implemented it. It didn't allow for the fact that some option values
6968 may legitimately be negative (e.g. size_addition), and it didn't even do
6969 the right test for positive values.
6971 48. Domain names in DNS records are case-independent. Exim always looks them up
6972 in lower case. Some resolvers return domain names in exactly the case they
6973 appear in the zone file, that is, they may contain uppercase letters. Not
6974 all resolvers do this - some return always lower case. Exim was treating a
6975 change of case by a resolver as a change of domain, similar to a widening
6976 of a domain abbreviation. This triggered its re-routing code and so it was
6977 trying to route what was effectively the same domain again. This normally
6978 caused routing to fail (because the router wouldn't handle the domain
6979 twice). Now Exim checks for this case specially, and just changes the
6980 casing of the domain that it ultimately uses when it transmits the message
6983 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6986 50. If a filter generated a file delivery with a non-absolute name (possible if
6987 no home directory exists for the router), the forbid_file option was not
6990 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6991 the existing equality matching.
6993 52. Exim was using ints instead of ino_t variables in some places where it was
6994 dealing with inode numbers.
6996 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6997 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6998 called TMPDIR, and if it finds it is different, it changes its value.
7000 54. The smtp_printf() function is now made available to local_scan() so
7001 additional output lines can be written before returning. There is also an
7002 smtp_fflush() function to enable the detection of a dropped connection.
7003 The variables smtp_input and smtp_batched_input are exported to
7006 55. Changed the default runtime configuration: the message "Unknown user"
7007 has been removed from the ACL, and instead placed on the localuser router,
7008 using the cannot_route_message feature. This means that any verification
7009 failures that generate their own messages won't get overridden. Similarly,
7010 the "Unrouteable address" message that was in the ACL for unverifiable
7011 relay addresses has also been removed.
7013 56. Added hosts_avoid_esmtp to the smtp transport.
7015 57. The exicyclog script was not checking for the esoteric option
7016 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
7017 will work only if exicyclog is run under the appropriate euid.
7019 58. Following a discussion on the list, the rules by which Exim recognises line
7020 endings on incoming messages have been changed. The -dropcr and drop_cr
7021 options are now no-ops, retained only for backwards compatibility. The
7022 following line terminators are recognized: LF CRLF CR. However, special
7023 processing applies to CR:
7025 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7026 nor a local message in the state where . is a terminator.
7028 (ii) If a bare CR is encountered in a header line, an extra space is added
7029 after the line terminator so as not to end the header. The reasoning
7030 behind this is that bare CRs in header lines are most likely either
7031 to be mistakes, or people trying to play silly games.
7033 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7034 was being incorrectly given as 18 bytes larger than it should have been.
7035 This is a VOB (very old bug).
7037 60. This may never have affected anything current, but just in case it has:
7038 When the local host is found other than at the start of a list of hosts,
7039 the local host, those with the same MX, and any that follow, are discarded.
7040 When the list in question was part of a longer list of hosts, the following
7041 hosts (not currently being processed) were also being discarded. This no
7042 longer happens. I'm not sure if this situation could ever has previously
7045 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7048 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7049 header. This is something that is recommended in a new Internet Draft, and
7050 is something that is documented as being done by Sendmail. There are two
7051 possible values. For messages generated by the autoreply transport, Exim
7054 Auto-Submitted: auto-replied
7056 whereas for all other generated messages (e.g. bounces) it adds
7058 Auto-Submitted: auto-generated
7060 63. The "personal" condition in filters now includes a test for the
7061 Auto-Submitted: header. If it contains the string "auto-" the message it
7062 not considered personal.
7064 64. Added rcpt_include_affixes as a generic transport option.
7066 65. Added queue_only_override (default true).
7068 66. Added the syslog_duplication option.
7070 67. If what should have been the first header line of a message consisted of
7071 a space followed by a colon, Exim was mis-interpreting it as a header line.
7072 It isn't of course - it is syntactically invalid and should therefore be
7073 treated as the start of the message body. The misbehaviour could have
7074 caused a number of strange effects, including loss of data in subsequent
7075 header lines, and spool format errors.
7077 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7078 client host had authenticated. This control can now be exercised by an ACL
7079 for more flexibility.
7081 69. By default, callouts do not happen when testing with -bh. There is now a
7082 variant, -bhc, which does actually run the callout code, including
7083 consulting and updating the callout cache.
7085 70. Added support for saslauthd authentication, courtesy of Alexander
7088 71. If statvfs() failed on the spool or log directories while checking their
7089 size for availability, Exim confusingly gave the error "space shortage".
7090 Furthermore, in debugging mode it crashed with a floating point exception.
7091 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7092 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7093 serious problem, Exim now writes to the main and panic logs when this
7094 happens, with details of the failure. It then refuses to accept the
7095 incoming message, giving the message "spool directory problem" or "log
7096 directory problem" with a 421 code for SMTP messages.
7098 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7099 0, 1, and 2 exist, because some OS complain for execs without them (see
7100 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7101 descriptors. However, the code omitted to check that the open succeeded,
7102 causing mysterious errors if for some reason the permissions on /dev/null
7103 got screwed. Now Exim writes a message to the main and panic logs, and
7104 bombs out if it can't open /dev/null.
7106 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7107 interact so that it is all more flexible. It is supposed to remain
7108 backwards compatible. Also added extra_local_interfaces.
7110 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7111 to bomb out with an assertion failure - to the client this appears as a
7112 connection drop. This problem occurs in the part of the code that was taken
7113 from the Samba project. Fortunately, the assertion is in a very simple
7114 function, so I have fixed this by reproducing the function inline in the
7115 one place where it is called, and arranging for authentication to fail
7116 instead of killing the process with assert().
7118 75. The SPA client code was not working when the server requested OEM rather
7119 than Unicode encoding.
7121 76. Added code to make require_files with a specific uid setting more usable in
7122 the case where statting the file as root fails - usually a non-root-mounted
7123 NFS file system. When this happens and the failure is EACCES, Exim now
7124 forks a subprocess and does the per-uid checking as the relevant uid.
7126 77. Added process_log_path.
7128 78. If log_file_path was not explicitly set, a setting of check_log_space or
7129 check_log_inodes was ignored.
7131 79. If a space check for the spool or log partitions fails, the incident is now
7132 logged. Of course, in the latter case the data may get lost...
7134 80. Added the %p formatting code to string_format() so that it can be used to
7135 print addresses in debug_print(). Adjusted all the address printing in the
7136 debugging in store.c to use %p rather than %d.
7138 81. There was a concern that a line of code in smtp_in.c could overflow a
7139 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7140 initially expressed, the concern was not well-founded, because trailing
7141 spaces are removed early. However, if the trailing spaces were followed by
7142 a NULL, they did not get removed, so the overflow was possible. Two fixes
7145 (a) I re-wrote the offending code in a cleaner fashion.
7146 (b) If an incoming SMTP command contains a NULL character, it is rejected
7149 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7150 runs initgroups(), so that if the Exim user is in any additional groups,
7151 they will be used during message reception.
7157 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob