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 JH/01 Rework error string handling in TLS interface so that the caller in
12 more cases is responsible for logging. This permits library-sourced
13 string to be attached to addresses during delivery, and collapses
14 pairs of long lines into single ones.
16 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
17 during configuration. Wildcards are allowed and expanded.
19 JH/02 Rework error string handling in DKIM to pass more info back to callers.
20 This permits better logging.
22 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
23 do not close it down and have the child transport start it up again on
24 the passed-on TCP connection. Instead, proxy the child (and any
25 subsequent ones) for TLS via a unix-domain socket channel. Logging is
26 affected: the continued delivery log lines do not have any DNSSEC, TLS
27 Certificate or OCSP information. TLS cipher information is still logged.
29 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
30 identical IP addresses on different listening ports. Will also affect
33 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
34 add noisy ifdef guards to special-case this sillyness.
35 Patch from Bernd Kuhls.
37 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
38 than 255 are no longer allowed.
40 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
41 Disable the session-cache too, which might reduce our load. Since we
42 currrectly use a new context for every connection, both as server and
43 client, there is no benefit for these.
44 GnuTLS appears to not support tickets server-side by default (we don't
45 call gnutls_session_ticket_enable_server()) but client side is enabled
46 by default on recent versions (3.1.3 +) unless the PFS priority string
49 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
50 <https://reproducible-builds.org/specs/source-date-epoch/>.
52 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
53 the check for any unsuccessful recipients did not notice the limit, and
54 erroneously found still-pending ones.
56 JH/08 Pipeline CHUNKING command and data together, on kernels that support
57 MSG_MORE. Only in-clear (not on TLS connections).
59 JH/09 Avoid using a temporary file during transport using dkim. Unless a
60 transport-filter is involved we can buffer the headers in memory for
61 creating the signature, and read the spool data file once for the
62 signature and again for transmission.
64 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
65 4.77 as the kernel support then wasn't solid, having issues in 64bit
66 mode. Now, it's been long enough. Add support for FreeBSD also.
68 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
69 case where the routing stage had gathered several addresses to send to
70 a host before calling the transport for the first, we previously failed
71 to close down TLS in the old transport process before passing the TCP
72 connection to the new process. The new one sent a STARTTLS command
73 which naturally failed, giving a failed delivery and bloating the retry
74 database. Investigation and fix prototype from Wolfgang Breyha.
76 JH/12 Fix check on SMTP command input synchronisation. Previously there were
77 false-negatives in the check that the sender had not preempted a response
78 or prompt from Exim (running as a server), due to that code's lack of
79 awareness of the SMTP input buffering.
81 PP/04 Add commandline_checks_require_admin option.
82 Exim drops privileges sanely, various checks such as -be aren't a
83 security problem, as long as you trust local users with access to their
84 own account. When invoked by services which pass untrusted data to
85 Exim, this might be an issue. Set this option in main configuration
86 AND make fixes to the calling application, such as using `--` to stop
89 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
90 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
91 responses to those, into a single TLS record each way (this usually means
92 a single packet). As a side issue, smtp_enforce_sync now works on TLS
95 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
96 affects you only if you're dancing at the edge of the param size limits.
97 If you are, and this message makes sense to you, then: raise the
98 configured limit or use OpenSSL 1.1. Nothing we can do for older
101 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
102 cmdline element to get the documented default one. Previously it was
105 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
106 are unsuable for non-message-specific reasons - eg. connection timeout,
107 banner-time rejection.
109 JH/16 Fix logging of delivery remote port, when specified by router, under
112 PP/06 Repair manualroute's ability to take options in any order, even if one
113 is the name of a transport.
116 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
118 JH/17 Change the list-building routines interface to use the expanding-string
119 triplet model, for better allocation and copying behaviour.
121 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
122 Previously it was constructed the first time a possibly-matching string
123 was met in the configuration file input during startup; now it is done
126 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
127 compatible one, to avoid the (poorly documented) possibility of a config
128 file in the working directory redirecting the DB files, possibly correpting
129 some existing file. CVE-2017-10140 assigned for BDB.
131 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
132 cache-hot. Previously, although the result was properly cached, the
133 initial verify call returned a defer.
135 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
136 the main verify for receipient in uncached-mode.
138 JH/22 Retire historical build files to an "unsupported" subdir. These are
139 defined as "ones for which we have no current evidence of testing".
141 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
142 if present. Previously it was ignored.
144 JH/24 Start using specified-initialisers in C structure init coding. This is
145 a C99 feature (it's 2017, so now considered safe).
147 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
148 if was a fixed-sized field and bitmask ops via macros; it is now more
151 PP/07 GitHub PR 56: Apply MariaDB build fix.
152 Patch provided by Jaroslav Škarvada.
154 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
155 during Coverity cleanups [4.87 JH/47]
156 Diagnosis and fix provided by Michael Fischer v. Mollard.
158 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
159 the right size to place the terminating semicolon on its own folded
160 line, the header hash was calculated to an incorrect value thanks to
161 the (relaxed) space the fold became.
163 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
164 and confused the parent.
166 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
167 which could crash as a result. This could lead to undeliverable messages.
169 JH/28 Logging: "next input sent too soon" now shows where input was truncated
172 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
173 matters on fast-turnover and PID-randomising systems, which were getting
174 out-of-order delivery.
176 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
177 a possibly-overlapping copy. The symptom was that "Remote host closed
178 connection in response to HELO" was logged instead of the actual 4xx
181 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
182 Previously only that bufferd was discarded, resulting in SYMTP command
185 JH/32 DKIM: when a message has multiple signatures matching an identity given
186 in dkim_verify_signers, run the dkim acl once for each. Previously only
187 one run was done. Bug 2189.
193 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
194 than -2003 did; needs libidn2 in addition to libidn.
196 JH/02 The path option on a pipe transport is now expanded before use.
198 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
199 Patch provided by "Björn", documentation fix added too.
201 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
202 missing a wire-to-host endian conversion.
204 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
205 close after a BDAT command line could be taken as a following command,
206 giving a synch failure. Fix by only checking for synch immediately
207 before acknowledging the chunk.
209 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
210 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
211 macro. Patches provided by Josh Soref.
213 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
214 Previously we did not; the RFC seems ambiguous and VRFY is not listed
215 by IANA as a service extension. However, John Klensin suggests that we
218 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
219 the dkim code may be unix-mode line endings rather than smtp wire-format
220 CRLF, so prepend a CR to any bare LF.
222 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
223 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
225 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
226 the first were themselves being wrongly included in the feed into dkim
227 processing; with most chunk sizes in use this resulted in an incorrect
228 body hash calculated value.
230 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
231 DKIM signature block, for verification. Although advised against by
232 standards it is specifically not ruled illegal.
234 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
236 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
237 missing a body hash (the bh= tag).
239 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
240 It seems that HAProxy sends the Proxy Protocol information in clear and
241 only then does a TLS startup, so do the same.
243 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
244 TCP connections (such as for Spamd) unless the daemon successfully set
245 Fast Open mode on its listening sockets. This fixes breakage seen on
246 too-old kernels or those not configured for Fast Open, at the cost of
247 requiring both directions being enabled for TFO, and TFO never being used
248 by non-daemon-related Exim processes.
250 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
251 endings, at least on the first header line. Try to canonify any that get
252 past that check, despite the cost.
254 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
255 now limited to an arbitrary five deep, while parsing addresses with the
256 strip_excess_angle_brackets option enabled.
258 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
259 instead leave the unprompted TLS handshake in socket buffer for the
260 TLS library to consume.
262 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
264 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
266 JH/16 Drop variables when they go out of scope. Memory management drops a whole
267 region in one operation, for speed, and this leaves assigned pointers
268 dangling. Add checks run only under the testsuite which checks all
269 variables at a store-reset and panics on a dangling pointer; add code
270 explicitly nulling out all the variables discovered. Fixes one known
271 bug: a transport crash, where a dangling pointer for $sending_ip_address
272 originally assigned in a verify callout, is re-used.
274 PP/06 Drop '.' from @INC in various Perl scripts.
276 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
278 PP/08 Reduce a number of compilation warnings under clang; building with
279 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
280 should be warning-free.
282 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
284 HS/01 Fix portability problems introduced by PP/08 for platforms where
285 realloc(NULL) is not equivalent to malloc() [SunOS et al].
287 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
288 chunk. This allows us to accept broken chunked messages. We need a more
289 general solution here.
291 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
292 already-broken messages in the queue.
294 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
296 JH/19 Fix reference counting bug in routing-generated-address tracking.
302 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
303 supports it and a size is available (ie. the sending peer gave us one).
305 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
306 years of being deprecated). The replacements are the ACLs
307 acl_smtp_mime and acl_not_smtp_mime.
309 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
310 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
311 or even an in-clear connection were permitted. Now, if the host lookup
312 was dnssec and dane was requested then the host is only used if the
313 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
314 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
315 if one fails this test.
316 This means that a poorly-configured remote DNS will make it incommunicado;
317 but it protects against a DNS-interception attack on it.
319 JH/04 Bug 1810: make continued-use of an open smtp transport connection
320 non-noisy when a race steals the message being considered.
322 JH/05 If main configuration option tls_certificate is unset, generate a
323 self-signed certificate for inbound TLS connections.
325 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
326 in rewrites and routers.
328 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
329 and logged a warning sing 4.83; now they are a configuration file error.
331 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
332 (lacking @domain). Apply the same qualification processing as RCPT.
334 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
336 JH/10 Support ${sha256:} applied to a string (as well as the previous
339 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
340 a cutthrough deliver is pending, as we always want to make a connection.
341 This also avoids re-routing the message when later placing the cutthrough
342 connection after a verify cache hit.
343 Do not update it with the verify result either.
345 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
346 when routing results in more than one destination address.
348 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
349 signing (which inhibits the cutthrough capability). Previously only
350 the presence of an option was tested; now an expansion evaluating as
351 empty is permissible (obviously it should depend only on data available
352 when the cutthrough connection is made).
354 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
355 the relevant preceding SMTP command did not note the pipelining mode.
357 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
358 Previously they were not counted.
360 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
361 as one having no matching records. Previously we deferred the message
362 that needed the lookup.
364 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
365 distinguished as "(=".
367 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
368 for missing MX records. Previously it only worked for missing A records.
370 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
372 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
373 after the data-go-ahead and data-ack. Patch from Jason Betts.
375 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
376 even for a "none" policy. Patch from Tony Meyer.
378 JH/22 Fix continued use of a connection for further deliveries. If a port was
379 specified by a router, it must also match for the delivery to be
382 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
383 When one of the recipients of a message was unsuitable for the connection
384 (has no matching addresses), we lost track of needing to mark it
385 deferred. As a result mail would be lost.
387 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
389 JH/25 Decoding ACL controls is now done using a binary search; the source code
390 takes up less space and should be simpler to maintain. Merge the ACL
391 condition decode tables also, with similar effect.
393 JH/26 Fix problem with one_time used on a redirect router which returned the
394 parent address unchanged. A retry would see the parent address marked as
395 delivered, so not attempt the (identical) child. As a result mail would
398 JH/27 Fix a possible security hole, wherein a process operating with the Exim
399 UID can gain a root shell. Credit to http://www.halfdog.net/ for
400 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
403 JH/28 Enable {spool,log} filesystem space and inode checks as default.
404 Main config options check_{log,spool}_{inodes,space} are now
405 100 inodes, 10MB unless set otherwise in the configuration.
407 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
408 Previously it only applied to the main-section connection policy
411 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
413 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
414 by me. Added RFC7919 DH primes as an alternative.
416 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
417 are not in the system include path.
419 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
420 GnuTLS, when a session startup failed (eg because the client disconnected)
421 Exim did stdio operations after fclose. This was exposed by a recent
422 change which nulled out the file handle after the fclose.
424 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
425 signed directly by the cert-signing cert, rather than an intermediate
426 OCSP-signing cert. This is the model used by LetsEncrypt.
428 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
430 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
431 an incoming connection.
433 HS/02 Bug 1802: Do not half-close the connection after sending a request
436 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
437 fallback to "prime256v1".
439 JH/34 SECURITY: Use proper copy of DATA command in error message.
440 Could leak key material. Remotely exploitable. CVE-2016-9963.
446 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
447 and 3.4.4 - once the server is enabled to respond to an OCSP request
448 it does even when not requested, resulting in a stapling non-aware
449 client dropping the TLS connection.
451 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
452 support variable-length bit vectors. No functional change.
454 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
455 The I= interface field on outgoing lines is now after the H= remote
456 host field, same as incoming lines. There is a separate
457 outgoing_interface log selector which allows you to disable the
460 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
461 If not running log_selector +smtp_connection the mainlog would be held
462 open indefinitely after a "too many connections" event, including to a
463 deleted file after a log rotate. Leave the per net connection logging
464 leaving it open for efficiency as that will be quickly detected by the
465 check on the next write.
467 HS/01 Bug 1671: Fix post transport crash.
468 Processing the wait-<transport> messages could crash the delivery
469 process if the message IDs didn't exist for some reason. When
470 using 'split_spool_directory=yes' the construction of the spool
471 file name failed already, exposing the same netto behaviour.
473 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
474 mime_regex ACL conditions.
476 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
477 to DSN fail messages (bounces): remote IP, remote greeting, remote response
478 to HELO, local diagnostic string.
480 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
481 log line to debug. Even when configured with a tls authenticator many
482 client connections are expected to not authenticate in this way, so
483 an authenticate fail is not an error.
485 HS/02 Add the Exim version string to the process info. This way exiwhat
486 gives some more detail about the running daemon.
488 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
489 matter for fast-change records such as DNSBLs.
491 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
492 retry record, even if constant. There may be multiple transports with
493 different interface settings and the retry behaviour needs to be kept
496 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
498 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
500 JH/10 Bug 840: fix log_defer_output option of pipe transport
502 JH/11 Bug 830: use same host for all RCPTS of a message, even under
503 hosts_randomize. This matters a lot when combined with mua_wrapper.
505 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
506 ${quote_pgsql:<string>} operator.
508 JH/13 Bug 1708: avoid misaligned access in cached lookup.
510 JH/14 Change header file name for freeradius-client. Relevant if compiling
511 with Radius support; from the Gentoo tree and checked under Fedora.
513 JH/15 Bug 1712: Introduce $prdr_requested flag variable
515 JH/16 Bug 1714: Permit an empty string as expansion result for transport
516 option transport_filter, meaning no filtering.
518 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
520 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
521 defaults to "*" (all hosts). The variable is now available when not built
522 with TLS, default unset, mainly to enable keeping the testsuite sane.
523 If a server certificate is not supplied (via tls_certificate) an error is
524 logged, and clients will find TLS connections fail on startup. Presumably
525 they will retry in-clear.
526 Packagers of Exim are strongly encouraged to create a server certificate
527 at installation time.
529 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
530 with the $config_file variable.
532 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
533 in transport context, after the attempt, and per-recipient. The latter type
534 is per host attempted. The event data is the error message, and the errno
535 information encodes the lookup type (A vs. MX) used for the (first) host,
536 and the trailing two digits of the smtp 4xx response.
538 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
539 to write to mainlog (or rejectlog, paniclog) in the window between file
540 creation and permissions/ownership being changed. Particularly affects
541 installations where exicyclog is run as root, rather than exim user;
542 result is that the running daemon panics and dies.
544 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
546 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
547 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
548 "pri" and "weight". Note that the previous implicit priority given by the
549 list order is no longer honoured.
551 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
554 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
555 by defining SUPPORT_SOCKS.
557 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
558 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
559 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
560 variables are renamed to proxy_{local,external}_{address,port}.
562 JH/27 Move Internationalisation support from Experimental to mainline, enabled
563 for a build by defining SUPPORT_I18N
565 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
566 of the query string, and make ${quote_redis:} do that quoting.
568 JH/29 Move Events support from Experimental to mainline, enabled by default
569 and removable for a build by defining DISABLE_EVENT.
571 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
573 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
574 cached by the daemon.
576 JH/32 Move Redis support from Experimental to mainline, enabled for a build
577 by defining LOOKUP_REDIS. The libhiredis library is required.
579 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
580 keys are given for lookup.
582 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
583 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
584 only supported when built with TLS support. The PolarSSL SHA routines
585 are still used when the TLS library is too old for convenient support.
587 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
588 openssl_options), for security. OpenSSL forces this from version 1.1.0
589 server-side so match that on older versions.
591 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
592 allocation for $value could be released as the expansion processing
593 concluded, but leaving the global pointer active for it.
595 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
596 and to use the domains and local_parts ACL conditions.
598 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
599 incorrectly not doubled on cutthrough transmission, hence seen as a
600 body-termination at the receiving system - resulting in truncated mails.
601 Commonly the sender saw a TCP-level error, and retransmitted the message
602 via the normal store-and-forward channel. This could result in duplicates
603 received - but deduplicating mailstores were liable to retain only the
604 initial truncated version.
606 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
608 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
610 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
611 we're in there, support oversigning also; bug 1309.
613 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
615 HS/04 Add support for keep_environment and add_environment options.
617 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
618 either intentional arithmetic overflow during PRNG, or testing config-
621 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
622 delivery resulted in actual delivery. Cancel cutthrough before DATA
625 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
626 rejects a recipient: pass the reject to the originator.
628 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
629 Many were false-positives and ignorable, but it's worth fixing the
632 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
633 for the new environment-manipulation done at startup. Move the routines
634 from being local to tls.c to being global via the os.c file.
636 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
637 an extract embedded as result-arg for a map, the first arg for extract
638 is unavailable so we cannot tell if this is a numbered or keyed
639 extraction. Accept either.
645 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
648 JH/02 The smtp transport option "multi_domain" is now expanded.
650 JH/03 The smtp transport now requests PRDR by default, if the server offers
653 JH/04 Certificate name checking on server certificates, when exim is a client,
654 is now done by default. The transport option tls_verify_cert_hostnames
655 can be used to disable this per-host. The build option
656 EXPERIMENTAL_CERTNAMES is withdrawn.
658 JH/05 The value of the tls_verify_certificates smtp transport and main options
659 default to the word "system" to access the system default CA bundle.
660 For GnuTLS, only version 3.0.20 or later.
662 JH/06 Verification of the server certificate for a TLS connection is now tried
663 (but not required) by default. The verification status is now logged by
664 default, for both outbound TLS and client-certificate supplying inbound
667 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
670 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
671 Status Notification (bounce) messages are now MIME format per RFC 3464.
672 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
673 under the control of the dsn_advertise_hosts option, and routers may
674 have a dsn_lasthop option.
676 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
677 default, modifiable by a malware= option. The list separator for
678 the options can now be changed in the usual way. Bug 68.
680 JH/10 The smtp_receive_timeout main option is now expanded before use.
682 JH/11 The incoming_interface log option now also enables logging of the
683 local interface on delivery outgoing connections.
685 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
686 if the interface and destination host and port all match.
688 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
691 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
692 Patch from Andrew Lewis.
694 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
695 now supports optional time-restrictions, weighting, and priority
696 modifiers per server. Patch originally by <rommer@active.by>.
698 JH/16 The spamd_address main option now supports a mixed list of local
699 and remote servers. Remote servers can be IPv6 addresses, and
700 specify a port-range.
702 JH/17 Bug 68: The spamd_address main option now supports an optional
703 timeout value per server.
705 JH/18 Bug 1581: Router and transport options headers_add/remove can
706 now have the list separator specified.
708 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
711 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
714 JH/21 Support for the A6 type of dns record is withdrawn.
716 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
717 rather than the verbs used.
719 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
720 from 255 to 1024 chars.
722 JH/24 Verification callouts now attempt to use TLS by default.
724 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
725 are generic router options now. The defaults didn't change.
727 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
728 Original patch from Alexander Shikoff, worked over by JH.
730 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
731 files if ZCAT_COMMAND is not executable.
733 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
735 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
737 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
738 Normally benign, it bites when the pair was led to by a CNAME;
739 modern usage is to not canonicalize the domain to a CNAME target
740 (and we were inconsistent anyway for A-only vs AAAA+A).
742 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
744 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
745 when evaluating $sender_host_dnssec.
747 JH/31 Check the HELO verification lookup for DNSSEC, adding new
748 $sender_helo_dnssec variable.
750 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
752 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
754 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
756 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
757 documented as working, but never had. Support all but $spam_report.
759 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
760 added for tls authenticator.
762 HS/03 Add perl_taintmode main config option
768 TL/01 When running the test suite, the README says that variables such as
769 no_msglog_check are global and can be placed anywhere in a specific
770 test's script, however it was observed that placement needed to be near
771 the beginning for it to behave that way. Changed the runtest perl
772 script to read through the entire script once to detect and set these
773 variables, reset to the beginning of the script, and then run through
774 the script parsing/test process like normal.
776 TL/02 The BSD's have an arc4random API. One of the functions to induce
777 adding randomness was arc4random_stir(), but it has been removed in
778 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
779 function when detected.
781 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
782 cause callback expansion.
784 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
785 syntax errors in an expansion can be treated as a string instead of
786 logging or causing an error, due to the internal use of bool_lax
787 instead of bool when processing it.
789 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
790 server certificates when making smtp deliveries.
792 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
794 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
796 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
798 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
799 Merged patch from Sebastian Wiedenroth.
801 JH/05 Fix results-pipe from transport process. Several recipients, combined
802 with certificate use, exposed issues where response data items split
803 over buffer boundaries were not parsed properly. This eventually
804 resulted in duplicates being sent. This issue only became common enough
805 to notice due to the introduction of connection certificate information,
806 the item size being so much larger. Found and fixed by Wolfgang Breyha.
808 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
809 size buffer was used, resulting in syntax errors when an expansion
812 JH/07 Add support for directories of certificates when compiled with a GnuTLS
813 version 3.3.6 or later.
815 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
816 is EXPERIMENTAL_EVENT, the main-configuration and transport options
817 both become "event_action", the variables become $event_name, $event_data
818 and $event_defer_errno. There is a new variable $verify_mode, usable in
819 routers, transports and related events. The tls:cert event is now also
820 raised for inbound connections, if the main configuration event_action
823 TL/06 In test suite, disable OCSP for old versions of openssl which contained
824 early OCSP support, but no stapling (appears to be less than 1.0.0).
826 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
827 server certificate names available under the smtp transport option
828 "tls_verify_cert_hostname" now do not permit multi-component wildcard
831 JH/10 Time-related extraction expansions from certificates now use the main
832 option "timezone" setting for output formatting, and are consistent
833 between OpenSSL and GnuTLS compilations. Bug 1541.
835 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
836 encoded parameter in the incoming message. Bug 1558.
838 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
839 include certificate info, eximon was claiming there were spoolfile
842 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
844 JH/14 Log delivery-related information more consistently, using the sequence
845 "H=<name> [<ip>]" wherever possible.
847 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
848 are problematic for Debian distribution, omit them from the release
851 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
853 JH/16 Fix string representation of time values on 64bit time_t architectures.
856 JH/17 Fix a null-indirection in certextract expansions when a nondefault
857 output list separator was used.
862 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
863 checkers that were complaining about end of non-void function with no
866 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
867 This was a regression introduced in 4.83 by another bugfix.
869 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
871 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
872 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
878 TF/01 Correctly close the server side of TLS when forking for delivery.
880 When a message was received over SMTP with TLS, Exim failed to clear up
881 the incoming connection properly after forking off the child process to
882 deliver the message. In some situations the subsequent outgoing
883 delivery connection happened to have the same fd number as the incoming
884 connection previously had. Exim would try to use TLS and fail, logging
885 a "Bad file descriptor" error.
887 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
888 utilities have not been installed.
890 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
891 temporary space as the ACL may create new global variables.
893 TL/01 LDAP support uses per connection or global context settings, depending
894 upon the detected version of the libraries at build time.
896 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
897 to extract and use the src ip:port in logging and expansions as if it
898 were a direct connection from the outside internet. PPv2 support was
899 updated based on HAProxy spec change in May 2014.
901 JH/02 Add ${listextract {number}{list}{success}{fail}}.
903 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
904 Properly escape header and check for NULL return.
906 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
909 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
911 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
912 characters in header names, implemented as a verify condition.
913 Contributed by Michael Fischer v. Mollard.
915 TL/05 Rename SPF condition results err_perm and err_temp to standardized
916 results permerror and temperror. Previous values are deprecated but
917 still accepted. In a future release, err_perm and err_temp will be
918 completely removed, which will be a backward incompatibility if the
919 ACL tests for either of these two old results. Patch contributed by
920 user bes-internal on the mailing list.
922 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
924 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
925 selectors, in both main and reject logs.
927 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
930 JH/07 Add malware type "sock" for talking to simple daemon.
932 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
934 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
935 routers/transports under cutthrough routing.
937 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
938 numbers. Touch up "bool" conditional to keep the same definition.
940 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
942 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
944 JH/12 Expand items in router/transport headers_add or headers_remove lists
945 individually rather than the list as a whole. Bug 1452.
947 Required for reasonable handling of multiple headers_ options when
948 they may be empty; requires that headers_remove items with embedded
949 colons must have them doubled (or the list-separator changed).
951 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
952 view the policy declared in the DMARC record. Currently, $dmarc_status
953 is a combined value of both the record presence and the result of the
956 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
958 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
959 dnslookup router and the smtp transport (applying to the forward
962 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
963 of ldap servers used for a specific lookup. Patch provided by Heiko
966 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
967 New variable $lookup_dnssec_authenticated for observability.
969 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
970 Patch submitted by Lars Timman.
972 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
974 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
975 Requires trusted mode and valid format message id, aborts otherwise.
976 Patch contributed by Heiko Schlichting.
978 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
979 certextract with support for various fields. Bug 1358.
981 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
982 is requested by default, modifiable by smtp transport option
985 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
986 operate on certificate variables to give certificate fingerprints
987 Also new ${sha256:cert_variable}.
989 JH/23 The PRDR feature is moved from being Experimental into the mainline.
991 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
992 Christian Aistleitner.
994 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
996 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
997 file. Patch from Wolfgang Breyha.
999 JH/25 Expand the coverage of the delivery $host and $host_address to
1000 client authenticators run in verify callout. Bug 1476.
1002 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1003 align with daemon_smtp_ports. Bug 72.
1005 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1006 support and error reporting did not work properly.
1008 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1009 and is readable. Patch from Andrew Colin Kissa.
1011 TL/14 Enhance documentation of ${run expansion and how it parses the
1012 commandline after expansion, particularly in the case when an
1013 unquoted variable expansion results in an empty value.
1015 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1017 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1018 by renaming away from it. Addresses GH issue 15, reported by
1021 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1023 TL/15 SECURITY: prevent double expansion in math comparison functions
1024 (can expand unsanitized data). Not remotely exploitable.
1031 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1033 PP/02 Make -n do something, by making it not do something.
1034 When combined with -bP, the name of an option is not output.
1036 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1039 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1040 $sender_host_name and config options to manage this, and basic check
1043 PP/05 DSCP support for outbound connections and control modifier for inbound.
1045 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1046 (Only plugin which currently uses this is kerberos4, which nobody should
1047 be using, but we should make it available and other future plugins might
1048 conceivably use it, even though it would break NAT; stuff *should* be
1049 using channel bindings instead).
1051 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1052 name; added for Sendmail compatibility; requires admin caller.
1053 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1054 just ignore it); requires trusted caller.
1055 Also parse but ignore: -Ac -Am -X<logfile>
1058 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1060 TL/02 Add +smtp_confirmation as a default logging option.
1062 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1063 Patch by Magnus Holmgren from 2007-02-20.
1065 TL/04 Bugzilla 1281 - Spec typo.
1066 Bugzilla 1283 - Spec typo.
1067 Bugzilla 1290 - Spec grammar fixes.
1069 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1071 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1073 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1074 mailing list by by Dmitry Isaikin.
1076 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1078 JH/02 Support "G" suffix to numbers in ${if comparisons.
1080 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1082 NM/01 Bugzilla 1197 - Spec typo
1083 Bugzilla 1196 - Spec examples corrections
1085 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1087 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1088 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1091 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1092 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1094 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1095 "acl {{name}{arg}...}", and optional args on acl condition
1098 JH/05 Permit multiple router/transport headers_add/remove lines.
1100 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1102 JH/07 Avoid using a waiting database for a single-message-only transport.
1103 Performance patch from Paul Fisher. Bugzilla 1262.
1105 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1108 JH/09 Add $headers_added variable, with content from use of ACL modifier
1109 add_header (but not yet added to the message). Bugzilla 199.
1111 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1112 Pulled from Bugzilla 817 by Wolfgang Breyha.
1114 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1116 (nb: this is the same fix as in Exim 4.80.1)
1118 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1121 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1124 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1126 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1127 advertise SMTP AUTH mechanism to us, instead of a generic
1128 protocol violation error. Also, make Exim more robust to bad
1129 data from the Dovecot auth socket.
1131 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1133 When a queue runner is handling a message, Exim first routes the
1134 recipient addresses, during which it prunes them based on the retry
1135 hints database. After that it attempts to deliver the message to
1136 any remaining recipients. It then updates the hints database using
1139 So if a recipient address works intermittently, it can get repeatedly
1140 deferred at routing time. The retry hints record remains fresh so the
1141 address never reaches the final cutoff time.
1143 This is a fairly common occurrence when a user is bumping up against
1144 their storage quota. Exim had some logic in its local delivery code
1145 to deal with this. However it did not apply to per-recipient defers
1146 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1148 This change adds a proper retry rule check during routing so that the
1149 final cutoff time is checked against the message's age. We only do
1150 this check if there is an address retry record and there is not a
1151 domain retry record; this implies that previous attempts to handle
1152 the address had the retry_use_local_parts option turned on. We use
1153 this as an approximation for the destination being like a local
1154 delivery, as in LMTP.
1156 I suspect this new check makes the old local delivery cutoff check
1157 redundant, but I have not verified this so I left the code in place.
1159 TF/02 Correct gecos expansion when From: is a prefix of the username.
1161 Test 0254 submits a message to Exim with the header
1165 When I ran the test suite under the user fanf2, Exim expanded
1166 the header to contain my full name, whereas it should have added
1167 a Resent-Sender: header. It erroneously treats any prefix of the
1168 username as equal to the username.
1170 This change corrects that bug.
1172 GF/01 DCC debug and logging tidyup
1173 Error conditions log to paniclog rather than rejectlog.
1174 Debug lines prefixed by "DCC: " to remove any ambiguity.
1176 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1178 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1179 Bug spotted by Jeremy Harris; was flawed since initial commit.
1180 Would have resulted in OCSP responses post-SNI triggering an Exim
1181 NULL dereference and crash.
1183 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1185 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1186 Bug detection, analysis and fix by Samuel Thibault.
1187 Bugzilla 1331, Debian bug #698092.
1189 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1191 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1192 Server implementation by Todd Lyons, client by JH.
1193 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1194 config variable "prdr_enable" controls whether the server
1195 advertises the facility. If the client requests PRDR a new
1196 acl_data_smtp_prdr ACL is called once for each recipient, after
1197 the body content is received and before the acl_smtp_data ACL.
1198 The client is controlled by both of: a hosts_try_prdr option
1199 on the smtp transport, and the server advertisement.
1200 Default client logging of deliveries and rejections involving
1201 PRDR are flagged with the string "PRDR".
1203 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1204 fclose(). Diagnosis by Todd Lyons.
1206 PP/17 Update configure.default to handle IPv6 localhost better.
1207 Patch by Alain Williams (plus minor tweaks).
1210 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1211 This is now consistent with GnuTLS, and is now documented: the
1212 previous undocumented portable approach to treating the option as
1213 unset was to force an expansion failure. That still works, and
1214 an empty string is now equivalent.
1216 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1217 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1218 not performing validation itself.
1220 PP/20 Added force_command boolean option to pipe transport.
1221 Patch from Nick Koston, of cPanel Inc.
1223 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1226 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1228 PP/21 Fix eximon continuous updating with timestamped log-files.
1229 Broken in a format-string cleanup in 4.80, missed when I repaired the
1230 other false fix of the same issue.
1231 Report and fix from Heiko Schlichting.
1234 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1235 Report from Prashanth Katuri.
1237 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1238 It's SecureTransport, so affects any MacOS clients which use the
1239 system-integrated TLS libraries, including email clients.
1241 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1242 using a MIME ACL for non-SMTP local injection.
1243 Report and assistance in diagnosis by Warren Baker.
1245 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1247 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1249 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1250 last id that failed so it may be referenced in subsequent ACL's.
1252 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1255 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1256 ldap library initialization, allowing self-signed CA's to be
1257 used. Also properly sets require_cert option later in code by
1258 using NULL (global ldap config) instead of ldap handle (per
1259 session). Bug diagnosis and testing by alxgomz.
1261 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1262 the src/util/ subdirectory.
1264 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1265 renamed to Transport Post Delivery Action by Jeremy Harris, as
1268 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1269 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1270 redis_servers = needs to be configured which will be used by the redis
1271 lookup. Patch from Warren Baker, of The Packet Hub.
1273 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1275 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1276 hostname or reverse DNS when processing a host list. Used suggestions
1277 from multiple comments on this bug.
1279 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1281 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1282 Missed a few lines, added it to make the runtest require no keyboard
1285 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1286 contains upper case chars. Make router use caseful_local_part.
1288 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1289 support when GnuTLS has been built with p11-kit.
1295 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1297 This, or similar/improved, will also be change PP/11 of 4.82.
1303 PP/01 Handle short writes when writing local log-files.
1304 In practice, only affects FreeBSD (8 onwards).
1305 Bugzilla 1053, with thanks to Dmitry Isaikin.
1307 NM/01 Bugzilla 949 - Documentation tweak
1309 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1312 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1314 PP/02 Implemented gsasl authenticator.
1316 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1318 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1319 `pkg-config foo` for cflags/libs.
1321 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1322 with rest of GSASL and with heimdal_gssapi.
1324 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1325 `pkg-config foo` for cflags/libs for the TLS implementation.
1327 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1328 properties get this fed in as external SSF. A number of robustness
1329 and debugging improvements to the cyrus_sasl authenticator.
1331 PP/08 cyrus_sasl server now expands the server_realm option.
1333 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1334 Patch by Jeremy Harris.
1336 PP/10 Added dbmjz lookup type.
1338 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1340 PP/12 MAIL args handles TAB as well as SP, for better interop with
1341 non-compliant senders.
1342 Analysis and variant patch by Todd Lyons.
1344 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1345 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1346 Patch from Dirk Mueller <dmueller@suse.com>
1348 PP/13 tls_peerdn now print-escaped for spool files.
1349 Observed some $tls_peerdn in wild which contained \n, which resulted
1350 in spool file corruption.
1352 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1353 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1354 or write after TLS renegotiation, which otherwise led to messages
1357 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1358 as a tracking header (ie: a signed header comes before the signature).
1359 Patch from Wolfgang Breyha.
1361 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1362 comma-sep list; embedded commas doubled.
1364 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1366 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1368 Report and patch from Dmitry Banschikov.
1370 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1371 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1372 failures. We appear to now support TLS1.1+ with Exim.
1374 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1375 lets Exim select keys and certificates based upon TLS SNI from client.
1376 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1377 before an outbound SMTP session. New log_selector, +tls_sni.
1379 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1380 NULL dereference. Report and patch from Alun Jones.
1382 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1383 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1384 Not seeing resolver debug output on NetBSD, but suspect this is a
1385 resolver implementation change.
1387 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1388 Left warnings. Added "eximon gdb" invocation mode.
1390 PP/21 Defaulting "accept_8bitmime" to true, not false.
1392 PP/22 Added -bw for inetd wait mode support.
1394 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1395 locate the relevant includes and libraries. Made this the default.
1397 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1398 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1400 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1401 This may cause build issues on older platforms.
1403 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1404 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1405 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1406 Added SNI support via GnuTLS too.
1407 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1409 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1411 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1412 Applied second patch from Janne, implementing suggestion to default
1413 multiple-strings-in-record handling to match SPF spec.
1415 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1417 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1418 read-only, out of scope).
1419 Patch from Wolfgang Breyha, report from Stuart Northfield.
1421 PP/29 Fix three issues highlighted by clang analyser static analysis.
1422 Only crash-plausible issue would require the Cambridge-specific
1423 iplookup router and a misconfiguration.
1424 Report from Marcin Mirosław.
1426 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1428 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1429 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1430 As part of this, removing so much warning spew let me fix some minor
1431 real issues in debug logging.
1433 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1434 assignment on my part. Fixed.
1436 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1437 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1438 Janne Snabb (who went above and beyond: thank you).
1440 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1441 string otherwise requires a connection and a bunch more work and it's
1442 relatively easy to get wrong. Should also expose TLS library linkage
1445 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1446 64-bit ${eval} (JH/03).
1448 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1449 GNU libc to support some of the 64-bit stuff, should not lead to
1450 conflicts. Defined before os.h is pulled in, so if a given platform
1451 needs to override this, it can.
1453 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1454 protection layer was required, which is not implemented.
1455 Bugzilla 1254, patch from Wolfgang Breyha.
1457 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1458 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1459 tls_dhparam take prime identifiers. Also unbreak combination of
1460 OpenSSL+DH_params+TLSSNI.
1462 PP/39 Disable SSLv2 by default in OpenSSL support.
1468 PP/01 Solaris build fix for Oracle's LDAP libraries.
1469 Bugzilla 1109, patch from Stephen Usher.
1471 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1473 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1476 TF/02 Fix a couple more cases where we did not log the error message
1477 when unlink() failed. See also change 4.74-TF/03.
1479 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1480 lock up or crash if it happened to be inside a call to libc when it
1481 got a SIGUSR1 from exiwhat.
1483 The SIGUSR1 handler appends the current process status to the process
1484 log which is later printed by exiwhat. It used to use the general
1485 purpose logging code to do this, but several functions it calls are
1486 not safe for signals.
1488 The new output code in the SIGUSR1 handler is specific to the process
1489 log, and simple enough that it's easy to inspect for signal safety.
1490 Removing some special cases also simplifies the general logging code.
1491 Removing the spurious timestamps from the process log simplifies
1494 TF/04 Improved ratelimit ACL condition.
1496 The /noupdate option has been deprecated in favour of /readonly which
1497 has clearer semantics. The /leaky, /strict, and /readonly update modes
1498 are mutually exclusive. The update mode is no longer included in the
1499 database key; it just determines when the database is updated. (This
1500 means that when you upgrade Exim will forget old rate measurements.)
1502 Exim now checks that the per_* options are used with an update mode that
1503 makes sense for the current ACL. For example, when Exim is processing a
1504 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1505 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1506 must specify per_mail/readonly. If you omit the update mode it defaults to
1507 /leaky where that makes sense (as before) or /readonly where required.
1509 The /noupdate option is now undocumented but still supported for
1510 backwards compatibility. It is equivalent to /readonly except that in
1511 ACLs where /readonly is required you may specify /leaky/noupdate or
1512 /strict/noupdate which are treated the same as /readonly.
1514 A useful new feature is the /count= option. This is a generalization
1515 of the per_byte option, so that you can measure the throughput of other
1516 aggregate values. For example, the per_byte option is now equivalent
1517 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1519 The per_rcpt option has been generalized using the /count= mechanism
1520 (though it's more complicated than the per_byte equivalence). When it is
1521 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1522 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1523 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1524 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1525 Note that using per_rcpt with a non-readonly update mode in more than
1526 one ACL will cause the recipients to be double-counted. (The per_mail
1527 and per_byte options don't have this problem.)
1529 The handling of very low rates has changed slightly. If the computed rate
1530 is less than the event's count (usually one) then this event is the first
1531 after a long gap. In this case the rate is set to the same as this event's
1532 count, so that the first message of a spam run is counted properly.
1534 The major new feature is a mechanism for counting the rate of unique
1535 events. The new per_addr option counts the number of different
1536 recipients that someone has sent messages to in the last time period. It
1537 behaves like per_rcpt if all the recipient addresses are different, but
1538 duplicate recipient addresses do not increase the measured rate. Like
1539 the /count= option this is a general mechanism, so the per_addr option
1540 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1541 example, measure the rate that a client uses different sender addresses
1542 with the options per_mail/unique=$sender_address. There are further
1543 details in the main documentation.
1545 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1547 TF/06 Removed a few PCRE remnants.
1549 TF/07 Automatically extract Exim's version number from tags in the git
1550 repository when doing development or release builds.
1552 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1553 Bugzilla 879. Patch from Paul Fisher.
1555 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1556 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1559 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1561 PP/05 Variable $av_failed, true if the AV scanner deferred.
1562 Bugzilla 1078. Patch from John Horne.
1564 PP/06 Stop make process more reliably on build failure.
1565 Bugzilla 1087. Patch from Heiko Schlittermann.
1567 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1568 Bugzilla 1089. Patch from Heiko Schlittermann.
1570 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1571 Bugzilla 1131. Patch from Holger Weiß.
1573 PP/09 Handle IPv6 addresses with SPF.
1574 Bugzilla 860. Patch from Wolfgang Breyha.
1576 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1578 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1581 PP/11 match_* no longer expand right-hand-side by default.
1582 New compile-time build option, EXPAND_LISTMATCH_RHS.
1583 New expansion conditions, "inlist", "inlisti".
1585 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1587 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1589 PP/14 fix log_write() format string regression from TF/03.
1590 Bugzilla 1152. Patch from Dmitry Isaikin.
1596 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1598 PP/02 Harmonised TLS library version reporting; only show if debugging.
1599 Layout now matches that introduced for other libraries in 4.74 PP/03.
1601 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1603 PP/04 New "dns_use_edns0" global option.
1605 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1608 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1609 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1611 TK/01 Updated PolarSSL code to 0.14.2.
1612 Bugzilla 1097. Patch from Andreas Metzler.
1614 PP/07 Catch divide-by-zero in ${eval:...}.
1615 Fixes bugzilla 1102.
1617 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1620 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1621 format-string attack -- SECURITY: remote arbitrary code execution.
1623 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1624 time unintentionally subject to list matching rules, letting the header
1625 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1626 arbitrary string expansion). This allowed for information disclosure.
1628 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1629 INT_MIN/-1 -- value coerced to INT_MAX.
1635 NM/01 Workaround for PCRE version dependency in version reporting
1638 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1639 This fixes portability to compilers other than gcc, notably
1640 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1642 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1643 makefiles for portability to HP-UX and POSIX correctness.
1645 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1646 Also via indented variable definition in the Makefile.
1647 (Debugging by Oliver Heesakkers).
1649 PP/02 Restore caching of spamd results with expanded spamd_address.
1650 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1652 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1653 Improves build reliability. Fix from: Frank Elsner
1655 NM/02 Fix wide character breakage in the rfc2047 coding
1656 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1658 NM/03 Allow underscore in dnslist lookups
1659 Fixes bug 1026. Patch from Graeme Fowler
1661 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1662 Code patches from Adam Ciarcinski of NetBSD.
1664 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1667 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1668 is logged, to avoid truncation. Patch from John Horne.
1670 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1671 Patch from Jakob Hirsch.
1673 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1674 SQL string expansion failure details.
1675 Patch from Andrey Oktyabrski.
1677 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1678 Patch from Simon Arlott.
1680 PP/09 New lookups functionality failed to compile on old gcc which rejects
1681 extern declarations in function scope.
1682 Patch from Oliver Fleischmann
1684 PP/10 Use sig_atomic_t for flags set from signal handlers.
1685 Check getgroups() return and improve debugging.
1686 Fixed developed for diagnosis in bug 927 (which turned out to be
1689 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1690 Patch from Mark Zealey.
1692 PP/12 Bugzilla 1056: Improved spamd server selection.
1693 Patch from Mark Zealey.
1695 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1696 Based on patch from Heiko Schlittermann.
1698 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1699 Patch from Uwe Doering, sign-off by Michael Haardt.
1701 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1702 variable declaration deep within a block. Bug and patch from
1705 PP/15 lookups-Makefile IRIX compatibility coercion.
1707 PP/16 Make DISABLE_DKIM build knob functional.
1709 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1710 Patch by Simon Arlott
1712 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1713 variable argument macros. Our copy now differs from upstream.
1719 TF/01 Failure to get a lock on a hints database can have serious
1720 consequences so log it to the panic log.
1722 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1723 controlled using the smtp_confirmation log selector.
1725 TF/03 Include the error message when we fail to unlink a spool file.
1727 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1728 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1729 for maintaining out-of-tree patches for some time.
1731 PP/01 Bugzilla 139: Documentation and portability issues.
1732 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1733 Handle per-OS dynamic-module compilation flags.
1735 PP/02 Let /dev/null have normal permissions.
1736 The 4.73 fixes were a little too stringent and complained about the
1737 permissions on /dev/null. Exempt it from some checks.
1738 Reported by Andreas M. Kirchwitz.
1740 PP/03 Report version information for many libraries, including
1741 Exim version information for dynamically loaded libraries. Created
1742 version.h, now support a version extension string for distributors
1743 who patch heavily. Dynamic module ABI change.
1745 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1746 privilege escalation vulnerability whereby the Exim run-time user
1747 can cause root to append content of the attacker's choosing to
1750 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1753 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1754 If dropping privileges for untrusted macros, we disabled normal logging
1755 on the basis that it would fail; for the Exim run-time user, this is not
1756 the case, and it resulted in successful deliveries going unlogged.
1757 Fixed. Reported by Andreas Metzler.
1763 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1764 only prepend for the Resent-* case. Fixes regression introduced in
1765 Exim 4.70 by NM/22 for Bugzilla 607.
1767 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1768 increasing numbers of administrators be bitten by this.
1770 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1772 PP/03 Bugzilla 994: added openssl_options main configuration option.
1774 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1776 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1778 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1780 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1781 without a peer certificate, leading to a segfault because of an
1782 assumption that peers always have certificates. Be a little more
1783 paranoid. Problem reported by Martin Tscholak.
1785 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1786 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1787 NB: ClamAV planning to remove STREAM in "middle of 2010".
1788 CL also introduces -bmalware, various -d+acl logging additions and
1789 more caution in buffer sizes.
1791 PP/09 Implemented reverse_ip expansion operator.
1793 PP/10 Bugzilla 937: provide a "debug" ACL control.
1795 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1797 PP/12 Bugzilla 973: Implement --version.
1799 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1801 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1803 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1805 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1806 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1807 ignore trailing whitespace.
1809 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1811 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1814 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1815 Notification from Dr Andrew Aitchison.
1817 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1818 ExtendedDetectionInfo response format.
1819 Notification from John Horne.
1821 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1824 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1825 XSL and documented dependency on system catalogs, with examples of how
1828 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1831 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1832 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1833 configuration file which is writeable by the Exim user or group.
1835 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1836 of configuration files to cover files specified with the -C option if
1837 they are going to be used with root privileges, not just the default
1840 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1841 option (effectively making it always true).
1843 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1844 files to be used while preserving root privileges.
1846 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1847 that rogue child processes cannot use them.
1849 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1850 run-time user, instead of root.
1852 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1853 Exim run-time user without dropping privileges.
1855 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1856 result string, instead of calling string_vformat() twice with the same
1859 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1860 for other users. Others should always drop root privileges if they use
1861 -C on the command line, even for a whitelisted configure file.
1863 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1865 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1871 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1872 $data_path, and $header_path variables; fixed documentation bugs and
1875 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1876 exipick to access non-standard spools, including the "frozen" queue
1879 NM/01 Bugzilla 965: Support mysql stored procedures.
1880 Patch from Alain Williams
1882 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1884 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1885 Patch from Andreas Metzler
1887 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1888 Patch from Kirill Miazine
1890 NM/05 Bugzilla 671: Added umask to procmail example.
1892 JJ/03 installed exipick 20100323.0, fixing doc bug
1894 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1895 directory. Notification and patch from Dan Rosenberg.
1897 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1899 TK/02 Improve log output when DKIM signing operation fails.
1901 MH/01 Treat the transport option dkim_domain as a colon separated
1902 list, not as a single string, and sign the message with each element,
1903 omitting multiple occurences of the same signer.
1905 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1906 Bugzilla 985, 986. Patch by Simon Arlott
1908 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1909 Patch by Simon Arlott
1911 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1912 MBX locking. Notification from Dan Rosenberg.
1918 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1920 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1922 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1924 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1926 NM/04 Bugzilla 915: Flag broken perl installation during build.
1932 TK/01 Added patch by Johannes Berg that expands the main option
1933 "spamd_address" if it starts with a dollar sign.
1935 TK/02 Write list of recipients to X-Envelope-Sender header when building
1936 the mbox-format spool file for content scanning (suggested by Jakob
1939 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1940 (http://www.dcc-servers.net/) support via dccifd. Activated by
1941 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1943 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1944 by Mark Daniel Reidel <mr@df.eu>.
1946 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1947 When building exim an external PCRE library is now needed -
1948 PCRE is a system library on the majority of modern systems.
1949 See entry on PCRE_LIBS in EDITME file.
1951 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1952 conversation. Added nologin parameter to request.
1953 Patch contributed by Kirill Miazine.
1955 TF/01 Do not log submission mode rewrites if they do not change the address.
1957 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1959 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1960 log files in place. Contributed by Roberto Lima.
1962 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1964 TF/03 Bugzilla 615: When checking the local_parts router precondition
1965 after a local_part_suffix or local_part_prefix option, Exim now
1966 does not use the address's named list lookup cache, since this
1967 contains cached lookups for the whole local part.
1969 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1970 Robert Millan. Documentation is in experimental-spec.txt.
1972 TF/04 Bugzilla 668: Fix parallel build (make -j).
1974 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1976 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1977 Patch provided by Jan Srzednicki.
1979 TF/05 Leading white space used to be stripped from $spam_report which
1980 wrecked the formatting. Now it is preserved.
1982 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1983 that they are available at delivery time.
1985 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1987 TF/08 TLS error reporting now respects the incoming_interface and
1988 incoming_port log selectors.
1990 TF/09 Produce a more useful error message if an SMTP transport's hosts
1991 setting expands to an empty string.
1993 NM/06 Bugzilla 744: EXPN did not work under TLS.
1994 Patch provided by Phil Pennock.
1996 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1997 Patch provided by Richard Godbee.
1999 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2000 acl_smtp_notquit, added index entry.
2002 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2003 Patch provided by Eugene Bujak.
2005 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2006 accept(). Patch provided by Maxim Dounin.
2008 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2009 Patch provided by Phil Pennock.
2011 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2013 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2014 Patch provided by Brad "anomie" Jorsch.
2016 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2017 Patch provided by Dean Brooks.
2019 TK/05 Add native DKIM support (does not depend on external libraries).
2021 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2022 Patch provided by Graeme Fowler.
2024 NM/16 Bugzilla 851: Documentation example syntax fix.
2026 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2028 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2031 NM/19 Bugzilla 745: TLS version reporting.
2032 Patch provided by Phil Pennock.
2034 NM/20 Bugzilla 167: bool: condition support.
2035 Patch provided by Phil Pennock.
2037 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2038 clients. Patch provided by Phil Pennock.
2040 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2041 Patch provided by Brad "anomie" Jorsch.
2043 NM/23 Bugzilla 687: Fix misparses in eximstats.
2044 Patch provided by Heiko Schlittermann.
2046 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2047 Patch provided by Heiko Schlittermann.
2049 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2050 plus update to original patch.
2052 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2054 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2055 Patch provided by David Brownlee.
2057 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2059 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2061 NM/30 Bugzilla 888: TLS documentation bugfixes.
2063 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2065 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2066 Unlike the original bugzilla I have changed all shell scripts in src tree.
2068 NM/33 Bugzilla 898: Transport filter timeout fix.
2069 Patch by Todd Rinaldo.
2071 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2072 Patch by Serge Demonchaux.
2074 NM/35 Bugzilla 39: Base64 decode bug fixes.
2075 Patch by Jakob Hirsch.
2077 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2079 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2081 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2083 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2089 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2090 ALT-N's libdkim that I have put here:
2091 http://duncanthrax.net/exim-experimental/
2093 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2094 were called 'true' and it seems that C99 defines that as a reserved
2095 keyword to be used with 'bool' variable types. That means you could
2096 not include C99-style headers which use bools without triggering
2097 build errors in sieve.c.
2099 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2100 as mailq or other aliases. Changed the --help handling significantly
2101 to do whats expected. exim_usage() emits usage/help information.
2103 SC/01 Added the -bylocaldomain option to eximstats.
2105 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2107 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2109 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2115 PH/01 Another patch from the Sieve maintainer.
2117 PH/02 When an IPv6 address is converted to a string for single-key lookup
2118 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2119 dots are used instead of colons so that keys in lsearch files need not
2120 contain colons. This was done some time before quoting was made available
2121 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2122 (notated using the quote facility) so as to distinguish them from IPv4
2123 keys. This meant that lookups for IP addresses in host lists did not work
2124 for iplsearch lookups.
2126 This has been fixed by arranging for IPv6 addresses to be expressed with
2127 colons if the lookup type is iplsearch. This is not incompatible, because
2128 previously such lookups could never work.
2130 The situation is now rather anomalous, since one *can* have colons in
2131 ordinary lsearch keys. However, making the change in all cases is
2132 incompatible and would probably break a number of configurations.
2134 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2137 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2138 conversion specification without a maximum field width, thereby enabling
2139 a rogue spamd server to cause a buffer overflow. While nobody in their
2140 right mind would setup Exim to query an untrusted spamd server, an
2141 attacker that gains access to a server running spamd could potentially
2142 exploit this vulnerability to run arbitrary code as the Exim user.
2144 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2145 $primary_hostname instead of what libspf2 thinks the hosts name is.
2147 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2148 a directory entry by the name of the lookup key. Previously, if a
2149 symlink pointed to a non-existing file or a file in a directory that
2150 Exim lacked permissions to read, a lookup for a key matching that
2151 symlink would fail. Now it is enough that a matching directory entry
2152 exists, symlink or not. (Bugzilla 503.)
2154 PH/03 The body_linecount and body_zerocount variables are now exported in the
2157 PH/04 Added the $dnslist_matched variable.
2159 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2160 This means they are set thereafter only if the connection becomes
2163 PH/06 Added the client_condition to authenticators so that some can be skipped
2164 by clients under certain conditions.
2166 PH/07 The error message for a badly-placed control=no_multiline_responses left
2167 "_responses" off the end of the name.
2169 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2171 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2172 (without spaces) instead of just copying the configuration text.
2174 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2176 PH/11 Added $max_received_linelength.
2178 PH/12 Added +ignore_defer and +include_defer to host lists.
2180 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2181 way in which PCRE > 7.0 is built.
2183 PH/14 Implemented queue_only_load_latch.
2185 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2186 MAIL command. The effect was to mangle the value on 64-bit systems.
2188 PH/16 Another patch from the Sieve maintainer.
2190 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2192 PH/18 If a system quota error occurred while trying to create the file for
2193 a maildir delivery, the message "Mailbox is full" was not appended to the
2194 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2195 only to a quota excession during the actual writing of the file.
2197 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2198 characters?) which causes problems in log lines. The DN values are now
2199 passed through string_printing() before being added to log lines.
2201 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2202 and InterBase are left for another time.)
2204 PH/21 Added message_body_newlines option.
2206 PH/22 Guard against possible overflow in moan_check_errorcopy().
2208 PH/23 POSIX allows open() to be a macro; guard against that.
2210 PH/24 If the recipient of an error message contained an @ in the local part
2211 (suitably quoted, of course), incorrect values were put in $domain and
2212 $local_part during the evaluation of errors_copy.
2218 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2219 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2222 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2223 issue a MAIL command.
2225 PH/02 In an ACL statement such as
2227 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2229 if a client was not listed at all, or was listed with a value other than
2230 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2231 the condition was not true (as it should be), so access was not denied.
2232 The bug was that the ! inversion was incorrectly passed on to the second
2233 item. This has been fixed.
2235 PH/03 Added additional dnslists conditions == and =& which are different from
2236 = and & when the dns lookup returns more than one IP address.
2238 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2239 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2241 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2242 FSYNC, which compiles an option called disable_fsync that allows for
2243 bypassing fsync(). The documentation is heavily laced with warnings.
2245 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2247 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2248 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2249 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2250 including adding "make clean"; (3) Added -fPIC when compiling the test
2251 dynamically loaded module, to get rid of a warning.
2253 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2254 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2255 = 0s. The bug is otherwise harmless.
2257 PH/07 There was a bug in the dovecot authenticator such that the value of
2258 $auth1 could be overwritten, and so not correctly preserved, after a
2259 successful authentication. This usually meant that the value preserved by
2260 the server_setid option was incorrect.
2262 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2264 PH/09 Installed PCRE release 7.0.
2266 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2267 run for batched SMTP input. It is now run at the start of every message
2268 in the batch. While fixing this I discovered that the process information
2269 (output by running exiwhat) was not always getting set for -bs and -bS
2270 input. This is fixed, and it now also says "batched" for BSMTP.
2272 PH/11 Added control=no_pipelining.
2274 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2275 patch, slightly modified), and move the expansion of helo_data till after
2276 the connection is made in the smtp transport (so it can use these
2279 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2281 PH/14 Added log_selector = +pid.
2283 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2285 PH/16 Add ${if forany and ${if forall.
2287 PH/17 Added dsn_from option to vary the From: line in DSNs.
2289 PH/18 Flush SMTP output before performing a callout, unless control =
2290 no_callout_flush is set.
2292 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2293 was true (the default) a successful delivery failed to delete the retry
2294 item, thus causing premature timeout of the address. The bug is now
2297 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2299 PH/21 Long custom messages for fakedefer and fakereject are now split up
2300 into multiline responses in the same way that messages for "deny" and
2301 other ACL rejections are.
2303 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2304 with slight modification.
2306 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2307 draft, changing the syntax and factoring some duplicate code".
2309 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2310 for deliveries of the second and subsequent messages over the same SMTP
2313 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2314 ${reduce, with only minor "tidies".
2316 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2318 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2319 expansion side effects.
2321 PH/27 When a message times out after an over-quota error from an Exim-imposed
2322 quota, the bounce message says "mailbox is full". This message was not
2323 being given when it was a system quota that was exceeded. It now should
2326 MH/03 Made $recipients available in local_scan(). local_scan() already has
2327 better access to the recipient list through recipients_list[], but
2328 $recipients can be useful in postmaster-provided expansion strings.
2330 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2331 in the case of a MAIL command with additional options following the
2332 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2333 were accidentally chopped off.
2335 PH/29 SMTP synchronization checks are implemented when a command is read -
2336 there is a check that no more input is waiting when there shouldn't be
2337 any. However, for some commands, a delay in an ACL can mean that it is
2338 some time before the response is written. In this time, more input might
2339 arrive, invalidly. So now there are extra checks after an ACL has run for
2340 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2341 pipelining has not been advertised.
2343 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2345 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2346 returned in the SMTP response when smtp_return_error_details was false.
2347 This has been fixed.
2349 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2350 instead of the C I/O that was originally supplied, because problems were
2351 reported on Solaris.
2353 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2354 Exim which did not show up earlier: it was assuming that a call to
2355 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2356 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2357 was a macro that became an assignment, so it seemed to work. This has
2358 changed to a proper function call with a void return, hence the compile
2359 error. Exim's code has been fixed.
2361 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2364 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2366 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2368 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2369 because it turns out to be too big or there is a timeout) there is no
2370 "Completed" line in the log. When some messages of this type were
2371 selected by exigrep, they were listed as "not completed". Others were
2372 picked up by some special patterns. I have improved the selection
2373 criteria to be more general.
2375 PH/38 The host_find_failed option in the manualroute router can now be set
2376 to "ignore", to completely ignore a host whose IP address cannot be
2377 found. If all hosts are ignored, the behaviour is controlled by the new
2378 host_all_ignored option.
2380 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2381 homing or because of multiple MX records with /mx) generated more than
2382 one IP address, and the following item turned out to be the local host,
2383 all the secondary addresses of the first item were incorrectly removed
2384 from the list, along with the local host and any following hosts (which
2385 is what is supposed to happen).
2387 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2388 whoever called Exim into the -H file. In the case of the daemon it was
2389 behaving confusingly. When first started, it used values for whoever
2390 started the daemon, but after a SIGHUP it used the Exim user (because it
2391 calls itself on a restart). I have changed the code so that it now always
2394 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2395 message are rejected with the same error (e.g. no authentication or bad
2396 sender address), and a DATA command is nevertheless sent (as can happen
2397 with PIPELINING or a stupid MUA), the error message that was given to the
2398 RCPT commands is included in the rejection of the DATA command. This is
2399 intended to be helpful for MUAs that show only the final error to their
2402 PH/42 Another patch from the Sieve maintainer.
2404 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2405 Eximstats - Fixed some broken HTML links and added missing column headers
2407 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2408 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2410 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2412 PH/43 Yet another patch from the Sieve maintainer.
2414 PH/44 I found a way to check for a TCP/IP connection going away before sending
2415 the response to the final '.' that terminates a message, but only in the
2416 case where the client has not sent further data following the '.'
2417 (unfortunately, this is allowed). However, in many cases there won't be
2418 any further data because there won't be any more messages to send. A call
2419 to select() can be used: if it shows that the input is "ready", there is
2420 either input waiting, or the socket has been closed. An attempt to read
2421 the next input character can distinguish the two cases. Previously, Exim
2422 would have sent an OK response which the client would never have see.
2423 This could lead to message repetition. This fix should cure that, at
2424 least in a lot of common cases.
2426 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2427 advertised in response to EHLO.
2433 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2434 fixed by 4.65/MH/01 (is this a record?) are fixed:
2436 (i) An empty string was always treated as zero by the numeric comparison
2437 operators. This behaviour has been restored.
2439 (ii) It is documented that the numeric comparison operators always treat
2440 their arguments as decimal numbers. This was broken in that numbers
2441 starting with 0 were being interpreted as octal.
2443 While fixing these problems I realized that there was another issue that
2444 hadn't been noticed. Values of message_size_limit (both the global option
2445 and the transport option) were treated as octal if they started with 0.
2446 The documentation was vague. These values are now always treated as
2447 decimal, and I will make that clear in the documentation.
2453 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2454 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2457 MH/01 Don't check that the operands of numeric comparison operators are
2458 integers when their expansion is in "skipping" mode (fixes bug
2459 introduced by 4.64-PH/07).
2461 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2462 child addresses, Exim now panics and dies. Previously, because the count
2463 is held in a short int, deliveries were likely to be lost. As such a
2464 large number of recipients for a single message is ridiculous
2465 (performance will be very, very poor), I have chosen to impose a limit
2466 rather than extend the field.
2472 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2473 leftover -K file (the existence of which was triggered by #402).
2474 While we were at it, introduced process PID as part of the -K
2475 filename. This should rule out race conditions when creating
2478 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2479 processing considerably. Previous code took too long for large mails,
2480 triggering a timeout which in turn triggers #401.
2482 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2483 in the DK code in transports.c. sendfile() is not really portable,
2484 hence the _LINUX specificness.
2486 TF/01 In the add_headers option to the mail command in an Exim filter,
2487 there was a bug that Exim would claim a syntax error in any
2488 header after the first one which had an odd number of characters
2491 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2492 callout verification, Exim cached a "reject" for the entire domain. This
2493 is correct for most verifications, but it is not correct for a recipient
2494 verification with use_sender or use_postmaster set, because in that case
2495 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2496 case of MAIL FROM:<> rejection from other early rejections (e.g.
2497 rejection of HELO). When verifying a recipient using a non-null MAIL
2498 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2499 Whatever the result of the callout, the value of the domain cache is
2500 left unchanged (for any other kind of callout, getting as far as trying
2501 RCPT means that the domain itself is ok).
2503 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2506 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2507 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2508 session, the daemon ignores ECONNECT errors and logs others; it now
2509 ignores EPIPE as well.
2511 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2512 (quoted-printable decoding).
2514 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2515 later the small subsequent patch to fix an introduced bug.
2517 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2519 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2521 PH/08 An error is now given if message_size_limit is specified negative.
2523 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2524 to be given (somewhat) arbitrary names.
2526 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2529 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2530 miscellaneous code fixes
2532 PH/10 Added the log_reject_target ACL modifier to specify where to log
2535 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2536 hostname. This is wrong, because it relates to the incoming message (and
2537 probably the interface on which it is arriving) and not to the outgoing
2538 callout (which could be using a different interface). This has been
2539 changed to use the value of the helo_data option from the smtp transport
2540 instead - this is what is used when a message is actually being sent. If
2541 there is no remote transport (possible with a router that sets up host
2542 addresses), $smtp_active_hostname is used.
2544 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2545 tweaks were necessary in order to get it to work (see also 21 below):
2546 (a) The code assumed that strncpy() returns a negative number on buffer
2547 overflow, which isn't the case. Replaced with Exim's string_format()
2549 (b) There were several signed/unsigned issues. I just did the minimum
2550 hacking in of casts. There is scope for a larger refactoring.
2551 (c) The code used strcasecmp() which is not a standard C function.
2552 Replaced with Exim's strcmpic() function.
2553 (d) The code set only $1; it now sets $auth1 as well.
2554 (e) A simple test gave the error "authentication client didn't specify
2555 service in request". It would seem that Dovecot has changed its
2556 interface. Fortunately there's a specification; I followed it and
2557 changed what the client sends and it appears to be working now.
2559 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2562 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2563 address A is aliased to B and C, where B exists and C does not. Without
2564 -v the output is "A verified" because verification stops after a
2565 successful redirection if more than one address is generated. However,
2566 with -v the child addresses are also verified. Exim was outputting "A
2567 failed to verify" and then showing the successful verification for C,
2568 with its parentage. It now outputs "B failed to verify", showing B's
2569 parentage before showing the successful verification of C.
2571 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2572 look up a TXT record in a specific list after matching in a combined
2575 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2576 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2577 they consult the DNS. I had assumed they would set it the way they
2578 wanted; and indeed my experiments on Linux seem to show that in some
2579 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2580 To be on the safe side, however, I have now made the interface to
2581 host_find_byname() similar to host_find_bydns(), with an argument
2582 containing the DNS resolver options. The host_find_byname() function now
2583 sets these options at its start, just as host_find_bydns() does. The smtp
2584 transport options dns_qualify_single and dns_search_parents are passed to
2585 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2586 of host_find_byname() use the default settings of RES_DEFNAMES
2587 (qualify_single) but not RES_DNSRCH (search_parents).
2589 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2590 spool_read_header() do less string testing, by means of a preliminary
2591 switch on the second character of optional "-foo" lines. (This is
2592 overdue, caused by the large number of possibilities that now exist.
2593 Originally there were few.) While I was there, I also converted the
2594 str(n)cmp tests so they don't re-test the leading "-" and the first
2595 character, in the hope this might squeeze out yet more improvement.
2597 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2598 flag allowing group syntax was set by the header_syntax check but not
2599 turned off, possible causing trouble later; (2) The flag was not being
2600 set at all for the header_verify test, causing "group"-style headers to
2601 be rejected. I have now set it in this case, and also caused header_
2602 verify to ignore an empty address taken from a group. While doing this, I
2603 came across some other cases where the code for allowing group syntax
2604 while scanning a header line wasn't quite right (mostly, not resetting
2605 the flag correctly in the right place). These bugs could have caused
2606 trouble for malformed header lines. I hope it is now all correct.
2608 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2609 with the "reply" argument non-NULL. The code, however (which originally
2610 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2611 but it didn't always do it. This confused somebody who was copying the
2612 code for some other use. I have removed all the tests.
2614 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2615 feature that was used to support insecure browsers during the U.S. crypto
2616 embargo. It requires special client support, and Exim is probably the
2617 only MTA that supported it -- and would never use it because real RSA is
2618 always available. This code has been removed, because it had the bad
2619 effect of slowing Exim down by computing (never used) parameters for the
2620 RSA_EXPORT functionality.
2622 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2623 authenticator to fail if there's a tab character in the incoming data
2624 (there should never be unless someone is messing about, as it's supposed
2625 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2626 if the connection is using TLS or if the remote IP is the same as the
2627 local IP, and the "valid-client-cert option" if a client certificate has
2630 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2631 authenticators. This can be used for authorization after authentication
2632 succeeds. (In the case of plaintext, it servers for both authentication
2635 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2636 if any retry times were supplied.
2638 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2639 connection, where there is no client IP address to check. In this
2640 situation, the verify now always succeeds.
2642 PH/25 Applied John Jetmore's -Mset patch.
2644 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2646 PH/27 In a string expansion for a processed (not raw) header when multiple
2647 headers of the same name were present, leading whitespace was being
2648 removed from all of them, but trailing whitespace was being removed only
2649 from the last one. Now trailing whitespace is removed from each header
2650 before concatenation. Completely empty headers in a concatenation (as
2651 before) are ignored.
2653 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2654 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2656 PH/29 [Removed. This was a change that I later backed out, and forgot to
2657 correct the ChangeLog entry (that I had efficiently created) before
2658 committing the later change.]
2660 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2661 address errors (4xx response to RCPT) over the same connection as other
2662 messages routed to the same hosts. Such deliveries are always "forced",
2663 so retry times are not inspected. This resulted in far too many retries
2664 for the affected addresses. The effect occurred only when there were more
2665 hosts than the hosts_max_try setting in the smtp transport when it had
2666 the 4xx errors. Those hosts that it had tried were not added to the list
2667 of hosts for which the message was waiting, so if all were tried, there
2668 was no problem. Two fixes have been applied:
2670 (i) If there are any address or message errors in an SMTP delivery, none
2671 of the hosts (tried or untried) are now added to the list of hosts
2672 for which the message is waiting, so the message should not be a
2673 candidate for sending over the same connection that was used for a
2674 successful delivery of some other message. This seems entirely
2675 reasonable: after all the message is NOT "waiting for some host".
2676 This is so "obvious" that I'm not sure why it wasn't done
2677 previously. Hope I haven't missed anything, but it can't do any
2678 harm, as the worst effect is to miss an optimization.
2680 (ii) If, despite (i), such a delivery is accidentally attempted, the
2681 routing retry time is respected, so at least it doesn't keep
2682 hammering the server.
2684 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2685 in ${readsocket because some servers need this prod.
2687 PH/32 Added some extra debug output when updating a wait-xxx database.
2689 PH/33 The hint "could be header name not terminated by colon", which has been
2690 given for certain expansion errors for a long time, was not being given
2691 for the ${if def:h_colon_omitted{... case.
2693 PH/34 The spec says: "With one important exception, whenever a domain list is
2694 being scanned, $domain contains the subject domain." There was at least
2695 one case where this was not true.
2697 PH/35 The error "getsockname() failed: connection reset by peer" was being
2698 written to the panic log as well as the main log, but it isn't really
2699 panic-worthy as it just means the connection died rather early on. I have
2700 removed the panic log writing for the ECONNRESET error when getsockname()
2703 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2704 runs only) independently of the message's sender address. This meant
2705 that, if the 4xx error was in fact related to the sender, a different
2706 message to the same recipient with a different sender could confuse
2707 things. In particular, this can happen when sending to a greylisting
2708 server, but other circumstances could also provoke similar problems.
2709 I have changed the default so that the retry time for these errors is now
2710 based a combination of the sender and recipient addresses. This change
2711 can be overridden by setting address_retry_include_sender=false in the
2714 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2715 remote server are returned as part of bounce messages. This was not
2716 happening for LMTP over a pipe (the lmtp transport), but now it is the
2717 same for both kinds of LMTP.
2719 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2720 in header lines that were in fact CNAMEs. This is no longer the case.
2722 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2723 and queue runs started by the daemon processed all messages. This has
2724 been fixed so that -R and -S can now usefully be given with -q<time>.
2726 PH/40 Import PCRE release 6.7 (fixes some bugs).
2728 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2730 PH/42 Give an error if -q is specified more than once.
2732 PH/43 Renamed the variables $interface_address and $interface_port as
2733 $received_ip_address and $received_port, to make it clear that these
2734 values apply to message reception, and not to the outgoing interface when
2735 a message is delivered. (The old names remain recognized, of course.)
2737 PH/44 There was no timeout on the connect() call when using a Unix domain
2738 socket in the ${readsocket expansion. There now is.
2740 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2741 be meaningful with "accept".
2743 SC/01 Eximstats V1.43
2744 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2746 SC/02 Eximstats V1.44
2747 Use a glob alias rather than an array ref in the generated
2748 parser. This improves both readability and performance.
2750 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2751 Collect SpamAssassin and rejection statistics.
2752 Don't display local sender or destination tables unless
2753 there is data to show.
2754 Added average volumes into the top table text output.
2756 SC/04 Eximstats V1.46
2757 Collect data on the number of addresses (recipients)
2758 as well as the number of messages.
2760 SC/05 Eximstats V1.47
2761 Added 'Message too big' to the list of mail rejection
2762 reasons (thanks to Marco Gaiarin).
2764 SC/06 Eximstats V1.48
2765 Mainlog lines which have GMT offsets and are too short to
2766 have a flag are now skipped.
2768 SC/07 Eximstats V1.49 (Alain Williams)
2769 Added the -emptyok flag.
2771 SC/08 Eximstats V1.50
2772 Fixes for obtaining the IP address from reject messages.
2774 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2775 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2776 whitespace changes from 4.64-PH/27
2778 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2781 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2782 are found, allow negative numbers in numeric criteria)
2784 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2786 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2787 to match changes made in 4.64-PH/43
2789 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2791 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2792 "domain" is actually an IP address, and force a failure. This locks out
2793 those revolvers/nameservers that support "A-for-A" lookups, in
2794 contravention of the specifications.
2796 PH/48 When a host name was looked up from an IP address, and the subsequent
2797 forward lookup of the name timed out, the host name was left in
2798 $sender_host_name, contrary to the specification.
2800 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2801 restricted to single-key lookups, Exim was not diagnosing an error if
2802 * or *@ was used with a query-style lookup.
2804 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2806 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2807 long ago, but noone interested enough thought of it. Let's just say that
2808 the "1.1" means that there are some new functions that weren't there at
2809 some point in the past.
2811 PH/51 Error processing for expansion failure of helo_data from an smtp
2812 transport during callout processing was broken.
2814 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2815 tested/used via the -bh/-bhc/-bs options.
2817 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2818 bug, fixed in subsequent PCRE releases).
2820 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2821 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2823 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2829 SC/01 Use a glob alias rather than an array ref in eximstats generated
2830 parser. This improves both readability and performance.
2832 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2833 Don't display local sender or destination tables in eximstats unless
2834 there is data to show.
2835 Added average volumes into the eximstats top table text output.
2837 SC/03 Collect data on the number of addresses (recipients) as well
2838 as the number of messages in eximstats.
2840 TF/01 Correct an error in the documentation for the redirect router. Exim
2841 does (usually) call initgroups() when daemonizing.
2843 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2844 with consistent privilege compared to when running as a daemon.
2846 TF/03 Note in the spec that $authenticated_id is not set for local
2847 submissions from trusted users.
2849 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2850 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2852 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2853 by adding some example configuration directives to the default
2854 configuration file. A little bit of work is required to uncomment the
2855 directives and define how usernames and passwords are checked, but
2856 there is now a framework to start from.
2858 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2859 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2860 without this. I don't know how relevant this is to other LDAP libraries.
2862 PH/02 Add the verb name to the "unknown ACL verb" error.
2864 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2866 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2868 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2869 directory not expanded when it should be if an expanded home directory
2870 was set for the address (which is overridden by the transport).
2872 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2875 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2876 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2877 because it is too late at that time, and has no effect.
2879 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2880 security issue with \' (bugzilla #107). I could not use the
2881 PQescapeStringConn() function, because it needs a PGconn value as one of
2884 PH/08 When testing addresses using -bt, indicate those final addresses that
2885 are duplicates that would not cause an additional delivery. At least one
2886 person was confused, thinking that -bt output corresponded to deliveries.
2887 (Suppressing duplicates isn't a good idea as you lose the information
2888 about possibly different redirections that led to the duplicates.)
2890 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2891 systems where poll() doesn't work, in particular OS X.
2893 PH/10 Added more information to debugging output for retry time not reached.
2895 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2896 operations in malware.c.
2898 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2901 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2902 syslog_duplication set false, log lines that would normally be written
2903 both the the main log and to the reject log were not written to syslog at
2906 PH/14 In the default configuration, change the use of "message" in ACL warn
2907 statements to "add_header".
2909 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2910 not followed by a command (e.g. "seen endif").
2912 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2913 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2916 PH/17 Added extra conditions to the default value of delay_warning_condition
2920 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2921 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2922 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2925 The Auto-Submitted: and various List- headers are standardised, whereas I
2926 don't think Precedence: ever was.
2928 PH/18 Refactored debugging code in route_finduser() to show more information,
2929 in particular, the error code if getpwnam() issues one.
2931 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2932 This is apparently needed in addition to the PH/07 change above to avoid
2933 any possible encoding problems.
2935 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2936 but not after initializing Perl.
2938 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2939 output them only if debugging. By default they are written stderr,
2940 apparently, which is not desirable.
2942 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2945 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2948 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2950 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2951 authenticated or an ident call has been made. Suppress the default
2952 values for $authenticated_id and $authenticated_sender (but permit -oMai
2953 and -oMas) when testing with -bh.
2955 PH/24 Re-jigged the order of the tests in the default configuration so that the
2956 tests for valid domains and recipients precede the DNS black list and CSA
2957 tests, on the grounds that those ones are more expensive.
2959 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2960 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2961 This bug exists in every version of Exim that I still have, right back to
2964 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2965 However, an attempt to turn on case-sensitivity in a regex key by
2966 including (?-i) didn't work because the subject string was already
2967 lowercased, and the effects were non-intuitive. It turns out that a
2968 one-line patch can be used to allow (?-i) to work as expected.
2974 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2975 other effects) broke the use of negated acl sub-conditions.
2977 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2980 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2981 "Deny" causes Exim to reject the incoming connection with a 554 error.
2982 Unfortunately, if there is a major crisis, such as a disk failure,
2983 tcp-wrappers gives "deny", whereas what one would like would be some
2984 kind of temporary error. A kludge has been added to help with this.
2985 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2986 554 error is used if errno is still zero or contains ENOENT (which occurs
2987 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2990 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2992 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2993 errors. Otherwise a message that provokes a temporary error (when other
2994 messages do not) can cause a whole host to time out.
2996 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2997 addresses were routed directly to files or pipes from a redirect router.
2998 File deliveries just didn't batch; pipe deliveries might have suffered
3001 PH/06 A failure to get a lock for a hints database would erroneously always say
3002 "Failed to get write lock", even when it was really a read lock.
3004 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3005 of 0600. This has been changed to use the value of the lockfile_mode
3006 option (which defaults to 0600).
3008 PH/08 Applied small patch from the Sieve maintainer.
3010 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3011 folder from quota calculations, a direct delivery into this folder messed
3012 up the contents of the maildirsize file. This was because the regex was
3013 used only to exclude .Trash (or whatever) when the size of the mailbox
3014 was calculated. There was no check that a delivery was happening into an
3015 excluded directory. This bug has been fixed by ignoring all quota
3016 processing for deliveries into excluded directories.
3018 PH/10 Added the maildirfolder_create_regex option to appendfile.
3024 PH/01 The code for finding all the local interface addresses on a FreeBSD
3025 system running IPv6 was broken. This may well have applied to all BSD
3026 systems, as well as to others that have similar system calls. The broken
3027 code found IPv4 interfaces correctly, but gave incorrect values for the
3028 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3029 that it would not match correctly against @[] and not recognize the IPv6
3032 PH/02 The ipliteral router was not recognizing addresses of the form user@
3033 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3035 PH/03 Added disable_ipv6.
3037 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3038 decoded form, because it is most often used to construct To: headers
3039 lines in autoreplies, and the decoded form may well be syntactically
3040 invalid. However, $reply_address has leading white space removed, and all
3041 newlines turned into spaces so that the autoreply transport does not
3044 PH/05 If group was specified without a user on a router, and no group or user
3045 was specified on a transport, the group from the router was ignored.
3047 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3048 for visible compile-time settings that can be used to change these
3049 numbers, for those that want even more. Backwards compatibility with old
3050 spool files has been maintained. However, going back to a previous Exim
3051 release will lost any variables that are in spool files.
3053 PH/07 Two small changes when running in the test harness: increase delay when
3054 passing a TCP/IP connection to a new process, in case the original
3055 process has to generate a bounce, and remove special handling of
3056 127.0.0.2 (sic), which is no longer necessary.
3058 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3059 be the same on different OS.
3061 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3064 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3065 whether --show-vars was specified or not
3067 JJ/02 exipick: Added support for new ACL variable spool format introduced
3070 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3071 syntactically invalid From: or Reply-to: line, and a filter used this to
3072 generate an autoreply, and therefore failed to obtain an address for the
3073 autoreply, Exim could try to deliver to a non-existent relative file
3074 name, causing unrelated and misleading errors. What now happens is that
3075 it logs this as a hard delivery error, but does not attempt to create a
3078 PH/11 The exinext utility has a -C option for testing purposes, but although
3079 the given file was scanned by exinext itself; it wasn't being passed on
3080 when Exim was called.
3082 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3083 an end-of-file indication when reading a command response.
3085 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3086 compiled. In many other places in Exim, IPv6 addresses are always
3087 recognized, so I have changed this. It also means that IPv4 domain
3088 literals of the form [IPV4:n.n.n.n] are now always recognized.
3090 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3091 used if the router is not running as root, for example, when verifying at
3092 ACL time, or when using -bh. The debugging output from this situation was
3093 non-existent - all you got was a failure to exec. I have made two
3096 (a) Failures to set uid/gid, the current directory, or a process leader
3097 in a subprocess such as that created by queryprogram now generate
3098 suitable debugging output when -d is set.
3100 (b) The queryprogram router detects when it is not running as root,
3101 outputs suitable debugging information if -d is set, and then runs
3102 the subprocess without attempting to change uid/gid.
3104 PH/15 Minor change to Makefile for building test_host (undocumented testing
3107 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3108 additional section of a DNS packet that returns MX or SRV records.
3109 Instead, it always explicitly searches for A/AAAA records. This avoids
3110 major problems that occur when a DNS server includes only records of one
3111 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3112 fixed another bug: if SRV records were looked up and the corresponding
3113 address records were *not* found in the additional section, the port
3114 values from the SRV records were lost.
3116 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3117 using the correct key (the original address) when searching the retry
3118 rules in order to find which one to use for generating the retry hint.
3120 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3121 adding the default one. Similarly, if it contains a Reply-To: header, the
3122 errors_reply_to option, if set, is not used.
3124 PH/19 When calculating a retry time, Exim used to measure the "time since
3125 failure" by looking at the "first failed" field in the retry record. Now
3126 it does not use this if it is later than than the arrival time of the
3127 message. Instead it uses the arrival time. This makes for better
3128 behaviour in cases where some deliveries succeed, thus re-setting the
3129 "first failed" field. An example is a quota failure for a huge message
3130 when small messages continue to be delivered. Without this change, the
3131 "time since failure" will always be short, possible causing more frequent
3132 delivery attempts for the huge message than are intended.
3133 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3135 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3136 $1, $2, $3) because the numerical variables can be reset during some
3137 expansion items (e.g. "match"), thereby losing the authentication data.
3139 PH/21 Make -bV show the size of off_t variables so that the test suite can
3140 decide whether to run tests for quotas > 2G.
3142 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3143 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3144 filecount value is greater than 2G or if a quota value is greater than 2G
3145 on a system where the size of off_t is not greater than 4, a panic error
3148 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3149 never match. The debug and -bh output now contains an explicit error
3150 message indicating a malformed IPv4 address or mask.
3152 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3153 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3154 PH/23 above applies.
3156 PH/25 Do not write to syslog when running in the test harness. The only
3157 occasion when this arises is a failure to open the main or panic logs
3158 (for which there is an explicit test).
3160 PH/26 Added the /no_tell option to "control=freeze".
3162 PH/27 If a host name lookup failed very early in a connection, for example, if
3163 the IP address matched host_lookup and the reverse lookup yielded a name
3164 that did not have a forward lookup, an error message of the form "no IP
3165 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3166 could be logged. Now it outputs the IP address instead of "NULL".
3168 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3169 allows the sender and the authenticated sender to be set when
3170 submitting a message from within Exim. Since child_open_exim() is
3171 documented for local_scan(), the new function should be too.
3173 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3174 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3175 results in an empty string is now treated as unset.
3177 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3179 PH/31 Added sender_verify_fail logging option.
3181 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3182 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3183 tidied the source and removed it altogether.
3185 PH/33 When a queue run was abandoned because the load average was too high, a
3186 log line was always written; now it is written only if the queue_run log
3187 selector is set. In addition, the log line for abandonment now contains
3188 information about the queue run such as the pid. This is always present
3189 in "start" and "stop" lines but was omitted from the "abandon" line.
3191 PH/34 Omit spaces between a header name and the colon in the error message that
3192 is given when verify = headers_syntax fails (if there are lots of them,
3193 the message gets confusing).
3195 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3196 names, as there are now some PTR records that contain slashes. This check
3197 is only to protect against broken name servers that fall over on strange
3198 characters, so the fact that it applies to all lookups doesn't matter.
3200 PH/36 Now that the new test suite is complete, we can remove some of the
3201 special code in Exim that was needed for the old test suite. For example,
3202 sorting DNS records because real resolvers return them in an arbitrary
3203 order. The new test suite's fake resolver always returns records in the
3206 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3207 bounces) except when queue_only is set, to avoid logging races between
3208 the different processes.
3210 PH/38 Panic-die if .include specifies a non-absolute path.
3212 PH/39 A tweak to the "H" retry rule from its user.
3214 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3215 a label. They prevented compilation on older perls.
3217 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3218 a warning to be raised on newish perls.
3220 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3221 on queue. Changes to match documented behaviour of showing count of
3222 messages matching specified criteria.
3224 PH/40 Changed the default ident timeout from 30s to 5s.
3226 PH/41 Added support for the use of login_cap features, on those BSD systems
3227 that have them, for controlling the resources used by pipe deliveries.
3229 PH/42 The content-scanning code uses fopen() to create files in which to put
3230 message data. Previously it was not paying any attention to the mode of
3231 the files. Exim runs with umask(0) because the rest of the code creates
3232 files with open(), and sets the required mode explicitly. Thus, these
3233 files were ending up world-writeable. This was not a big issue, because,
3234 being within the spool directory, they were not world-accessible. I have
3235 created a function called modefopen, which takes an additional mode
3236 argument. It sets umask(777), creates the file, chmods it to the required
3237 mode, then resets the umask. All the relevant calls to fopen() in the
3238 content scanning code have been changed to use this function.
3240 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3241 to 24 hours. This avoids potential overflow problems when processing G
3242 and H retry rules. I suspect nobody ever tinkers with this value.
3244 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3246 PH/45 When the plaintext authenticator is running as a client, the server's
3247 challenges are checked to ensure they are valid base64 strings. By
3248 default, the authentication attempt is cancelled if an invalid string is
3249 received. Setting client_ignore_invalid_base64 true ignores these errors.
3250 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3251 they are received. Thus, the responses can be made to depend on the
3252 challenges. If an invalid string is ignored, an empty string is placed in
3255 PH/46 Messages that are created by the autoreply transport now contains a
3256 References: header, in accordance with RFCs 2822 and 3834.
3258 PH/47 Added authenticated_sender_force to the smtp transport.
3260 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3262 PH/49 Installed latest patch from the Sieve maintainer.
3264 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3265 was also set, the appendfile transport was unnecessarily scanning a
3266 directory of message files (e.g. for maildir delivery) to find the count
3267 of files (along with the size), even though it did not need this
3268 information. It now does the scan only if it needs to find either the
3269 size of the count of files.
3271 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3273 PH/52 Two bugs concerned with error handling when the smtp transport is
3276 (i) Exim was not creating retry information for temporary errors given
3277 for individual recipients after the DATA command when the smtp transport
3278 was used in LMTP mode. This meant that they could be retried too
3279 frequently, and not timed out correctly.
3281 (ii) Exim was setting the flag that allows error details to be returned
3282 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3283 recipients that were returned after the DATA command.
3285 PH/53 This is related to PH/52, but is more general: for any failing address,
3286 when detailed error information was permitted to be returned to the
3287 sender, but the error was temporary, then after the final timeout, only
3288 "retry timeout exceeded" was returned. Now it returns the full error as
3289 well as "retry timeout exceeded".
3291 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3292 do this, and (what is worse) MTAs that accept it.
3294 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3295 will now be deprecated.
3297 PH/56 New os.c-cygwin from the Cygwin maintainer.
3299 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3300 formats (previously only available in exim formats via -bpr, -bpru,
3301 and -bpra. Now also available in native and exiqgrep formats)
3303 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3304 with very large, slow to parse queues
3306 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3308 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3310 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3311 responses to authentication challenges, though it was showing the
3312 challenges; (ii) I've removed the CR characters from the debug output for
3315 PH/58 Allow for the insertion of a newline as well as a space when a string
3316 is turned into more than one encoded-word during RFC 2047 encoding. The
3317 Sieve code now uses this.
3319 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3320 data_4xx, lost_connection, tls_required.
3322 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3323 message was being sent as an SMTP response.
3325 PH/61 Add -l and -k options to exicyclog.
3327 PH/62 When verifying, if an address was redirected to one new address, so that
3328 verification continued, and the new address failed or deferred after
3329 having set something in $address_data, the value of $address_data was not
3330 passed back to the ACL. This was different to the case when no
3331 redirection occurred. The value is now passed back in both cases.
3333 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3334 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3335 use login_cap.h, so on its own it isn't the distinguishing feature. The
3336 new name refers directly to the setclassresources() function.
3338 PH/65 Added configuration files for NetBSD3.
3340 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3342 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3343 is preferred over IPv4.
3345 PH/68 The bounce_return_message and bounce_return_body options were not being
3346 honoured for bounces generated during the reception of non-SMTP messages.
3347 In particular, this applied to messages rejected by the ACL. This bug has
3348 been fixed. However, if bounce_return_message is true and bounce_return_
3349 body is false, the headers that are returned for a non-SMTP message
3350 include only those that have been read before the error was detected.
3351 (In the case of an ACL rejection, they have all been read.)
3353 PH/69 The HTML version of the specification is now built in a directory called
3354 spec_html instead of spec.html, because the latter looks like a path with
3355 a MIME-type, and this confuses some software.
3357 PH/70 Catch two compiler warnings in sieve.c.
3359 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3360 function verify_get_ident() calls ip_connect() to connect a socket, but
3361 if the "connect()" function timed out, ip_connect() used to close the
3362 socket. However, verify_get_ident() also closes the socket later, and in
3363 between Exim writes to the log, which may get opened at this point. When
3364 the socket was closed in ip_connect(), the log could get the same file
3365 descriptor number as the socket. This naturally causes chaos. The fix is
3366 not to close the socket in ip_connect(); the socket should be closed by
3367 the function that creates it. There was only one place in the code where
3368 this was missing, in the iplookup router, which I don't think anybody now
3369 uses, but I've fixed it anyway.
3371 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3372 well as to direct DNS lookups. Otherwise the handling of names in host
3373 lists is inconsistent and therefore confusing.
3379 PH/01 Two changes to the default runtime configuration:
3381 (1) Move the checks for relay_from_hosts and authenticated clients from
3382 after to before the (commented out) DNS black list checks.
3384 (2) Add control=submission to the relay_from_hosts and authenticated
3385 clients checks, on the grounds that messages accepted by these
3386 statements are most likely to be submissions.
3388 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3390 (1) Generate an error if the third argument for the ${prvs expansion is
3393 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3396 (3) Reset the variables that are obtained from the first argument of
3397 ${prvscheck and used in the second argument before leaving the code,
3398 because their memory is reclaimed, so using them afterwards may do
3401 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3402 one (it's much easier than Tom thought :-).
3404 (5) Because of (4), we can now allow for the use of $prvscheck_result
3405 inside the third argument.
3407 PH/03 For some reason, the default setting of PATH when running a command from
3408 a pipe transport was just "/usr/bin". I have changed it to
3411 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3412 anything to be listed in the output from -bV.
3414 PH/05 When a filter generated an autoreply, the entire To: header line was
3415 quoted in the delivery log line, like this:
3417 => >A.N.Other <ano@some.domain> <original@ddress> ...
3419 This has been changed so that it extracts the operative address. There
3420 may be more than one such address. If so, they are comma-separated, like
3423 => >ano@some.domain,ona@other.domain <original@ddress> ...
3425 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3426 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3427 not being looked up in the rDNS to get a host name, Exim was showing the
3428 IP address twice in Received: lines, even though the IP addresses were
3429 identical. For example:
3431 Received: from [1.2.3.4] (helo=[1.2.3.4])
3433 However, if the real host name was known, it was omitting the HELO data
3434 if it matched the actual IP address. This has been tidied up so that it
3435 doesn't show the same IP address twice.
3437 PH/07 When both +timestamp and +memory debugging was on, the value given by
3438 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3439 called by the debug printing, thereby overwriting the timestamp buffer.
3440 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3442 PH/08 When the original message was included in an autoreply transport, it
3443 always said "this is a copy of the message, including all the headers",
3444 even if body_only or headers_only was set. It now gives an appropriate
3447 PH/09 Applied a patch from the Sieve maintainer which:
3449 o fixes some comments
3450 o adds the (disabled) notify extension core
3451 o adds some debug output for the result of if/elsif tests
3452 o points to the current vacation draft in the documentation
3453 and documents the missing references header update
3457 o fixes a bug in processing the envelope test (when testing
3458 multiple envelope elements, the last element determined the
3461 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3462 Electronic Mail") by including:
3464 Auto-submitted: auto-generated
3466 in the messages that it generates (bounce messages and others, such as
3467 warnings). In the case of bounce messages for non-SMTP messages, there was
3468 also a typo: it was using "Auto_submitted" (underscore instead of
3469 hyphen). Since every message generated by Exim is necessarily in response
3470 to another message, thes have all been changed to:
3472 Auto-Submitted: auto-replied
3474 in accordance with these statements in the RFC:
3476 The auto-replied keyword:
3478 - SHOULD be used on messages sent in direct response to another
3479 message by an automatic process,
3481 - MUST NOT be used on manually-generated messages,
3483 - MAY be used on Delivery Status Notifications (DSNs) and Message
3484 Disposition Notifications (MDNs),
3486 - MUST NOT be used on messages generated by automatic or periodic
3487 processes, except for messages which are automatic responses to
3490 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3491 to the default Received: header definition.
3493 PH/12 Added log selector acl_warn_skipped (default on).
3495 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3496 variables because (a) they are in the wrong storage pool and (b) even if
3497 they were copied, it wouldn't work properly because of the caching.
3499 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3500 checking when decoding. Apparently there are clients that generate
3501 overlong encoded strings. Why am I not surprised?
3503 PH/15 If the first argument of "${if match_address" was not empty, but did not
3504 contain an "@" character, Exim crashed. Now it writes a panic log message
3505 and treats the condition as false.
3507 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3509 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3510 extension "envelope-auth". The code is finished and in agreement with
3511 other implementations, but there is no documentation so far and in fact,
3512 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3513 not changing the active code.
3515 Print executed "if" and "elsif" statements when debugging is used. This
3516 helps a great deal to understand what a filter does.
3518 Document more things not specified clearly in RFC3028. I had all this
3519 sorted out, when out of a sudden new issues came to my mind. Oops."
3521 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3524 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3525 canonical form (as documented). However, after a host name lookup from
3526 the IP address, check_host() was doing a simple string comparison with
3527 addresses acquired from the DNS when checking that the found name did
3528 have the original IP as one of its addresses. Since any found IPv6
3529 addresses are likely to be in abbreviated form, the comparison could
3530 fail. Luckily, there already exists a function for doing the comparison
3531 by converting both addresses to binary, so now that is used instead of
3532 the text comparison.
3534 PH/20 There was another similar case to PH/19, when a complete host name was
3535 given in a host list; looking up its IP address could give an abbreviated
3536 form, whereas the current host's name might or might not be abbreviated.
3537 The same fix has been applied.
3543 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3544 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3547 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3548 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3550 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3552 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3553 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3554 Also, for "Auto-Submitted", treat anything other than "no" as
3555 non-personal, in accordance with RFC 3834. (Previously it treated
3556 anything starting "auto-" as non-personal.)
3558 TF/01 The control=submission/name=... option had a problem with syntax
3559 errors if the name included a slash character. The /name= option
3560 now slurps the rest of the string, so it can include any characters
3561 but it must come last in the list of options (after /sender_retain
3564 PH/05 Some modifications to the interface to the fake nameserver for the new
3572 TK/01 Added the "success_on_redirect" address verification option. See
3573 NewStuff for rationale and an example.
3575 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3577 PH/02 Patch to exigrep to allow it to work on syslog lines.
3579 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3580 fread() to skip over the body file's header line, because in Cygwin the
3581 header line is locked and is inaccessible.
3583 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3584 co-exist for some time) to make it clear that it is the Exim ID that is
3585 referenced, not the Message-ID: header line.
3587 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3588 string_format() function, because snprintf() does not exist on all
3591 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3592 ${stat: expansion item.
3594 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3595 protocol synchronization error", to keep the pedants happy.
3597 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3598 well as for IRIX systems, when gcc is being used. See the host.c source
3601 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3603 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3606 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3607 transport and to the smtp transport in LMTP mode.
3609 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3611 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3612 run a filter in a subprocess. This could lead to confusion in subsequent
3613 lookups in the parent process. There should also be a search_tidyup() at
3614 the end of the subprocess.
3616 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3617 only if the host matched helo_try_verify_hosts, which caused the
3618 verification to occur when the EHLO/HELO command was issued. The ACL just
3619 tested the remembered result. Now, if a previous verification attempt has
3620 not happened, "verify = helo" does it there and then.
3622 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3624 TK/03 Fix log output including CR from clamd.
3626 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3627 exist provoked a memory error which could cause a segfault.
3629 PH/15 Installed PCRE 6.2
3631 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3633 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3634 of the problem. Specifically, suggested +O2 rather than +O1 for the
3637 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3639 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3640 record, the port was not correctly logged when the outgoing_port log
3641 selector was set (it logged the transort's default port).
3643 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3644 fallback_hosts, and "hosts" in the smtp transport.
3646 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3647 host errors such as "Connection refused".
3649 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3650 authentication with radiusclient 0.4.9:
3652 - Error returned from rc_read_config was caught wrongly
3653 - Username/password not passed on to radius server due to wrong length.
3655 The presumption is that some radiusclient API changes for 4.51/PH/17
3656 were not taken care of correctly. The code is still untested by me (my
3657 Linux distribution still has 0.3.2 of radiusclient), but it was
3658 contributed by a Radius user.
3660 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3661 expanding the "port" option of the smtp transport.
3663 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3664 while reading a MIME header. Thanks to Tom Hughes for a patch.
3666 PH/24 Include config.h inside local_scan.h so that configuration settings are
3669 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3670 that in an ACL for RCPT (for example), you can examine exactly what was
3673 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3674 commands, but it was not correctly comparing the address with the actual
3675 client host address. Thus, it would show the EHLO address in Received:
3676 header lines when this was not necessary.
3678 PH/27 Added the % operator to ${eval:}.
3680 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3681 it should be ignoring failures (because with -C, for example, it has lost
3682 privilege). It wasn't ignoring creation failures other than "already
3685 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3688 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3689 because an input line was too long, either on its own, or by virtue of
3690 too many continuations, the temporary file was not being removed, and the
3691 return code was incorrect.
3693 PH/31 Missing "BOOL" in function definition in filtertest.c.
3695 PH/32 Applied Sieve patches from the maintainer.
3697 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3699 PH/33 Added "verify = not_blind".
3701 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3702 Local/Makefile (with some defaults set). These are used in built scripts
3703 such as exicyclog, but they have never been used in the exim_install
3704 script (though there are many overriding facilities there). I have
3705 arranged that the exim_install script now takes note of these two
3708 PH/35 Installed configuration files for Dragonfly.
3710 PH/36 When a locally submitted message by a trusted user did not contain a
3711 From: header, and the sender address was obtained from -f or from an SMTP
3712 MAIL command, and the trusted user did not use -F to supply a sender
3713 name, $originator_name was incorrectly used when constructing a From:
3714 header. Furthermore, $originator_name was used for submission mode
3715 messages from external hosts without From: headers in a similar way,
3716 which is clearly wrong.
3718 PH/37 Added control=suppress_local_fixups.
3720 PH/38 When log_selector = +received_sender was set, and the addition of the
3721 sender made the log line's construction buffer exactly full, or one byte
3722 less than full, an overflow happened when the terminating "\n" was
3725 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3726 when the result of a list match is failure because a DNS lookup failed.
3728 PH/40 RM_COMMAND is now used in the building process.
3730 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3731 the "build-* directories that it finds.
3733 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3734 address in a domain literal was a prefix of an interface address.
3736 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3737 when verifying a sender address, unless rewrite_headers is false.
3739 PH/44 Wrote a long comment about why errors_to addresses are verified as
3740 recipients, not senders.
3742 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3743 the ratelimit ACL was added.
3745 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3747 PH/46 Added extra information about PostgreSQL errors to the error string.
3749 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3750 avoiding the need to install special zones in a real server. This is
3751 backwards compatible; if it can't find the fake resolver, it drops back.
3752 Thus, both old and new test suites can be run.
3754 TF/02 Added util/ratelimit.pl
3756 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3757 clock is set back in time.
3759 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3760 Candler <B.Candler@pobox.com>.
3762 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3763 OK for addresses that are the result of redirections.
3765 PH/48 A number of further additions for the benefit of the new test suite,
3766 including a fake gethostbyname() that interfaces to the fake DNS resolver
3769 TF/06 The fix for widen_domains has also been applied to qualify_single and
3770 search_parents which are the other dnslookup options that can cause
3773 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3776 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3778 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3779 service" if it encountered an MX record with an empty target hostname.
3780 The message is now "an MX or SRV record indicated no SMTP service".
3782 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3783 if the DNS of the sending site is misconfigured. This is quite a
3784 common situation. This change restores the behaviour of treating a
3785 helo verification defer as a failure.
3787 PH/51 If self=fail was set on a router, the bounce message did not include the
3788 actual error message.
3794 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3796 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3797 command itself ended in error, the underlying message about the transport
3798 filter timeout was being overwritten with the pipe command error. Now the
3799 underlying error message should be appended to the second error message.
3801 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3803 PH/02 Reduce the amount of output that "make" produces by default. Full output
3804 can still be requested.
3806 PH/03 The warning log line about a condition test deferring for a "warn" verb
3807 was being output only once per connection, rather than after each
3808 occurrence (because it was using the same function as for successful
3809 "warn" verbs). This seems wrong, so I have changed it.
3811 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3812 it should not have, which might have caused a crash in the right
3813 circumstances, but probably never did.
3815 PH/04 Installed a modified version of Tony Finch's patch to make submission
3816 mode fix the return path as well as the Sender: header line, and to
3817 add a /name= option so that you can make the user's friendly name appear
3820 TF/03 Added the control = fakedefer ACL modifier.
3822 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3823 Mark Lowes for thorough testing.
3825 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3827 TK/03 Merged latest SRS patch from Miles Wilton.
3829 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3830 with the definition in sysexits.h (which is #included earlier).
3831 Fortunately, Exim does not actually use EX_OK. The code used to try to
3832 preserve the sysexits.h value, by assuming that macro definitions were
3833 scanned for macro replacements. I have been disabused of this notion,
3834 so now the code just undefines EX_OK before #including unistd.h.
3836 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3837 in the smtp transport. When a block could not be written in a single
3838 write() function, the timeout was being re-applied to each part-write.
3839 This seems wrong - if the receiver was accepting one byte at a time it
3840 would take for ever. The timeout is now adjusted when this happens. It
3841 doesn't have to be particularly precise.
3843 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3844 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3846 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3847 without a domain if the check to <postmaster@domain> fails.
3849 SC/01 Eximstats: added -xls and the ability to specify output files
3850 (patch written by Frank Heydlauf).
3852 SC/02 Eximstats: use FileHandles for outputting results.
3854 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3856 SC/04 Eximstats: fixed display of large numbers with -nvr option
3858 SC/05 Eximstats: fixed merging of reports with empty tables.
3860 SC/06 Eximstats: added the -include_original_destination flag
3862 SC/07 Eximstats: removed tabs and trailing whitespace.
3864 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3866 TK/06 MBOX spool code: Add real "From " MBOX separator line
3867 so the .eml file is really in mbox format (even though
3868 most programs do not really care). Patch from Alex Miller.
3870 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3871 The latter is generated from $received_to and is only set if the
3872 message has one envelope recipient. SA can use these headers,
3873 obviously out-of-the-box. Patch from Alex Miller.
3875 PH/08 The ${def test on a variable was returning false if the variable's
3876 value was "0", contrary to what the specification has always said!
3877 The result should be true unless the variable is empty.
3879 PH/09 The syntax error of a character other than { following "${if
3880 def:variable_name" (after optional whitespace) was not being diagnosed.
3881 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3882 accidental colon was present, for example, could give incorrect results.
3884 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3885 result is used (not including appendfile, where other changes are about
3888 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3889 This involved changing a lot of size variables from int to off_t. It
3890 should work with maildirs and everything.
3892 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3893 spamd dying while we are connected to it.
3895 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3898 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3899 function for BDB 4.3.
3901 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3903 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3904 the sources for PCRE have been re-arranged and more files are now
3907 PH/15 The code I had for printing potentially long long variables in PH/11
3908 above was not the best (it lost precision). The length of off_t variables
3909 is now inspected at build time, and an appropriate printing format (%ld
3910 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3911 to be "long long int" or "long int". This is needed for the internal
3912 formatting function string_vformat().
3914 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3915 the configuration file to be ":syslog", then the script "guesses" where
3916 the logs files are, rather than using the compiled in default. In our
3917 case the guess is not the same as the compiled default, so the script
3918 suddenly stopped working when I started to use syslog. The patch checks
3919 to see if log_file_path is "". If so, it attempts to read it from exim
3920 with no configuration file to get the compiled in version, before it
3921 falls back to the previous guessing code."
3923 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3924 implementing BATV in an Exim configuration. See NewStuff for the gory
3927 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3928 Makefile that are specific to HP-UX.
3930 PH/18 If the "use_postmaster" option was set for a recipient callout together
3931 with the "random" option, the postmaster address was used as the MAIL
3932 FROM address for the random test, but not for the subsequent recipient
3933 test. It is now used for both.
3935 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3936 patch removes a few documentation additions to RFC 3028, because the
3937 latest draft now contains them. It adds the new en;ascii-case comparator
3938 and a new error check for 8bit text in MIME parts. Comparator and
3939 require names are now matched exactly. I enabled the subaddress
3940 extension, but it is not well tested yet (read: it works for me)."
3942 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3943 rework some of the code of TK/09 above to avoid the hardwired use of
3944 "%lld" and "long long". Replaced the call to snprintf() with a call to
3947 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3948 records point to non-existent hosts", "retry timeout exceeded", and
3949 "retry time not reached for any host after a long failure period".
3951 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3952 experimental DomainKeys support:
3954 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3955 (2) On an error such as an illegally used "control", the wrong name for
3956 the control was given.
3958 These problems did NOT occur unless DomainKeys support was compiled.
3960 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3962 PH/24 Added ${if match_ip condition.
3964 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3965 databases so that it will be absolutely obvious if a crash occurs in the
3966 DB library. This is a regular occurrence (often caused by mis-matched
3969 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3970 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3971 picked up on a user's system that detects such things. There doesn't seem
3972 to be a gcc warning option for this - only an attribute that has to be
3973 put on the function's prototype. It seems that in Fedora Core 4 they have
3974 set this on a number of new functions. No doubt there will be more in due
3977 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3978 specify a transport. However, if an address that was verified by such a
3979 router was the subject of a callout, Exim crashed because it tried to
3980 read the rcpt_include_affixes from the non-existent transport. Now it
3981 just assumes that the setting of that option is false. This bug was
3982 introduced by 4.51/PH/31.
3984 PH/28 Changed -d+all to exclude +memory, because that information is very
3985 rarely of interest, but it makes the output a lot bigger. People tend to
3986 do -d+all out of habit.
3988 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3989 code in os-type was giving problems when libc.so lives in lib64, like on
3992 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3993 aren't the modern standard, and it seems that some systems' include files
3994 don't always have them. Exim was already checking for some of the newer
3995 ones like T_AAAA, and defining it itself. I've added checks for all the
3996 record types that Exim uses.
3998 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3999 not automatically generating a new one, as it is supposed to. This
4000 prevented TLS from working. If the file did exist, but contained invalid
4001 data, a new version was generated, as expected. It was only the case of a
4002 non-existent file that was broken.
4004 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4005 with a change in libdomainkeys > 0.64.
4007 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4008 from DNS. If the selector record carries the flag, it now has
4009 precedence over the domain-wide flag.
4011 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4013 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4014 the use of a port name that isn't defined in /etc/services) occurred, the
4015 message was deferred as in a normal delivery, and thus remained on the
4016 spool, instead of being failed because of the mua_wrapper setting. This
4017 is now fixed, and I tidied up some of the mua_wrapper messages at the
4020 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4021 the messages in a hash of arrays rather than using individual hashes.
4022 This is a bit cleaner and results in dramatic memory savings, albeit
4023 at a slight CPU cost.
4025 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4026 as requested by Marc Sherman.
4028 SC/10 Eximstats: added histograms for user specified patterns as requested
4031 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4033 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4034 fopen() in the content-scanning modules that did not already have it.
4040 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4041 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4043 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4045 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4047 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4048 to test Sieve filters that use "vacation".
4050 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4051 that changes the way the GnuTLS parameters are stored in the cache file.
4052 The new format can be generated externally. For backward compatibility,
4053 if the data in the cache doesn't make sense, Exim assumes it has read an
4054 old-format file, and it generates new data and writes a new file. This
4055 means that you can't go back to an older release without removing the
4058 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4059 work if there are any delivery delays because "one_time" forces the
4060 parent to be marked "delivered", so its unseen clone is never tried
4061 again. For this reason, Exim now forbids the simultaneous setting of
4064 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4065 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4066 entry if you want to know the details. The fix, however, neglected to
4067 consider the case when local delivery batching is involved. The test for
4068 "previously delivered" was not happening when checking to see if an
4069 address could be batched with a previous (undelivered) one; under
4070 certain circumstances this could lead to multiple deliveries to the same
4073 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4074 in its include files, and this causes problems building Exim.
4076 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4077 header_syntax) but Exim was just ignoring anything given after a slash.
4078 In particular, this caused confusion with an attempt to use "verify =
4079 reverse_host_lookup/defer_ok". An error is now given when options are
4080 supplied for verify items that do not have them. (Maybe reverse_host_
4081 lookup should have a defer_ok option, but that's a different point.)
4083 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4084 defined by RFC 821) to 2048, because there were problems with some AUTH
4085 commands, and RFC 1869 says the size should be increased for extended
4086 SMTP commands that take arguments.
4088 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4091 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4092 "unknown" error; now it says that the functionality isn't in the binary.
4094 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4095 an address' error message when a string expansion fails (syntax or
4096 whatever). Otherwise the password may appear in the log. Following change
4097 PH/42 below, there is no longer a chance of it appearing in a bounce
4100 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4102 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4103 of its addresses was ever tried. (Bugzilla bug #2.)
4105 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4106 the result incorrectly in the debug output. (It correctly added a newline
4107 to what was transported.)
4109 TF/01 Added $received_time.
4111 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4112 commented out examples of how to interface to a virus scanner and to
4113 SpamAssassin. Also added commented examples of av_scanner and
4114 spamd_address settings.
4116 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4117 and controls are allowed in which ACLs. There were a couple of minor
4118 errors. Some of the entries in the conditions table (which is a table of
4119 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4120 negation of where the condition IS allowed.
4122 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4124 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4125 header file does not have a version number, so I've had to invent a new
4126 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4127 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4128 radiusclient), but it was contributed by a Radius user.
4130 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4131 files or directories, for OpenSSL.
4133 PH/19 When an Exim process that is running as root has to create an Exim log
4134 file, it does so in a subprocess that runs as exim:exim so as to get the
4135 ownership right at creation (otherwise, other Exim processes might see
4136 the file with the wrong ownership). There was no test for failure of this
4137 fork() call, which would lead to the process getting stuck as it waited
4138 for a non-existent subprocess. Forks do occasionally fail when resources
4139 run out. I reviewed all the other calls to fork(); they all seem to check
4142 PH/20 When checking for unexpected SMTP input at connect time (before writing
4143 the banner), Exim was not dealing correctly with a non-positive return
4144 from the read() function. If the client had disconnected by this time,
4145 the result was a log entry for a synchronization error with an empty
4146 string after "input=" when read() returned zero. If read() returned -1
4147 (an event I could not check), uninitialized data bytes were printed.
4148 There were reports of junk text (parts of files, etc) appearing after
4151 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4153 PH/22 Added support for macro redefinition, and (re)definition in between
4154 driver and ACL definitions.
4156 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4157 forgetting to use the resulting value; it was using the unexpanded value.
4159 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4160 hadn't been configured. The fix is from Juergen Kreileder, who
4161 understands it better than I do:
4163 "Here's what I see happening with three configured cyrus_sasl
4164 authenticators configured (plain, login, cram-md5):
4166 On startup auth_cyrus_sasl_init() gets called for each of these.
4167 This means three calls to sasl_listmech() without a specified mech_list.
4168 => SASL tests which mechs of all available mechs actually work
4169 => three warnings about OTP not working
4170 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4172 With the patch, sasl_listmech() also gets called three times. But now
4173 SASL's mech_list option is set to the server_mech specified in the the
4174 authenticator. Or in other words, the answer from sasl_listmech()
4175 gets limited to just the mech you're testing for (which is different
4177 => the return list contains just 'plain' or 'login', 'cram-md5' or
4178 nothing depending on the value of ob->server_mech.
4180 I've just tested the patch: Authentication still works fine,
4181 unavailable mechs specified in the exim configuration are still
4182 caught, and the auth.log warnings about OTP are gone."
4184 PH/25 When debugging is enabled, the contents of the command line are added
4185 to the debugging output, even when log_selector=+arguments is not
4188 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4189 answer is "GNU", and only if the return is "GNU/something" is the answer
4192 PH/27 $acl_verify_message is now set immediately after the failure of a
4193 verification in an ACL, and so is available in subsequent modifiers. In
4194 particular, the message can be preserved by coding like this:
4196 warn !verify = sender
4197 set acl_m0 = $acl_verify_message
4199 Previously, $acl_verify_message was set only while expanding "message"
4200 and "log_message" when a very denied access.
4202 PH/28 Modified OS/os.c-Linux with
4204 -#ifndef OS_LOAD_AVERAGE
4205 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4207 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4208 nomenclature these days.)
4210 PH/29 Installed patch from the Sieve maintainer that adds the options
4211 sieve_useraddress and sieve_subaddress to the redirect router.
4213 PH/30 In these circumstances:
4214 . Two addresses routed to the same list of hosts;
4215 . First host does not offer TLS;
4216 . First host accepts first address;
4217 . First host gives temporary error to second address;
4218 . Second host offers TLS and a TLS session is established;
4219 . Second host accepts second address.
4220 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4221 and peerdn, if requested) that were in fact used only for the second
4224 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4225 attention to any local part prefix or suffix that was matched by the
4226 router that accepted the address. It now behaves in the same way as it
4227 does for delivery: the affixes are removed from the local part unless
4228 rcpt_include_affixes is set on the transport.
4230 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4231 timeout during the DATA phase of an incoming message.
4233 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4234 applied a patch sanctioned by the Sieve maintainer.
4236 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4237 the uid or gid is negative. A case of a negative gid caused this to be
4238 noticed. The fix allows for either to be negative.
4240 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4241 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4244 PH/36 The change PH/12 above was broken. Fixed it.
4246 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4247 the grounds that routing the same address twice would always produce the
4248 same answer. This might have been true once, but it is certainly no
4249 longer true now. Routing a child address may depend on the previous
4250 routing that produced that child. Some complicated redirection strategies
4251 went wrong when messages had multiple recipients, and made Exim's
4252 behaviour dependent on the order in which the addresses were given.
4254 I have moved the duplicate checking until after the routing is complete.
4255 Exim scans the addresses that are assigned to local and remote
4256 transports, and removes any duplicates. This means that more work will be
4257 done, as duplicates will always all be routed, but duplicates are
4258 presumably rare, so I don't expect this is of any significance.
4260 For deliveries to pipes, files, and autoreplies, the duplicate checking
4261 still happens during the routing process, since they are not going to be
4264 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4265 It corrects a timeout issue with spamd. This is Ian's comment: "The
4266 background is that sometimes spamd either never reads data from a
4267 connection it has accepted, or it never writes response data. The exiscan
4268 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4269 blindly assumes that writes won't block so it may never time out."
4271 PH/39 Allow G after quota size as well as K and M.
4273 PH/40 The value set for $authenticated_id in an authenticator may not contain
4274 binary zeroes or newlines because the value is written to log lines and
4275 to spool files. There was no check on this. Now the value is run through
4276 the string_printing() function so that such characters are converted to
4277 printable escape sequences.
4279 PH/41 $message_linecount is a new variable that contains the total number of
4280 lines in the message. Compare $body_linecount, which is the count for the
4283 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4284 bounce and delay warning messages, except in certain special cases, which
4287 (a) An SMTP error message from a remote host;
4288 (b) A message specified in a :fail: redirection;
4289 (c) A message specified in a "fail" command in a system filter;
4290 (d) A message specified in a FAIL return from the queryprogram router;
4291 (e) A message specified by the cannot_route_message router option.
4293 In these cases only, Exim does include the error details in bounce and
4294 warning messages. There are also a few cases where bland messages such
4295 as "unrouteable address" or "local delivery error" are given.
4297 PH/43 $value is now also set for the "else" part of a ${run expansion.
4299 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4300 being worked on, but at least Exim now implements the latest version to
4303 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4304 process to complete was treated as a delivery failure, a timeout while
4305 writing the message to the pipe was logged, but erroneously treated as a
4306 successful delivery. Such timeouts include transport filter timeouts. For
4307 consistency with the overall process timeout, these timeouts are now
4308 treated as errors, giving rise to delivery failures by default. However,
4309 there is now a new Boolean option for the pipe transport called
4310 timeout_defer, which, if set TRUE, converts the failures into defers for
4311 both kinds of timeout. A transport filter timeout is now identified in
4314 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4315 systems where "make" and "gmake" are different, calling "gmake" at top
4316 level broke things. I've arranged for the value of $(MAKE) to be passed
4317 from the Makefile to this script so that it can call the same version of
4321 A note about Exim versions 4.44 and 4.50
4322 ----------------------------------------
4324 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4325 changes of various kinds. As a consequence, a big documentation update was
4326 needed. This delayed the release for rather longer than seemed good, especially
4327 in the light of a couple of (minor) security issues. Therefore, the changes
4328 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4329 release. So 4.44 and 4.50 are in effect two different branches that both start
4332 I have left the 4.50 change log unchanged; it contains all the changes since
4333 4.43. The change log for 4.44 is below; many of its items are identical to
4334 those for 4.50. This seems to be the most sensible way to preserve the
4335 historical information.
4341 1. Minor wording change to the doc/README.SIEVE file.
4343 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4344 computation of the current number of files was incorrect.
4346 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4347 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4350 4. Give more explanation in the error message when the command for a transport
4351 filter fails to execute.
4353 5. There are several places where Exim runs a non-Exim command in a
4354 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4355 was being done only for the command run by the queryprogram router. It is
4356 now done for all such subprocesses. The other cases are: ${run, transport
4357 filters, and the commands run by the lmtp and pipe transports.
4359 6. Added CONFIGURE_GROUP build-time option.
4361 7. Some older OS have a limit of 256 on the maximum number of file
4362 descriptors. Exim was using setrlimit() to set 1000 as a large value
4363 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4364 systems. I've change it so that if it can't get 1000, it tries for 256.
4366 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4367 was an oversight, and furthermore, ever since the addition of extra
4368 controls (e.g. 4.43/32), the checks on when to allow different forms of
4369 "control" were broken. There should now be diagnostics for all cases when a
4370 control that does not make sense is encountered.
4372 9. Added the /retain_sender option to "control=submission".
4374 10. $recipients is now available in the predata ACL (oversight).
4376 11. Tidy the search cache before the fork to do a delivery from a message
4377 received from the command line. Otherwise the child will trigger a lookup
4378 failure and thereby defer the delivery if it tries to use (for example) a
4379 cached ldap connection that the parent has called unbind on.
4381 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4382 of $address_data from the recipient verification was clobbered by the
4383 sender verification.
4385 13. The value of address_data from a sender verification is now available in
4386 $sender_address_data in subsequent conditions in the ACL statement.
4388 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4390 15. Added a new option "connect=<time>" to callout options, to set a different
4393 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4394 was its contents. (It was OK if the option was not defined at all.)
4396 17. A "Completed" log line is now written for messages that are removed from
4397 the spool by the -Mrm option.
4399 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4400 information about exactly what failed.
4402 19. Added -dd to debug only the daemon process.
4404 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4405 handles timeouts, both on the server side and network timeouts. Renamed the
4406 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4408 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4409 It is now set to "smtps".
4411 22. $host_address is now set to the target address during the checking of
4412 ignore_target_hosts.
4414 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4415 being passed; this would have caused $sender_host_name to have been used if
4416 matching the list had actually called for a host name (not very likely,
4417 since this list is usually IP addresses). A host name is now passed as
4420 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4421 code that specifies a non-restarting handler (typically sigaction() in
4422 modern systems) in an attempt to fix a rare and obscure crash bug.
4424 25. Narrowed the window for a race in the daemon that could cause it to ignore
4425 SIGCHLD signals. This is not a major problem, because they are used only to
4426 wake it up if nothing else does.
4428 26. A malformed maildirsize file could cause Exim to calculate negative values
4429 for the mailbox size or file count. Odd effects could occur as a result.
4430 The maildirsize information is now recalculated if the size or filecount
4433 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4434 support for a long time. Removed HAVE_SYS_VFS_H.
4436 28. Installed the latest version of exipick from John Jetmore.
4438 29. In an address list, if the pattern was not a regular expression, an empty
4439 subject address (from a bounce message) matched only if the pattern was an
4440 empty string. Non-empty patterns were not even tested. This was the wrong
4441 because it is perfectly reasonable to use an empty address as part of a
4442 database query. An empty address is now tested by patterns that are
4443 lookups. However, all the other forms of pattern expect the subject to
4444 contain a local part and a domain, and therefore, for them, an empty
4445 address still always fails if the pattern is not itself empty.
4447 30. Exim went into a mad DNS loop when attempting to do a callout where the
4448 host was specified on an smtp transport, and looking it up yielded more
4449 than one IP address.
4451 31. Re-factored the code for checking spool and log partition space into a
4452 function that finds that data and another that does the check. The former
4453 is then used to implement four new variables: $spool_space, $log_space,
4454 $spool_inodes, and $log_inodes.
4456 32. The RFC2047 encoding function was originally intended for short strings
4457 such as real names; it was not keeping to the 75-character limit for
4458 encoded words that the RFC imposes. It now respects the limit, and
4459 generates multiple encoded words if necessary. To be on the safe side, I
4460 have increased the buffer size for the ${rfc2047: expansion operator from
4463 33. It is now permitted to omit both strings after an "if" condition; if the
4464 condition is true, the result is "true". As before, when the second string
4465 is omitted, a false condition yields an empty string. This makes it less
4466 cumbersome to write custom ACL and router conditions.
4468 34. Failure to deliver a bounce message always caused it to be frozen, even if
4469 there was an errors_to setting on the router. The errors_to setting is now
4472 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4473 canonical form (fully expanded) before being placed in
4474 $sender_host_address.
4476 36. The table in the code that translates DNS record types into text (T_A to
4477 "A" for instance) was missing entries for NS and CNAME. It is just possible
4478 that this could have caused confusion if both these types were looked up
4479 for the same domain, because the text type is used as part of Exim's
4480 per-process caching. But the chance of anyone hitting this buglet seems
4483 37. The dnsdb lookup has been extended in a number of ways.
4485 (1) There is a new type, "zns", which walks up the domain tree until it
4486 finds some nameserver records. It should be used with care.
4488 (2) There is a new type, "mxh", which is like "mx" except that it returns
4489 just the host names, not the priorities.
4491 (3) It is now possible to give a list of domains (or IP addresses) to be
4492 looked up. The behaviour when one of the lookups defers can be
4493 controlled by a keyword.
4495 (4) It is now possible to specify the separator character for use when
4496 multiple records are returned.
4498 38. The dnslists ACL condition has been extended: it is now possible to supply
4499 a list of IP addresses and/or domains to be looked up in a particular DNS
4502 39. Added log_selector=+queue_time_overall.
4504 40. When running the queue in the test harness, wait just a tad after forking a
4505 delivery process, to get repeatability of debugging output.
4507 41. Include certificate and key file names in error message when GnuTLS fails
4508 to set them up, because the GnuTLS error message doesn't include the name
4509 of the failing file when there is a problem reading it.
4511 42. Allow both -bf and -bF in the same test run.
4513 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4515 44. The "Exiscan patch" is now merged into the mainline Exim source.
4517 45. Sometimes the final signoff response after QUIT could fail to get
4518 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4519 < 0 before doing a fflush(). This bug looks as though it goes back to the
4520 introduction of TLS in release 3.20, but "sometimes" must have been rare
4521 because the tests only now provoked it.
4523 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4524 (this can affect the format of dates).
4526 47. exim_tidydb, when checking for the continued existence of a message for
4527 which it has found a message-specific retry record, was not finding
4528 messages that were in split spool directories. Consequently, it was
4529 deleting retry records that should have stayed in existence.
4531 48. Steve fixed some bugs in eximstats.
4533 49. The SPA authentication driver was not abandoning authentication and moving
4534 on to the next authenticator when an expansion was forced to fail,
4535 contradicting the general specification for all authenticators. Instead it
4536 was generating a temporary error. It now behaves as specified.
4538 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4539 (the order specifies the preference for clients). The order is now AES256,
4540 AES128, 3DES, ARCFOUR128.
4542 51. Small patch to Sieve code - explicitly set From: when generating an
4545 52. Exim crashed if a remote delivery caused a very long error message to be
4546 recorded - for instance if somebody sent an entire SpamAssassin report back
4547 as a large number of 550 error lines. This bug was coincidentally fixed by
4548 increasing the size of one of Exim's internal buffers (big_buffer) that
4549 happened as part of the Exiscan merge. However, to be on the safe side, I
4550 have made the code more robust (and fixed the comments that describe what
4553 53. Now that there can be additional text after "Completed" in log lines (if
4554 the queue_time_overall log selector is set), a one-byte patch to exigrep
4555 was needed to allow it to recognize "Completed" as not the last thing in
4558 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4559 patch that reportedly fixes this has been added. I am not expert enough to
4560 create a test for it. This is what the patch creator wrote:
4562 "I found a little strange behaviour of ldap code when working with
4563 Windows 2003 AD Domain, where users was placed in more than one
4564 Organization Units. When I tried to give exim partial DN, the exit code
4565 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4566 But simultaneously result of request was absolutely normal ldap result,
4567 so I produce this patch..."
4569 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4570 REFERENCE, so I have modified the code to exclude the patch when that macro
4573 55. Some experimental protocols are using DNS PTR records for new purposes. The
4574 keys for these records are domain names, not reversed IP addresses. The
4575 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4576 leaves it alone. Component reversal etc. now happens only for IP addresses.
4579 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4581 57. Double the size of the debug message buffer (to 2048) so that more of very
4582 long debug lines gets shown.
4584 58. The exicyclog utility now does better if the number of log files to keep
4585 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4587 59. Two changes related to the smtp_active_hostname option:
4589 (1) $smtp_active_hostname is now available as a variable.
4590 (2) The default for smtp_banner uses $smtp_active_hostname instead
4591 of $primary_hostname.
4593 60. The host_aton() function is supposed to be passed a string that is known
4594 to be a valid IP address. However, in the case of IPv6 addresses, it was
4595 not checking this. This is a hostage to fortune. Exim now panics and dies
4596 if the condition is not met. A case was found where this could be provoked
4597 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4598 components; fortuitously, this particular loophole had already been fixed
4599 by change 4.50/55 above.
4601 If there are any other similar loopholes, the new check in host_aton()
4602 itself should stop them being exploited. The report I received stated that
4603 data on the command line could provoke the exploit when Exim was running as
4604 exim, but did not say which command line option was involved. All I could
4605 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4606 running as the user.
4609 61. There was a buffer overflow vulnerability in the SPA authentication code
4610 (which came originally from the Samba project). I have added a test to the
4611 spa_base64_to_bits() function which I hope fixes it.
4614 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4615 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4617 63. The daemon start-up calls getloadavg() while still root for those OS that
4618 need the first call to be done as root, but it missed one case: when
4619 deliver_queue_load_max is set with deliver_drop_privilege. This is
4620 necessary for the benefit of the queue runner, because there is no re-exec
4621 when deliver_drop_privilege is set.
4623 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4624 This has been fixed.
4626 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4627 was in use, was not putting the data itself into the right store pool;
4628 consequently, it could be overwritten for a subsequent message in the same
4629 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4632 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4634 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4635 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4636 the function were treating the return as a boolean value, which happened to
4637 work because 0=false and not-0=true, but is not correct code.
4639 68. The host_aton() function was not handling scoped IPv6 addresses (those
4640 with, for example, "%eth0" on the end) correctly.
4642 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4643 negated items (that is, ~something) in unsigned ints. Some compilers
4644 apparently mutter when there is no cast.
4646 70. If an address verification called from an ACL failed, and did not produce a
4647 user-specific message (i.e. there was only a "system" message), nothing was
4648 put in $acl_verify_message. In this situation, it now puts the system
4651 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4652 session; change 4.31/43 added the unwanted input to the log line - except
4653 that it did not do this in the start of session case. It now does.
4655 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4656 This is wrong and can cause the other end to generate a synchronization
4657 error if it is another Exim or anything else that does the synchronization
4658 check. A QUIT command is no longer sent after a timeout.
4660 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4661 during host lookups.
4663 74. The defer_ok option of callout verification was not working if it was used
4664 when verifying addresses in header lines, that is, for this case:
4666 verify = header_sender/callout=defer_ok
4668 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4669 those file descriptors could be used for SMTP connections. If anything
4670 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4671 could be sent to the SMTP client, causing chaos. The daemon now opens
4672 stdin, stdout, and stderr to /dev/null when it puts itself into the
4675 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4676 log by default. The user can override this with suitable Perl magic.
4678 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4679 the log message when discard triggers, was not working for the DATA ACL or
4680 for the non-SMTP ACL.
4682 78. Error message wording change in sieve.c.
4684 79. If smtp_accept_max_per_host was set, the number of connections could be
4685 restricted to fewer than expected, because the daemon was trying to set up
4686 a new connection before checking whether the processes handling previous
4687 connections had finished. The check for completed processes is now done
4688 earlier. On busy systems, this bug wouldn't be noticed because something
4689 else would have woken the daemon, and it would have reaped the completed
4692 80. If a message was submitted locally by a user whose login name contained one
4693 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4694 It caused a spool format error. I have fixed the spool reading code. A
4695 related problem was that the "from" clause in the Received: line became
4696 illegal because of the space(s). It is now covered by ${quote_local_part.
4698 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4701 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4703 83. Patch from Sieve maintainer to fix unterminated string problem in
4704 "vacation" handling.
4706 84. Some minor changes to the Linux configuration files to help with other
4707 OS variants using glibc.
4709 85. One more patch for Sieve to update vacation handling to latest spec.
4712 ----------------------------------------------------
4713 See the note above about the 4.44 and 4.50 releases.
4714 ----------------------------------------------------
4720 1. Change 4.43/35 introduced a bug that caused file counts to be
4721 incorrectly computed when quota_filecount was set in an appendfile
4724 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4725 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4728 3. Give more explanation in the error message when the command for a transport
4729 filter fails to execute.
4731 4. There are several places where Exim runs a non-Exim command in a
4732 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4733 was being done only for the command run by the queryprogram router. It is
4734 now done for all such subprocesses. The other cases are: ${run, transport
4735 filters, and the commands run by the lmtp and pipe transports.
4737 5. Some older OS have a limit of 256 on the maximum number of file
4738 descriptors. Exim was using setrlimit() to set 1000 as a large value
4739 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4740 systems. I've change it so that if it can't get 1000, it tries for 256.
4742 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4743 was an oversight, and furthermore, ever since the addition of extra
4744 controls (e.g. 4.43/32), the checks on when to allow different forms of
4745 "control" were broken. There should now be diagnostics for all cases when a
4746 control that does not make sense is encountered.
4748 7. $recipients is now available in the predata ACL (oversight).
4750 8. Tidy the search cache before the fork to do a delivery from a message
4751 received from the command line. Otherwise the child will trigger a lookup
4752 failure and thereby defer the delivery if it tries to use (for example) a
4753 cached ldap connection that the parent has called unbind on.
4755 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4756 of $address_data from the recipient verification was clobbered by the
4757 sender verification.
4759 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4760 was its contents. (It was OK if the option was not defined at all.)
4762 11. A "Completed" log line is now written for messages that are removed from
4763 the spool by the -Mrm option.
4765 12. $host_address is now set to the target address during the checking of
4766 ignore_target_hosts.
4768 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4769 being passed; this would have caused $sender_host_name to have been used if
4770 matching the list had actually called for a host name (not very likely,
4771 since this list is usually IP addresses). A host name is now passed as
4774 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4775 code that specifies a non-restarting handler (typically sigaction() in
4776 modern systems) in an attempt to fix a rare and obscure crash bug.
4778 15. Narrowed the window for a race in the daemon that could cause it to ignore
4779 SIGCHLD signals. This is not a major problem, because they are used only to
4780 wake it up if nothing else does.
4782 16. A malformed maildirsize file could cause Exim to calculate negative values
4783 for the mailbox size or file count. Odd effects could occur as a result.
4784 The maildirsize information is now recalculated if the size or filecount
4787 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4788 support for a long time. Removed HAVE_SYS_VFS_H.
4790 18. Updated exipick to current release from John Jetmore.
4792 19. Allow an empty sender to be matched against a lookup in an address list.
4793 Previously the only cases considered were a regular expression, or an
4796 20. Exim went into a mad DNS lookup loop when doing a callout where the
4797 host was specified on the transport, if the DNS lookup yielded more than
4800 21. The RFC2047 encoding function was originally intended for short strings
4801 such as real names; it was not keeping to the 75-character limit for
4802 encoded words that the RFC imposes. It now respects the limit, and
4803 generates multiple encoded words if necessary. To be on the safe side, I
4804 have increased the buffer size for the ${rfc2047: expansion operator from
4807 22. Failure to deliver a bounce message always caused it to be frozen, even if
4808 there was an errors_to setting on the router. The errors_to setting is now
4811 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4812 canonical form (fully expanded) before being placed in
4813 $sender_host_address.
4815 24. Updated eximstats to version 1.33
4817 25. Include certificate and key file names in error message when GnuTLS fails
4818 to set them up, because the GnuTLS error message doesn't include the name
4819 of the failing file when there is a problem reading it.
4821 26. Expand error message when OpenSSL has problems setting up cert/key files.
4824 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4825 (this can affect the format of dates).
4827 28. exim_tidydb, when checking for the continued existence of a message for
4828 which it has found a message-specific retry record, was not finding
4829 messages that were in split spool directories. Consequently, it was
4830 deleting retry records that should have stayed in existence.
4832 29. eximstats updated to version 1.35
4833 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4834 1.35 - bugfix such that pie charts by volume are generated correctly
4836 30. The SPA authentication driver was not abandoning authentication and moving
4837 on to the next authenticator when an expansion was forced to fail,
4838 contradicting the general specification for all authenticators. Instead it
4839 was generating a temporary error. It now behaves as specified.
4841 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4842 (the order specifies the preference for clients). The order is now AES256,
4843 AES128, 3DES, ARCFOUR128.
4845 31. Small patch to Sieve code - explicitly set From: when generating an
4848 32. Exim crashed if a remote delivery caused a very long error message to be
4849 recorded - for instance if somebody sent an entire SpamAssassin report back
4850 as a large number of 550 error lines. This bug was coincidentally fixed by
4851 increasing the size of one of Exim's internal buffers (big_buffer) that
4852 happened as part of the Exiscan merge. However, to be on the safe side, I
4853 have made the code more robust (and fixed the comments that describe what
4856 33. Some experimental protocols are using DNS PTR records for new purposes. The
4857 keys for these records are domain names, not reversed IP addresses. The
4858 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4859 leaves it alone. Component reversal etc. now happens only for IP addresses.
4862 34. The host_aton() function is supposed to be passed a string that is known
4863 to be a valid IP address. However, in the case of IPv6 addresses, it was
4864 not checking this. This is a hostage to fortune. Exim now panics and dies
4865 if the condition is not met. A case was found where this could be provoked
4866 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4867 components; fortuitously, this particular loophole had already been fixed
4868 by change 4.50/55 or 4.44/33 above.
4870 If there are any other similar loopholes, the new check in host_aton()
4871 itself should stop them being exploited. The report I received stated that
4872 data on the command line could provoke the exploit when Exim was running as
4873 exim, but did not say which command line option was involved. All I could
4874 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4875 running as the user.
4878 35. There was a buffer overflow vulnerability in the SPA authentication code
4879 (which came originally from the Samba project). I have added a test to the
4880 spa_base64_to_bits() function which I hope fixes it.
4883 36. The daemon start-up calls getloadavg() while still root for those OS that
4884 need the first call to be done as root, but it missed one case: when
4885 deliver_queue_load_max is set with deliver_drop_privilege. This is
4886 necessary for the benefit of the queue runner, because there is no re-exec
4887 when deliver_drop_privilege is set.
4889 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4890 was in use, was not putting the data itself into the right store pool;
4891 consequently, it could be overwritten for a subsequent message in the same
4892 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4895 38. Sometimes the final signoff response after QUIT could fail to get
4896 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4897 < 0 before doing a fflush(). This bug looks as though it goes back to the
4898 introduction of TLS in release 3.20, but "sometimes" must have been rare
4899 because the tests only now provoked it.
4905 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4906 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4907 Now it returns an integer. A number of calls were still expecting a T/F
4908 return. Fortuitously, in all cases, the tests worked in OK situations,
4909 which is the norm. However, things would have gone wrong on any write
4910 failures on the smtp file descriptor. This function is used when sending
4911 messages over SMTP and also when doing verify callouts.
4913 2. When Exim is called to do synchronous delivery of a locally submitted
4914 message (the -odf or -odi options), it no longer closes stderr before doing
4917 3. Implemented the mua_wrapper option.
4919 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4921 5. Implemented the functions header_remove(), header_testname(),
4922 header_add_at_position(), and receive_remove_recipient(), and exported them
4925 6. If an ACL "warn" statement specified the addition of headers, Exim already
4926 inserted X-ACL-Warn: at the start if there was no header name. However, it
4927 was not making this test for the second and subsequent header lines if
4928 there were newlines in the string. This meant that an invalid header could
4929 be inserted if Exim was badly configured.
4931 7. Allow an ACL "warn" statement to add header lines at the start or after all
4932 the Received: headers, as well as at the end.
4934 8. Added the rcpt_4xx retry error code.
4936 9. Added postmaster_mailfrom=xxx to callout verification option.
4938 10. Added mailfrom=xxxx to the callout verification option, for verify=
4941 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4942 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4944 12. Inserted some casts to stop certain compilers warning when using pointer
4945 differences as field lengths or precisions in printf-type calls (mostly
4946 affecting debugging statements).
4948 13. Added optional readline() support for -be (dynamically loaded).
4950 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4951 same clock tick as a message's arrival, so that its received time was the
4952 same as the "first fail" time on the retry record, and that message
4953 remained on the queue past the ultimate address timeout, every queue runner
4954 would try a delivery (because it was past the ultimate address timeout) but
4955 after another failure, the ultimate address timeout, which should have then
4956 bounced the address, did not kick in. This was a "< instead of <=" error;
4957 in most cases the first failure would have been in the next clock tick
4958 after the received time, and all would be well.
4960 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4961 being recognized when the domain list was tested by the match_domain
4962 condition in an expansion string.
4964 16. Added the ${str2b64: operator.
4966 17. Exim was always calling setrlimit() to set a large limit for the number of
4967 processes, without checking whether the existing limit was already
4968 adequate. (It did check for the limit on file descriptors.) Furthermore,
4969 errors from getrlimit() and setrlimit() were being ignored. Now they are
4970 logged to the main and panic logs, but Exim does carry on, to try to do its
4971 job under whatever limits there are.
4973 18. Imported PCRE 5.0.
4975 19. Trivial typo in log message " temporarily refused connection" (the leading
4978 20. If the log selector return_path_on_delivery was set and an address was
4979 redirected to /dev/null, the delivery process crashed because it assumed
4980 that a return path would always be set for a "successful" delivery. In this
4981 case, the whole delivery is bypassed as an optimization, and therefore no
4984 21. Internal re-arrangement: the function for sending a challenge and reading
4985 a response while authentication was assuming a zero-terminated challenge
4986 string. It's now changed to take a pointer and a length, to allow for
4987 binary data in such strings.
4989 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4991 23. Exim was not respecting finduser_retries when seeking the login of the
4992 uid under which it was called; it was always trying 10 times. (The default
4993 setting of finduser_retries is zero.) Also, it was sleeping after the final
4994 failure, which is pointless.
4996 24. Implemented tls_on_connect_ports.
4998 25. Implemented acl_smtp_predata.
5000 26. If the domain in control=submission is set empty, Exim assumes that the
5001 authenticated id is a complete email address when it generates From: or
5002 Sender: header lines.
5004 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5005 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5006 chown and chgrp in /bin and hostname in /usr/bin.
5008 28. Exim was keeping the "process log" file open after each use, just as it
5009 does for the main log. This opens the possibility of it remaining open for
5010 long periods when the USR1 signal hits a daemon. Occasional processlog
5011 errors were reported, that could have been caused by this. Anyway, it seems
5012 much more sensible not to leave this file open at all, so that is what now
5015 29. The long-running daemon process does not normally write to the log once it
5016 has entered its main loop, and it closes the log before doing so. This is
5017 so that log files can straightforwardly be renamed and moved. However,
5018 there are a couple of unusual error situations where the daemon does write
5019 log entries, and I had neglected to close the log afterwards.
5021 30. The text of an SMTP error response that was received during a remote
5022 delivery was being truncated at 512 bytes. This is too short for some of
5023 the long messages that one sometimes sees. I've increased the limit to
5026 31. It is now possible to make retry rules that apply only when a message has a
5027 specific sender, in particular, an empty sender.
5029 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5030 it possible to be selective about when SMTP synchronization is enforced.
5032 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5034 32. Implemented hosts_connection_nolog.
5036 33. Added an ACL for QUIT.
5038 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5041 35. Added mailbox_size and mailbox_filecount to appendfile.
5043 36. Added control = no_multiline_responses to ACLs.
5045 37. There was a bug in the logic of the code that waits for the clock to tick
5046 in the case where the clock went backwards by a substantial amount such
5047 that the microsecond fraction of "now" was more than the microsecond
5048 fraction of "then" (but the whole seconds number was less).
5050 38. Added support for the libradius Radius client library this is found on
5051 FreeBSD (previously only the radiusclient library was supported).
5057 1. When certain lookups returned multiple values in the form name=value, the
5058 quoting of the values was not always being done properly. Specifically:
5059 (a) If the value started with a double quote, but contained no whitespace,
5061 (b) If the value contained whitespace other than a space character (i.e.
5062 tabs or newlines or carriage returns) it was not quoted.
5063 This fix has been applied to the mysql and pgsql lookups by writing a
5064 separate quoting function and calling it from the lookup code. The fix
5065 should probably also be applied to nisplus, ibase and oracle lookups, but
5066 since I cannot test any of those, I have not disturbed their existing code.
5068 2. A hit in the callout cache for a specific address caused a log line with no
5069 reason for rejecting RCPT. Now it says "Previous (cached) callout
5070 verification failure".
5072 3. There was an off-by-one bug in the queryprogram router. An over-long
5073 return line was truncated at 256 instead of 255 characters, thereby
5074 overflowing its buffer with the terminating zero. As well as fixing this, I
5075 have increased the buffer size to 1024 (and made a note to document this).
5077 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5078 when Exim is waiting for an SMTP response from a remote server, Exim
5079 restarts its select() call on the socket, thereby resetting its timeout.
5080 This is not a problem when such interrupts are rare. Somebody set up a cron
5081 job to run exiwhat every 2 minutes, which is less than the normal select()
5082 timeout (5 or 10 minutes). This meant that the select() timeout never
5083 kicked in because it was always reset. I have fixed this by comparing the
5084 time when an interrupt arrives with the time at the start of the first call
5085 to select(). If more time than the timeout has elapsed, the interrupt is
5086 treated as a timeout.
5088 5. Some internal re-factoring in preparation for the addition of Sieve
5089 extensions (by MH). In particular, the "personal" test is moved to a
5090 separate function, and given an option for scanning Cc: and Bcc: (which is
5091 not set for Exim filters).
5093 6. When Exim created an email address using the login of the caller as the
5094 local part (e.g. when creating a From: or Sender: header line), it was not
5095 quoting the local part when it contained special characters such as @.
5097 7. Installed new OpenBSD configuration files.
5099 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5100 try to make them clearer.
5102 9. Callout options, other than the timeout value, were being ignored when
5103 verifying sender addresses in header lines. For example, when using
5105 verify = header_sender/callout=no_cache
5107 the cache was (incorrectly) being used.
5109 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5110 only the Cygwin environment.
5112 11. When return_path_on_delivery was set as a log selector, if different remote
5113 addresses in the same message used different return paths and parallel
5114 remote delivery occurred, the wrong values would sometimes be logged.
5115 (Whenever a remote delivery process finished, the return path value from
5116 the most recently started remote delivery process was logged.)
5118 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5119 lines when AUTH and/or TLS are in use. This is the "received protocol"
5120 field. Exim used to use "asmtp" for authenticated SMTP, without any
5121 indication (in the protocol name) for TLS use. Now it follows the RFC and
5122 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5123 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5124 names appear in log lines as well as in Received: header lines.
5126 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5127 extensions, and comparison tests, and to fix some bugs.
5129 14. Changes to the "personal" filter test:
5131 (1) The test was buggy in that it was just doing the equivalent of
5132 "contains" tests on header lines. For example, if a user's address was
5133 anne@some.where, the "personal" test would incorrectly be true for
5135 To: susanne@some.where
5137 This test is now done by extracting each address from the header in turn,
5138 and checking the entire address. Other tests that are part of "personal"
5139 are now done using regular expressions (for example, to check local parts
5140 of addresses in From: header lines).
5142 (2) The list of non-personal local parts in From: addresses has been
5143 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5144 taken from the Sieve specification recommendations.
5146 (3) If the message contains any header line starting with "List-" it is
5147 treated as non-personal.
5149 (4) The test for "circular" in the Subject: header line has been removed
5150 because it now seems ill-conceived.
5152 15. Minor typos in src/EDITME comments corrected.
5154 16. Installed latest exipick from John Jetmore.
5156 17. If headers_add on a router specified a text string that was too long for
5157 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5158 of string_sprintf() is now avoided.
5160 18. $message_body_size was not set (it was always zero) when running the DATA
5161 ACL and the local_scan() function.
5163 19. For the "mail" command in an Exim filter, no default was being set for
5164 the once_repeat time, causing a random time value to be used if "once" was
5165 specified. (If the value happened to be <= 0, no repeat happened.) The
5166 default is now 0s, meaning "never repeat". The "vacation" command was OK
5167 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5168 (I found it when inspecting the code).
5170 20. There is now an overall timeout for performing a callout verification. It
5171 defaults to 4 times the callout timeout, which applies to individual SMTP
5172 commands during the callout. The overall timeout applies when there is more
5173 than one host that can be tried. The timeout is checked before trying the
5174 next host. This prevents very long delays if there are a large number of
5175 hosts and all are timing out (e.g. when the network connections are timing
5176 out). The value of the overall timeout can be changed by specifying an
5177 additional sub-option for "callout", called "maxwait". For example:
5179 verify = sender/callout=5s,maxwait=20s
5181 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5182 to the end before writing, but this should make it even safer).
5184 22. Exim was forgetting that it had advertised PIPELINING for the second and
5185 subsequent messages on an SMTP connection. It was also not resetting its
5186 memory on STARTTLS and an internal HELO.
5188 23. When Exim logs an SMTP synchronization error within a session, it now
5189 records whether PIPELINING has been advertised or not.
5191 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5192 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5193 rather than long int.
5195 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5197 26. Added the never_mail option to autoreply.
5203 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5204 crash if the getsockname() call failed; this can happen if a connection is
5205 closed very soon after it is established. The problem was simply in the
5206 order in which certain operations were done, causing Exim to try to write
5207 to the SMTP stream before it had set up the file descriptor. The bug has
5208 been fixed by making things happen in the correct order.
5214 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5215 before the connection was closed, thus losing the rejection response.
5217 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5218 some early Solaris releases, but causes trouble in current releases where
5219 socklen_t is defined.
5221 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5224 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5227 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5228 that releasing the top of it at the end releases what was used for sub-
5229 expansions (unless the block got too big). However, discard this block if
5230 the first thing is a variable or header, so that we can use its block when
5231 it is dynamic (useful for very large $message_headers, for example).
5233 6. Lookups now cache *every* query, not just the most recent. A new, separate
5234 store pool is used for this. It can be recovered when all lookup caches are
5235 flushed. Lookups now release memory at the end of their result strings.
5236 This has involved some general refactoring of the lookup sources.
5238 7. Some code has been added to the store_xxx() functions to reduce the amount
5239 of flapping under certain conditions.
5241 8. log_incoming_interface used to affect only the <= reception log lines. Now
5242 it causes the local interface and port to be added to several more SMTP log
5243 lines, for example "SMTP connection from", and rejection lines.
5245 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5247 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5249 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5250 could be overwritten at the end of the current message (or the start of a
5251 new message if it was set in a HELO ACL). The value is now preserved for
5252 the duration of the SMTP connection.
5254 12. If a transport had a headers_rewrite setting, and a matching header line
5255 contained an unqualified address, that address was qualified, even if it
5256 did not match any rewriting rules. The underlying bug was that the values
5257 of the flags that permit the existence of unqualified sender and recipient
5258 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5259 non-local messages, and by -bnq for local messages) were not being
5260 preserved with the message after it was received.
5262 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5263 "next input=" as part of the text comprising the host identity instead of
5264 the correct text. The code was using the same buffer for two different
5265 strings. However, depending on which order the printing function evaluated
5266 its arguments, the bug did not always show up. Under Linux, for example, my
5267 test suite worked just fine.
5269 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5270 doesn't work with some older versions of Perl. It has been changed to "my",
5271 which in any case is probably the better facility to use.
5273 15. A really picky compiler found some instances of statements for creating
5274 error messages that either had too many or two few arguments for the format
5277 16. The size of the buffer for calls to the DNS resolver has been increased
5278 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5279 for addresses that have a lot of PTR records. This alleviates a problem; it
5280 does not fully solve it.
5282 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5283 buffer now truncates the list and logs the incident, which is the same
5284 action as happens when Exim is looking up a host name and its aliases.
5285 Previously in this situation something unpredictable would happen;
5286 sometimes it was "internal error: store_reset failed".
5288 18. If a server dropped the connection unexpectedly when an Exim client was
5289 using GnuTLS and trying to read a response, the client delivery process
5290 crashed while trying to generate an error log message.
5292 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5293 string, for example:
5295 warn message = H1: something\nH2: something
5297 the text was added as a single header line from Exim's point of view
5298 though it ended up OK in the delivered message. However, searching for the
5299 second and subsequent header lines using $h_h2: did not work. This has been
5300 fixed. Similarly, if a system filter added multiple headers in this way,
5301 the routers could not see them.
5303 20. Expanded the error message when iplsearch is called with an invalid key to
5304 suggest using net-iplsearch in a host list.
5306 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5307 ACLs are no longer actually imposed (and a message to that effect is
5310 22. If a "gecos" field in a passwd entry contained escaped characters, in
5311 particular, if it contained a \" sequence, Exim got it wrong when building
5312 a From: or a Sender: header from that name. A second bug also caused
5313 incorrect handling when an unquoted " was present following a character
5314 that needed quoting.
5316 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5317 was not being matched caselessly.
5319 24. Arranged for all hyphens in the exim.8 source to be escaped with
5322 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5323 properly. Recipient callout cache records were still being keyed to include
5324 the sender, even when use_sender was set false. This led to far more
5325 callouts that were necessary. The sender is no longer included in the key
5326 when use_sender is false.
5328 26. Added "control = submission" modifier to ACLs.
5330 27. Added the ${base62d: operator to decode base 62 numbers.
5332 28. dnsdb lookups can now access SRV records.
5334 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5335 the configuration file.
5337 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5338 (-v) mode. This makes the output for a verbose queue run more intelligible.
5340 31. Added a use_postmaster feature to recipient callouts.
5342 32. Added the $body_zerocount variable, containing the number of binary zero
5343 bytes in the message body.
5345 33. The time of last modification of the "new" subdirectory is now used as the
5346 "mailbox time last read" when there is a quota error for a maildir
5349 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5351 35. Added +ignore_unknown as a special item in host lists.
5353 36. Code for decoding IPv6 addresses in host lists is now included, even if
5354 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5355 address was recognized as an IP address, but was then not correctly decoded
5356 into binary, causing unexpected and incorrect effects when compared with
5363 1. Very minor rewording of debugging text in manualroute to say "list of
5364 hosts" instead of "hostlist".
5366 2. If verify=header_syntax was set, and a header line with an unqualified
5367 address (no domain) and a large number of spaces between the end of the
5368 name and the colon was received, the reception process suffered a buffer
5369 overflow, and (when I tested it) crashed. This was caused by some obsolete
5370 code that should have been removed. The fix is to remove it!
5372 3. When running in the test harness, delay a bit after writing a bounce
5373 message to get a bit more predictability in the log output.
5375 4. Added a call to search_tidyup() just before forking a reception process. In
5376 theory, someone could use a lookup in the expansion of smtp_accept_max_
5377 per_host which, without the tidyup, could leave open a database connection.
5379 5. Added the variables $recipient_data and $sender_data which get set from a
5380 lookup success in an ACL "recipients" or "senders" condition, or a router
5381 "senders" option, similar to $domain_data and $local_part_data.
5383 6. Moved the writing of debug_print from before to after the "senders" test
5386 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5387 problems for message scanning, either using a data ACL, or using
5388 local_scan() because the Received: header was not generated till after they
5389 were called (in order to set the time as the time of reception completion).
5390 I have revised the way this works. The header is now generated after the
5391 body is received, but before the ACL or local_scan() are called. After they
5392 are run, the timestamp in the header is updated.
5398 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5399 before starting a queue runner without re-exec. This happened only when
5400 deliver_drop_privilege was set or when the Exim user was set to root. The
5401 effect of the bug was that timeouts during subsequent deliveries caused
5402 crashes instead of being properly handled. The handler is now left at its
5403 default (and expected) setting.
5405 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5406 message, again when deliver_drop_privilege is set or Exim is run as root.
5407 The bug described in (1) was not present in this case, but the tidying up
5408 of the other signals was missing. I have made the two cases consistent.
5410 3. The ignore_target_hosts setting on a manualroute router was being ignored
5411 for hosts that were looked up using the /MX notation.
5413 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5416 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5417 operated on the sender address. After changing the $sender_address to <>
5418 for the sender address verify, Exim was re-instated it as the original
5419 (before rewriting) address, but remembering that it had rewritten it, so it
5420 wasn't rewriting it again. This bug also had the effect of breaking the
5421 sender address verification caching when the sender address was rewritten.
5423 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5424 This has been changed so that if the ip literal address matches
5425 ignore_target_hosts, the router declines.
5427 7. Added expansion conditions match_domain, match_address, and match_local_
5428 part (NOT match_host).
5430 8. The placeholder for the Received: header didn't have a length field set.
5432 9. Added code to Exim itself and to exim_lock to test for a specific race
5433 condition that could lead to file corruption when using MBX delivery. The
5434 issue is with the lockfile that is created in /tmp. If this file is removed
5435 after a process has opened it but before that process has acquired a lock,
5436 there is the potential for a second process to recreate the file and also
5437 acquire a lock. This could lead to two Exim processes writing to the file
5438 at the same time. The added code performs the same test as UW imapd; it
5439 checks after acquiring the lock that its file descriptor still refers to
5440 the same named file.
5442 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5443 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5446 11. Added the smtp_active_hostname option. If used, this will typically be made
5447 to depend on the incoming interface address. Because $interface_address is
5448 not set up until the daemon has forked a reception process, error responses
5449 that can happen earlier (such as "too many connections") no longer contain
5452 12. If an expansion in a condition on a "warn" statement fails because a lookup
5453 defers, the "warn" statement is abandoned, and the next ACL statement is
5454 processed. Previously this caused the whole ACL to be aborted.
5456 13. Added the iplsearch lookup type.
5458 14. Added ident_timeout as a log selector.
5460 15. Added tls_certificate_verified as a log selector.
5462 16. Added a global option tls_require_ciphers (compare the smtp transport
5463 option of the same name). This controls incoming TLS connections.
5465 17. I finally figured out how to make tls_require_ciphers do a similar thing
5466 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5467 before starting the TLS session.
5469 18. Tabs are now shown as \t in -bP output.
5471 19. If the log selector return_path_on_delivery was set, Exim crashed when
5472 bouncing a message because it had too many Received: header lines.
5474 20. If two routers both had headers_remove settings, and the first one included
5475 a superfluous trailing colon, the final name in the first list and the
5476 first name in the second list were incorrectly joined into one item (with a
5477 colon in the middle).
5483 1. Added -C and -D options to the exinext utility, mainly to make it easier
5484 to include in the automated testing, but these could be helpful when
5485 multiple configurations are in use.
5487 2. The exinext utility was not formatting the output nicely when there was
5488 an alternate port involved in the retry record key, nor when there was a
5489 message id as well (for retries that were specific to a specific message
5490 and a specific host). It was also confused by IPv6 addresses, because of
5491 the additional colons they contain. I have fixed the IPv4 problem, and
5492 patched it up to do a reasonable job for IPv6.
5494 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5495 delivery, the log line now contains "pipelined" if PIPELINING was used.
5497 4. An SMTP transport process used to panic and die if the bind() call to set
5498 an explicit outgoing interface failed. This has been changed; it is now
5499 treated in the same way as a connect() failure.
5501 5. A reference to $sender_host_name in the part of a conditional expansion
5502 that was being skipped was still causing a DNS lookup. This no longer
5505 6. The def: expansion condition was not recognizing references to header lines
5506 that used bh_ and bheader_.
5508 7. Added the _cache feature to named lists.
5510 8. The code for checking quota_filecount in the appendfile transport was
5511 allowing one more file than it should have been.
5513 9. For compatibility with Sendmail, the command line option
5521 and sets the incoming protocol and host name (for trusted callers). The
5522 host name and its colon can be omitted when only the protocol is to be set.
5523 Note the Exim already has two private options, -pd and -ps, that refer to
5524 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5525 "s", but I don't think that's a major issue.
5527 10. A number of refactoring changes to the code, none of which should affect
5530 (a) The number of logging options was getting close to filling up the
5531 32-bit word that was used as a bit map. I have split them into two classes:
5532 those that are passed in the argument to log_write(), and those that are
5533 only ever tested independently outside of that function. These are now in
5534 separate 32-bit words, so there is plenty of room for expansion again.
5535 There is no change in the user interface or the logging behaviour.
5537 (b) When building, for example, log lines, the code previously used a
5538 macro that called string_cat() twice, in order to add two strings. This is
5539 not really sufficiently general. Furthermore, there was one instance where
5540 it was actually wrong because one of the argument was used twice, and in
5541 one call a function was used. (As it happened, calling the function twice
5542 did not affect the overall behaviour.) The macro has been replaced by a
5543 function that can join an arbitrary number of extra strings onto a growing
5546 (c) The code for expansion conditions now uses a table and a binary chop
5547 instead of a serial search (which was left over from when there were very
5548 few conditions). Also, it now recognizes conditions like "pam" even when
5549 the relevant support is not compiled in: a suitably worded error message is
5550 given if an attempt is made to use such a condition.
5552 11. Added ${time_interval:xxxxx}.
5554 12. A bug was causing one of the ddress fields not to be passed back correctly
5555 from remote delivery subprocesses. The field in question was not being
5556 subsequently used, so this caused to problems in practice.
5558 13. Added new log selectors queue_time and deliver_time.
5560 14. Might have fixed a bug in maildirsizefile handling that threw up
5561 "unexpected character" debug warnings, and recalculated the data
5562 unnecessarily. In any case, I expanded the warning message to give more
5565 15. Added the message "Restricted characters in address" to the statements in
5566 the default ACL that block characters like @ and % in local parts.
5568 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5569 Three changes have been made:
5571 (a) There was a serious bug; a negative response to MAIL caused the whole
5572 recipient domain to be cached as invalid, thereby blocking all messages
5573 to all local parts at the same domain, from all senders. This bug has
5574 been fixed. The domain is no longer cached after a negative response to
5575 MAIL if the sender used is not empty.
5577 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5580 (c) A new callout option, "use_sender" has been added for people who want
5581 the modified behaviour.
5587 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5590 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5591 indeed breaks things for older releases.
5593 3. Added additional logging to the case where there is a problem reading data
5594 from a filter that is running in a subprocess using a pipe, in order to
5595 try to track down a specific problem.
5597 4. Testing facility fudge: when running in the test harness and attempting
5598 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5599 getting "No route to host". Convert this to a timeout.
5601 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5604 6. Some OS don't have socklen_t but use size_t instead. This affects the
5605 fifth argument of getsockopt() amongst other things. This is now
5606 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5607 can be set for individual OS. I have set it for SunOS5, OSF1, and
5608 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5609 some earlier ones do not.
5611 7. Change 4.30/15 was not doing the test caselessly.
5613 8. The standard form for an IPv6 address literal was being rejected by address
5614 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5615 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5616 this, as well as the form without the "IPv6" on the front (but only when
5617 address literals are enabled, of course).
5619 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5621 10. Exim crashed if a message with an empty sender address specified by -f
5622 encountered a router with an errors_to setting. This could be provoked only
5623 by a command such as
5627 where an empty string was supplied; "<>" did not hit this bug.
5629 11. Installed PCRE release 4.5.
5631 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5632 remained set. It is now erased.
5634 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5635 times from message ids (which are base 36 rather than the normal 62).
5637 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5638 were being counted as actual protocol errors, and logged if the log
5639 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5640 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5641 and DATA following a set of rejected RCPTs do not count as protocol errors.
5642 In other words, Exim assumes they were pipelined, though this may not
5643 actually be the case. Of course, in all cases the client gets an
5644 appropriate error code.
5646 15. If a lookup fails in an ACL condition, a message about the failure may
5647 be available; it is used if testing the ACL cannot continue, because most
5648 such messages specify what the cause of the deferral is. However, some
5649 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5650 that caused an old message to be retained and used if a later statement
5651 caused a defer, replacing the real cause of the deferral.
5653 16. If an IP address had so many PTR records that the DNS lookup buffer
5654 was not large enough to hold them, Exim could crash while trying to process
5655 the truncated data. It now detects and logs this case.
5657 17. Further to 4.21/58, another change has been made: if (and only if) the
5658 first line of a message (the first header line) ends with CRLF, a bare LF
5659 in a subsequent header line has a space inserted after it, so as not to
5660 terminate the header.
5662 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5663 unnecessarily, used atoi() instead of strtol(), and didn't check the
5664 termination when getting file sizes from file names by regex.
5666 19. Completely re-implemented the support for maildirsize files, in the light
5667 of a number of problems with the previous contributed implementation
5668 (4.30/29). In particular:
5670 . If the quota is zero, the maildirsize file is maintained, but no quota is
5673 . If the maildir directory does not exist, it is created before any attempt
5674 to write a maildirsize file.
5676 . The quota value in the file is just a cache; if the quota is changed in
5677 the transport, the new value overrides.
5679 . A regular expression is available for excluding directories from the
5682 20. The autoreply transport checks the characters in options that define the
5683 message's headers; it allows continued headers, but it was checking with
5684 isspace() after an embedded newline instead of explicitly looking for a
5687 21. If all the "regular" hosts to which an address was routed had passed their
5688 expiry times, and had not reached their retry times, the address was
5689 bounced, even if fallback hosts were defined. Now Exim should go on to try
5692 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5693 equivalent code in the SMTP transport. Some hosts send humungous responses
5694 to HELO/EHLO, more than 1024 it seems.
5696 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5697 gives compiler warnings in some environments. I've now done it "properly",
5700 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5701 (because of problems with the built-in one) was declared to return uschar *
5702 instead of char *, causing compiler failure.
5704 25. Fixed a file descriptor leak when processing alias/forward files.
5706 26. Fixed a minor format string issue in dbfn.c.
5708 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5710 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5711 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5714 29. When a sender address is verified, it is cached, to save repeating the test
5715 when there is more than one recipient in a message. However, when the
5716 verification involves a callout, it is possible for different callout
5717 options to be set for different recipients. It is too complicated to keep
5718 track of this in the cache, so now Exim always runs a verification when a
5719 callout is required, relying on the callout cache for the optimization.
5720 The overhead is duplication of the address routing, but this should not be
5723 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5724 to be verified with callout=postmaster, and the main callout worked but the
5725 postmaster check failed, the verification correctly failed. However, if a
5726 subsequent RCPT command asked for sender verification *without* the
5727 postmaster check, incorrect caching caused this verification also to fail,
5730 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5731 it was not caching the DNS options (qualify_single, search_parents) that
5732 were used when the lookup failed. A subsequent lookup with different
5733 options therefore always gave the same answer, though there were cases
5734 where it should not have. (Example: a "domains = !$mx_any" option on a
5735 dnslookup router: the "domains" option is always processed without any
5736 widening, but the router might have qualify_single set.) Now Exim uses the
5737 cached value only when the same options are set.
5739 32. Added John Jetmore's "exipick" utility to the distribution.
5741 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5742 than a timeout (e.g. a certificate is required, and is not provided), an
5743 Exim server now closes the connection immediately. Previously it waited for
5744 the client to close - but if the client is SSL, it seems that they each
5745 wait for each other, leading to a delay before one of them times out.
5747 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5748 maintained 0.8.x compatibility because I don't think many are using it, and
5749 it is clearly obsolete.
5751 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5754 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5755 client certificate was expired. A simple patch fixes this, though I don't
5756 understand the full logic of why the verify callback is called multiple
5759 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5760 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5761 which causes problems with some clients (such as the Certicom SSL Plus
5762 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5763 disables the coutermeasure allowing Eudora to connect."
5765 38. Exim was not checking that a write() to a log file succeeded. This could
5766 lead to Bad Things if a log got too big, in particular if it hit a file
5767 size limit. Exim now panics and dies if it cannot write to a log file, just
5768 as it does if it cannot open a log file.
5770 39. Modified OS/Makefile-Linux so that it now contains
5772 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5774 The two -D definitions ensure that Exim is compiled with large file
5775 support, which makes it possible to handle log files that are bigger than
5778 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5779 instance) a domain was checked against a named list that involved a lookup,
5780 causing $domain_data to be set, then another domain was checked against the
5781 same list, then the first domain was re-checked, the value of $domain_data
5782 after the final check could be wrong. In particular, if the second check
5783 failed, it could be set empty. This bug probably also applied to
5786 41. The strip_trailing_dot option was not being applied to the address given
5787 with the -f command-line option.
5789 42. The code for reading a message's header from the spool was incrementing
5790 $received_count, but never initializing it. This meant that the value was
5791 incorrect (doubled) while delivering a message in the same process in which
5792 it was received. In the most common configuration of Exim, this never
5793 happens - a fresh exec is done - but it can happen when
5794 deliver_drop_privilege is set.
5796 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5797 it now includes up to 150 characters of the unexpected data in the log
5800 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5801 and building data strings. The size of both of these buffers was 10 000
5802 bytes - far larger than anybody would *ever* want, thought I. Needless to
5803 say, somebody hit the limit. I have increased the maximum line length to
5804 20 000 and the maximum data length of concatenated lines to 100 000. I have
5805 also fixed two bugs, because there was no checking on these buffers. Tsk,
5806 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5807 buffer is too small.
5809 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5810 lsearch lookups. Now it does.
5812 46. When parsing a route_list item in a manualroute router, a fixed-length
5813 buffer was used for the list of hosts. I made this 1024 bytes long,
5814 thinking that nobody would ever have a list of hosts that long. Wrong.
5815 Somebody had a whole pile of complicated expansion conditions, and the
5816 string was silently truncated, leading to an expansion error. It turns out
5817 that it is easier to change to an unlimited length (owing to other changes
5818 that have happened since this code was originally written) than to build
5819 structure for giving a limitation error. The length of the item that
5820 expands into the list of hosts is now unlimited.
5822 47. The lsearch lookup could not handle data where the length of text line was
5823 more than 4095 characters. Such lines were truncated, leading to shortened
5824 data being returned. It should now handle lines of any length.
5826 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5827 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5830 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5831 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5832 no longer happens in comment lines. A list of replacements is now placed
5833 at the head of all of the source files, except those whose only change is
5834 to replace PERL_COMMAND in the very first #! line.
5836 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5837 messages on the queue, with a bottom-up merge sort, using code contributed
5838 by Michael Haardt. This should make operations like -bp somewhat faster on
5839 large queues. It won't affect queue runners, except when queue_run_in_order
5842 51. Installed eximstats 1.31 in the distribution.
5844 52. Added support for SRV lookups to the dnslookup router.
5846 53. If an ACL referred to $message_body or $message_body_end, the value was not
5847 reset for any messages that followed in the same SMTP session.
5849 54. The store-handling optimization for building very long strings was not
5850 differentiating between the different store pools. I don't think this
5851 actually made any difference in practice, but I've tidied it.
5853 55. While running the routers to verify a sender address, $sender_address
5854 was still set to the sender address. This is wrong, because when routing to
5855 send a bounce to the sender, it would be empty. Therefore, I have changed
5856 it so that, while verifying a sender address, $sender_address is set to <>.
5857 (There is no change to what happens when verifying a recipient address.)
5859 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5860 target A or AAAA records (if not already returned) without resetting the
5861 qualify_single or search_parents options of the DNS resolver. These are
5862 inappropriate in this case because the targets of MX and SRV records must
5863 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5864 target that, when qualified, matched something in the local domain. These
5865 two options are now turned off when doing these lookups.
5867 57. It seems that at least some releases of Reiserfs (which does not have the
5868 concept of a fixed number of inodes) returns zero and not -1 for the
5869 number of available inodes. This interacted badly with check_spool_inodes,
5870 which assumed that -1 was the "no such thing" setting. What I have done is
5871 to check that the total number of inodes is greater than zero before doing
5872 the test of how many are available.
5874 58. When a "warn" ACL statement has a log_message modifier, the message is
5875 remembered, and not repeated. This is to avoid a lot of repetition when a
5876 message has many recipients that cause the same warning to be written.
5877 However, Exim was preserving the list of already written lines for an
5878 entire SMTP session, which doesn't seem right. The memory is now reset if a
5879 new message is started.
5881 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5882 debugging output unless log_rewrite was also set.
5884 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5885 of (int)(handle) when we know that handle contains (void *)(-1).
5887 61. The Exim daemon panic-logs an error return when it closes the incoming
5888 connection. However "connection reset by peer" seems to be common, and
5889 isn't really an error worthy of noting specially, so that particular error
5892 62. When Exim is trying to find all the local interfaces, it used to panic and
5893 die if the ioctl to get the interface flags failed. However, it seems that
5894 on at least one OS (Solaris 9) it is possible to have an interface that is
5895 included in the list of interfaces, but for which you get a failure error
5896 for this call. This happens when the interface is not "plumbed" into a
5897 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5898 failure of the "get flags" call assumes that the interface is down.
5900 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5901 makes life easier for people who are doing arithmetic on fields extracted
5902 from dates, where you often get leading zeros that should not be
5903 interpreted as octal.
5905 64. Added qualify_domain to the redirect router, to override the global
5908 65. If a pathologically long header line contained very many addresses (the
5909 report of this problem mentioned 10 000) and each of them was rewritten,
5910 Exim could use up a very large amount of memory. (It kept on making new
5911 copies of the header line as it rewrote, and never released the old ones.)
5912 At the expense of a bit more processing, the header rewriting function has
5913 been changed so that it no longer eats memory in this way.
5915 66. The generation of the Received: header has been moved from the time that a
5916 message starts to be received, to the time that it finishes. The timestamp
5917 in the Received: header should now be very close to that of the <= log
5918 line. There are two side-effects of this change:
5920 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5921 logged header lines no longer include the local Received: line, because
5922 it has not yet been created. The same applies to a copy of the message
5923 that is returned to a non-SMTP sender when a message is rejected.
5925 (b) When a filter file is tested using -bf, no additional Received: header
5926 is added to the test message. After some thought, I decided that this
5929 This change does not affect the value of $received_for. It is still set
5930 after address rewriting, but before local_scan() is called.
5932 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5934 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5935 gave an unhelpful panic error message, and a defer error. I have managed to
5936 change this behaviour so that it now rejects any supplied certificate,
5937 which seems right, as the list of acceptable certificates is empty.
5939 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5940 gave an unhelpful defer error. I have not managed to make this reject any
5941 supplied certificates, but the error message it gives is "no certificate
5942 supplied", which is not helpful.
5944 70. exigrep's output now also includes lines that are not associated with any
5945 message, but which match the given pattern. Implemented by a patch from
5946 Martin Sluka, which also tidied up the Perl a bit.
5948 71. Recipient callout verification, like sender verification, was using <> in
5949 the MAIL FROM command. This isn't really the right thing, since the actual
5950 sender may affect whether the remote host accepts the recipient or not. I
5951 have changed it to use the actual sender in the callout; this means that
5952 the cache record is now keyed on a recipient/sender pair, not just the
5953 recipient address. There doesn't seem to be a real danger of callout loops,
5954 since a callout by the remote host to check the sender would use <>.
5955 [SEE ABOVE: changed after hitting problems.]
5957 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5958 temporary errors. However, in the case of such a code being given after
5959 the end of a data transmission (i.e. after ".") Exim was failing to write
5960 a retry record for the message. (Yes, there was some broken host that was
5961 actually sending 8xx at this point.)
5963 73. An unknown lookup type in a host list could cause Exim to panic-die when
5964 the list was checked. (An example that provoked this was putting <; in the
5965 middle of a list instead of at the start.) If this happened during a DATA
5966 ACL check, a -D file could be left lying around. This kind of configuration
5967 error no longer causes Exim to die; instead it causes a defer error. The
5968 incident is still logged to the main and panic logs.
5970 74. Buglet left over from Exim 3 conversion. The message "too many messages
5971 in one connection" was written to the rejectlog but not the mainlog, except
5972 when address rewriting (yes!) was being logged.
5974 75. Added write_rejectlog option.
5976 76. When a system filter was run not as root (that is, when system_filter_user
5977 was set), the values of the $n variables were not being returned to the
5978 main process; thus, they were not subsequently available in the $sn
5981 77. Added +return_path_on_delivery log selector.
5983 78. A connection timeout was being treated differently from recipients deferred
5984 when testing hosts_max_try with a message that was older than the host's
5985 retry timeout. (The host should not be counted, thus allowing all hosts to
5986 be tried at least once before bouncing.) This may have been the cause of an
5987 occasionally reported bug whereby a message would remain on the queue
5988 longer than the retry timeout, but would be bounced if a delivery was
5989 forced. I say "may" because I never totally pinned down the problem;
5990 setting up timeout/retry tests is difficult. See also the next item.
5992 79. The ultimate address timeout was not being applied to errors that involved
5993 a combination of host plus message (for example, a timeout on a MAIL
5994 command). When an address resolved to a number of possible hosts, and they
5995 were not all tried for each delivery (e.g. because of hosts_max_try), a
5996 message could remain on the queue longer than the retry timeout.
5998 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6001 81. Fixed an obscure SMTP outgoing bug which required at least the following
6002 conditions: (a) there was another message waiting for the same server;
6003 (b) the server returned 5xx to all RCPT commands in the first message so
6004 that the message was not completed; (c) the server dropped the connection
6005 or gave a negative response to the RSET that Exim sends to abort the
6006 transaction. The observed case was a dropped connection after DATA that had
6007 been sent in pipelining mode. That is, the server had advertised PIPELINING
6008 but was not implementing it correctly. The effect of the bug was incorrect
6009 behaviour, such as trying another host, and this could lead to a crash.
6015 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6016 and daemon.c were passed as pointers to ints; they should have been
6017 pointers to socklen_t variables (which are typically unsigned ints).
6019 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6022 3. Fixed a really odd bug that affected only the testing scheme; patching a
6023 certain fixed string in the binary changed the value of another string that
6024 happened to be identical to the end of the original first string.
6026 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6027 name", it returns that address as the IP address. On some operating
6028 systems (e.g. Solaris), it also passes back the IP address string as the
6029 "host name". However, on others (e.g. Linux), it passes back an empty
6030 string. Exim wasn't checking for this, and was changing the host name to an
6031 empty string, assuming it had been canonicalized.
6033 5. Although rare, it is permitted to have more than one PTR record for a given
6034 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6035 all the names associated with an address, because they do in Solaris.
6036 However, it seems that they do not in Linux for data that comes from the
6037 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6038 I found this out when I moved to a new Linux workstation and tried to run
6039 the Exim test suite.
6041 To get round this problem I have changed the code so that it now does its
6042 own call to the DNS to look up PTR records when searching for a host name.
6043 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6044 addresses that are only in /etc/hosts are still found.
6046 This behaviour is, however, controlled by an option called host_lookup_
6047 order, which defaults to "bydns:byaddr". If people want to use the other
6048 order, or indeed, just use one or the other means of lookup, they can
6049 specify it in this variable.
6051 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6052 some operating systems, this comes back from gethostbyaddr() as an empty
6053 string, and this is what Exim used to test for. However, it seems that in
6054 other systems, "." is yielded. Exim now tests for this case too.
6056 7. The values of check_spool_space and check_log_space are now held internally
6057 as a number of kilobytes instead of an absolute number of bytes. If a
6058 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6059 kilobyte. This means that much larger values can be stored.
6061 8. Exim monitor: an attempt to get the action menu when not actually pointing
6062 at a message produces an empty menu entitled "No message selected". This
6063 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6064 no entries in it ("Shell widget menu has zero width and/or height"). So I
6065 have added a single, blank menu entry in this case.
6067 9. Added ${quote_local_part.
6069 10. MIME decoding is now applied to the contents of Subject: header lines when
6072 11. Now that a reference to $sender_host_address automatically causes a reverse
6073 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6074 host lookup before query-style lookups in lists that might use this
6075 variable. This has therefore been abolished, and the "net-" prefix is no
6076 longer necessary for query-style lookups.
6078 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6079 have been a typo for LFLAGS, so it has been changed.
6081 13. The install script calls Exim with "-C /dev/null" in order to find the
6082 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6083 to be output. However, since Exim outputs its version number before the
6084 error, it didn't break the script. It just looked ugly. I fixed this by
6085 always allowing "-C /dev/null" if the caller is root.
6087 14. Ignore overlarge ACL variable number when reading spool file - insurance
6088 against a later release with more variables having written the file.
6090 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6091 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6092 this, as well as the form without the "IPv6" on the front.
6094 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6095 OS/Makefile-Darwin file.
6097 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6098 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6100 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6101 variable. However, some OS use ints for inodes. I've added cast to long int
6102 to get rid of the compiler warning.
6104 19. I had forgotten to lock out "/../" in configuration file names when
6105 ALT_CONFIG_PREFIX was set.
6107 20. Routers used for verification do not need to specify transports. However,
6108 if such a router generated a host list, and callout was configured, Exim
6109 crashed, because it could not find a port number from the (non-existent)
6110 transport. It now assumes port 25 in this circumstance.
6112 21. Added the -t option to exigrep.
6114 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6115 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6116 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6117 them is set, LOOKUP_LSEARCH is forced.
6119 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6121 24. Added sender and host information to the "rejected by local_scan()" log
6122 line; previously there was no indication of these.
6124 25. Added .include_if_exists.
6126 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6127 Linux. It turns out that not all file systems support this. Apparently some
6128 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6129 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6130 not supported on the file descriptor, is now ignored when Exim is trying to
6131 sync a directory. This applies only to Linux.
6133 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6135 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6136 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6139 29. Added support for maildirsize files from supplied patch (modified a bit).
6141 30. The use of :fail: followed by an empty string could lead Exim to respond to
6142 sender verification failures with (e.g.):
6144 550 Verification failed for <xxx>
6145 550 Sender verify failed
6147 where the first response line was missing the '-' that indicates it is not
6148 the final line of the response.
6150 31. The loop for finding the name of the user that called Exim had a hardwired
6151 limit of 10; it now uses the value of finduser_retries, which is used for
6152 all other user lookups.
6154 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6157 33. Exim was neglecting to zero errno before one call of strtol() when
6158 expanding a string and expecting an integer value. On some systems this
6159 resulted in spurious "integer overflow" errors. Also, it was casting the
6160 result into an int without checking.
6162 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6163 did not work. The code looks as if it has *never* worked, though it appears
6164 to have been documented since at least release 1.62. I have made it work.
6166 35. The "timeout_DNS" error in retry rules, also documented since at least
6167 1.62, also never worked. As it isn't clear exactly what this means, and
6168 clearly it isn't a major issue, I have abolished the feature by treating it
6169 as "timeout", and writing a warning to the main and panic logs.
6171 36. The display of retry rules for -brt wasn't always showing the error code
6174 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6175 timeout_connect_A, timeout_connect_MX.
6177 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6178 to the empty sender.
6180 39. The daemon was not analysing the content of -oX till after it had closed
6181 stderr and disconnected from the controlling terminal. This meant that any
6182 syntax errors were only noted on the panic log, and the return code from
6183 the command was 0. By re-arranging the code a little, I've made the
6184 decoding happen first, so such errors now appear on stderr, and the return
6185 code is 1. However, the actual setting up of the sockets still happens in
6186 the disconnected process, so errors there are still only recorded on the
6189 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6190 connections (as happens on some IP stacks) was logged at start up time as
6191 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6192 it from "IPv6 and IPv4", which means that two separate sockets are being
6195 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6196 says whether AF_INET or AF_INET6 was passed as an argument.
6198 42. Exiwhat output was messed up when time zones were included in log
6201 43. Exiwhat now gives more information about the daemon's listening ports,
6202 and whether -tls-on-connect was used.
6204 44. The "port" option of the smtp transport is now expanded.
6206 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6207 silently ignored. Now an error message is written to the main and panic
6210 46. There's a new ACL modifier called "logwrite" which writes to a log file
6211 as soon as it is encountered.
6213 47. Added $local_user_uid and $local_user_gid at routing time.
6215 48. Exim crashed when trying to verify a sender address that was being
6218 49. Exim was recognizing only a space character after ".include". It now also
6219 recognizes a tab character.
6221 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6222 extracting the relevant information from the specification. The man page no
6223 longer contains scrambled data for the -d option, and I've added a section
6224 at the front about calling Exim under different names.
6226 51. Added "extra_headers" argument to the "mail" command in filter files.
6228 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6231 53. Installed eximstats 1.29.
6233 54. Added transport_filter_timeout as a generic transport option.
6235 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6236 Cc: header lines. This was required by RFC 822, but it not required by RFC
6239 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6240 incoming messages that did not have them. Now it does so only if the
6241 message originates locally, that is, if there is no associated remote host
6242 address. When Resent- header lines are present, this applies to the Resent-
6243 lines rather than the non-Resent- lines.
6245 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6246 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6248 58. Messages for configuration errors now include the name of the main
6249 configuration file - useful now that there may be more than one file in a
6250 list (.included file names were always shown).
6252 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6253 for those rare installations that do not start the daemon as root or run it
6254 setuid root. I've cut out the call to initgroups() if the daemon is not
6257 60. The Exim user and group can now be bound into the binary as text strings
6258 that are looked up at the start of Exim's processing.
6260 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6262 62. Added $mailstore_basename variable.
6264 63. Installed patch to sieve.c from Michael Haardt.
6266 64. When Exim failed to open the panic log after failing to open the main log,
6267 the original message it was trying to log was written to stderr and debug
6268 output, but if they were not available (the usual case in production), it
6269 was lost. Now it is written to syslog before the two lines that record the
6270 failures to open the logs.
6272 65. Users' Exim filters run in subprocesses under the user's uid. It is
6273 possible for a "deliver" command or an alias in a "personal" command to
6274 provoke an address rewrite. If logging of address rewriting is configured,
6275 this fails because the process is not running as root or exim. There may be
6276 a better way of dealing with this, but for the moment (because 4.30 needs
6277 to be released), I have disabled address rewrite logging when running a
6278 filter in a non-root, non-exim process.
6284 1. The buildconfig auxiliary program wasn't quoting the value set for
6285 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6286 not defined. This bug was masked in 4.22 by the effect that was fixed in
6289 2. Some messages that were rejected after a message id was allocated were
6290 shown as "incomplete" by exigrep. It no longer does this for messages that
6291 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6293 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6294 have allow_domain_literals set, the ID did not get logged in the <= line.
6295 Domain literals are now always recognized in Message-ID: header lines.
6297 4. The first argument for a ${extract expansion item is the key name or field
6298 number. Leading and trailing spaces in this item were not being ignored,
6299 causing some misleading effects.
6301 5. When deliver_drop_privilege was set, single queue runner processes started
6302 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6303 same command in the process it spins off) were not dropping privilege.
6305 6. When the daemon running as "exim" started a queue runner, it always
6306 re-executed Exim in the spun-off process. This is a waste of effort when
6307 deliver_drop_privilege is set. The new process now just calls the
6308 queue-runner function directly.
6314 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6317 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6318 ignored. Though the use of -oP was forcing the writing of a pid file, it
6319 was always written to the default place.
6321 3. If the message "no IP address found for host xxxx" is generated during
6322 incoming verification, it is now followed by identification of the incoming
6323 connection (so you can more easily find what provoked it).
6325 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6327 5. Added some features to "harden" Exim a bit more against certain attacks:
6329 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6330 be put in Local/Makefile. This is like the never_users runtime option,
6331 but it cannot be overridden. The default setting is "root".
6333 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6334 prefix string with which any file named in a -C command line option
6337 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6338 is retained for -C and -D only if the caller of Exim is root. Without
6339 it, the exim user may also use -C and -D and retain privilege.
6341 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6342 command line option is disabled.
6344 6. Macro names set by the -D option must start with an upper case letter, just
6345 like macro names defined in the configuration file.
6347 7. Added "dereference=" facility to LDAP.
6349 8. Two instances of the typo "uknown" in the source files are fixed.
6351 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6352 the Configure-Makefile script screwed up while processing it.
6354 10. Incorporated PCRE 4.4.
6356 11. The SMTP synchronization check was not operating right at the start of an
6357 SMTP session. For example, it could not catch a HELO sent before the client
6358 waited for the greeting. There is now a check for outstanding input at the
6359 point when the greeting is written. Because of the duplex, asynchronous
6360 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6361 way, but not yet received, when the check is performed.
6363 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6364 on TCP/IP sockets, because this apparently causes some broken clients to
6367 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6368 unchanged) from the Cygwin maintainer.
6370 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6371 was supported instead of testing for mbx. Effectively a typo.
6373 15. The spa authenticator server code was not checking that the input it
6374 received was valid base64.
6376 16. The debug output line for the "set" modifier in ACLs was not showing the
6377 name of the variable that was being set.
6379 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6381 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6382 lookup on its own. Something else was needed to trigger the lookup. For
6383 example, a match in host_lookup or the need for a host name in a host list.
6384 Now, if $sender_host_name is referenced and the host name has not yet been
6385 looked up, a lookup is performed. If the lookup fails, the variable remains
6386 empty, and $host_lookup_failed is set to "1".
6388 19. Added "eqi" as a case-independent comparison operator.
6390 20. The saslauthd authentication condition could segfault if neither service
6391 nor realm was specified.
6393 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6394 error message that was logged was misleading, and incoming SMTP
6395 connections were dropped. The message is now more accurate, and temporary
6396 errors are given to SMTP connections.
6398 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6399 (except RSET) with a 421 error, until QUIT is received. However, it was
6400 failing to send a response to QUIT.
6402 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6403 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6404 ACL is now run after the helo_try_verify_hosts code.
6406 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6407 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6408 case-independent, so other case variants are also recognized). Apparently
6409 some systems use these upper case variants.
6411 25. If more than two messages were waiting for the same host, and a transport
6412 filter was specified for the transport, Exim sent two messages over the
6413 same TCP/IP connection, and then failed with "socket operation on non-
6414 socket" when it tried to send the third.
6416 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6418 27. The extern definition of crypt16() in expand.c was not being excluded when
6419 the OS had its own crypt16() function.
6421 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6422 as a preferred synonym for return_size_limit, both as an option and as an
6425 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6427 30. Changed the default configuration ACL to relax the local part checking rule
6428 for addresses that are not in any local domains. For these addresses,
6429 slashes and pipe symbols are allowed within local parts, but the sequence
6430 /../ is explicitly forbidden.
6432 31. SPA server authentication was not clearing the challenge buffer before
6435 32. log_message in a "warn" ACL statement was writing to the reject log as
6436 well as to the main log, which contradicts the documentation and doesn't
6437 seem right (because no rejection is happening). So I have stopped it.
6439 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6440 However, I am unable to do any testing of this.
6442 34. Fixed an infelicity in the appendfile transport. When checking directories
6443 for a mailbox, to see if any needed to be created, it was accidentally
6444 using path names with one or more superfluous leading slashes; tracing
6445 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6447 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6448 contents are added to the log line that is written for every discarded
6449 recipient. (Previously a log_message setting was ignored.)
6451 36. The ${quote: operator now quotes the string if it is empty.
6453 37. The install script runs exim in order to find its version number. If for
6454 some reason other than non-existence or emptiness, which it checks, it
6455 could not run './exim', it was installing it with an empty version number,
6456 i.e. as "exim-". This error state is now caught, and the installation is
6459 38. An argument was missing from the function that creates an error message
6460 when Exim fails to connect to the socket for saslauthd authentication.
6461 This could cause Exim to crash, or give a corrupted message.
6463 39. Added isip, isip4, and isip6 to ${if conditions.
6465 40. The ACL variables $acl_xx are now saved with the message, and can be
6466 accessed later in routers, transports, and filters.
6468 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6469 strings in the file are not string-expanded.
6471 42. If a MAIL command specified a SIZE value that was too large to fit into an
6472 int variable, the check against message_size_limit failed. Such values are
6473 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6474 day this will have to be increased, but I don't think I want to be around
6475 when emails are that large.
6482 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6483 iconv() is not standard in FreeBSD.
6485 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6486 IPv6 enabled. The observed symptom was a segmentation fault on return from
6487 the function os_common_find_running_interfaces() in src/os.c.
6489 3. In the check_special_case() function in daemon.c I had used "errno" as an
6490 argument name, which causes warnings on some systems. This was basically a
6491 typo, since it was named "eno" in the comments!
6493 4. The code that waits for the clock to tick (at a resolution of some fraction
6494 of a second) so as to ensure message-id uniqueness was always waiting for
6495 at least one whole tick, when it could have waited for less. [This is
6496 almost certainly not relevant at current processor speeds, where it is
6497 unlikely to ever wait at all. But we try to future-proof.]
6499 5. The function that sleeps for a time interval that includes fractions of a
6500 second contained a race. It did not block SIGALRM between setting the
6501 timer, and suspending (a couple of lines later). If the interval was short
6502 and the sigsuspend() was delayed until after it had expired, the suspension
6503 never ended. On busy systems this could lead to processes getting stuck for
6506 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6507 process, before it forks any delivery processes. The open lookup caching
6508 mechanism meant that the open file or database connection was passed into
6509 the delivery process. The problem was that delivery processes always tidy
6510 up cached lookup data. This could cause a problem for the next delivery
6511 process started by the queue runner, because the external queue runner
6512 process does not know about the closure. So the next delivery process
6513 still has data in the lookup cache. In the case of a file lookup, there was
6514 no problem because closing a file descriptor in a subprocess doesn't affect
6515 the parent. However, if the lookup was caching a connection to a database,
6516 the connection was closed, and the second delivery process was likely to
6517 see errors such as "PGSQL: query failed: server closed the connection
6518 unexpectedly". The problem has been fixed by closing all cached lookups
6519 in a queue runner before running a delivery process.
6521 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6522 seem to have the "const" qualifier which it has on other OS. I've
6525 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6526 *of the same type* (client or server) with the same public name that an
6527 error should be diagnosed.
6529 9. When Exim looked up a host name for an IP address, but failed to find the
6530 original IP address when looking up the host name (a safety check), it
6531 output the message "<ip address> does not match any IP for NULL", which was
6532 confusing, to say the least. The bug was that the host name should have
6533 appeared instead of "NULL".
6535 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6536 user that is built into the binary, and the -C or -D options is used, root
6537 privilege is dropped before the configuration file is read. In addition,
6538 logging is switched to stderr instead of the normal log files. If the
6539 configuration then re-defines the Exim user, the unprivileged environment
6540 is probably not what is expected, so Exim logs a panic warning message (but
6543 However, if deliver_drop_privilege is set, the unprivileged state may well
6544 be exactly what is intended, so the warning has been cut out in that case,
6545 and Exim is allowed to try to write to its normal log files.
6551 1. smtp_return_error_details was not giving details for temporary sender
6552 or receiver verification errors.
6554 2. Diagnose a configuration error if two authenticators have the same public
6557 3. Exim used not to create the message log file for a message until the first
6558 delivery attempt. This could be confusing when incoming messages were held
6559 for policy or load reasons. The message log file is now created at the time
6560 the message is received, and an initial "Received" line is written to it.
6562 4. The automatically generated man page for command line options had a minor
6563 bug that caused no ill effects; however, a more serious problem was that
6564 the procedure for building the man page automatically didn't always
6565 operate. Consequently, release 4.20 contains an out-of-date version. This
6566 shouldn't happen again.
6568 5. When building Exim with embedded Perl support, the script that builds the
6569 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6570 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6572 6. The freeze_tell option was not being used for messages that were frozen on
6573 arrival, either by an ACL or by local_scan().
6575 7. Added the smtp_incomplete_transaction log selector.
6577 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6578 was accepting AUTH without a new EHLO.
6580 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6581 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6584 10. Logging of TCP/IP connections (when configured) now happens in the main
6585 daemon process instead of the child process, so that the TCP/IP connection
6586 count is more accurate (but it can never be perfect).
6588 11. The use of "drop" in a nested ACL was not being handled correctly in the
6589 outer ACL. Now, if condition failure induced by the nested "drop" causes
6590 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6591 or "require"), the connection is dropped.
6593 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6594 that yield "discard" can now be used with an "accept" or a "discard" verb,
6595 but an error is generated for any others (because I can't see a useful way
6596 to define what should happen).
6598 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6599 are now processed in the same way as lines in the Exim configuration file.
6600 In particular, continuation lines are supported.
6602 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6604 15. Added -ti meaning -t -i.
6606 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6607 domains, and warn by logging if others are found.
6609 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6610 returned from GIFCONF when Exim is trying to find the list of interfaces on
6611 a host. The code in os.c has been modified to copy each ifreq to an aligned
6612 structure in all cases.
6614 Also, in some cases, the returned ifreq's were being copied to a 'struct
6615 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6616 means the last couple of bytes of an IPv6 address could be chopped if the
6617 ifreq contained only a normal sockaddr (14 bytes storage).
6619 18. Named domain lists were not supported in the hosts_treat_as_local option.
6620 An entry such as +xxxx was not recognized, and was treated as a literal
6623 19. Ensure that header lines added by a DATA ACL are included in the reject log
6624 if the ACL subsequently rejects the message.
6626 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6627 MD5 (which is deprecated).
6629 21. When testing a filter file using -bf, Exim was writing a message when it
6630 took the sender from a "From " line in the message, but it was not doing so
6631 when it took $return_path from a Return-Path: header line. It now does.
6633 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6634 with a valid header line field name (a series of printing characters
6635 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6637 23. Changed "disc" in the source to "disk" to conform to the documentation and
6638 the book and for uniformity.
6640 24. Ignore Sendmail's -Ooption=value command line item.
6642 25. When execve() failed while trying to run a command in a pipe transport,
6643 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6644 could be confused with a return value of 69 from the command itself. This
6645 has been changed to 127, the value the shell returns if it is asked to run
6646 a non-existent command. The wording for the related log line suggests a
6647 non-existent command as the problem.
6649 26. If received_header_text expands to an empty string, do not add a Received:
6650 header line to the message. (Well, it adds a token one on the spool, but
6651 marks it "old" so that it doesn't get used or transmitted.)
6653 27. Installed eximstats 1.28 (addition of -nt option).
6655 28. There was no check for failure on the call to getsockname() in the daemon
6656 code. This can fail if there is a shortage of resources on the system, with
6657 ENOMEM, for example. A temporary error is now given on failure.
6659 29. Contrary to the C standard, it seems that in some environments, the
6660 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6661 program. Exim now does this explicitly; it affects the formatting of
6662 timestamps using strftime().
6664 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6665 complaints. I've now initialized all the variables, to avoid this.
6667 32. Header lines added by a system filter were not being "seen" during
6668 transport-time rewrites.
6670 33. The info_callback() function passed to OpenSSL is set up with type void
6671 (*)(SSL *, int, int), as described somewhere. However, when calling the
6672 function (actually a macro) that sets it up, the type void(*)() is
6673 expected. I've put in a cast to prevent warnings from picky compilers.
6675 34. If a DNS black list lookup found a CNAME record, but there were no A
6676 records associated with the domain it pointed at, Exim crashed.
6678 35. If a DNS black list lookup returned more than one A record, Exim ignored
6679 all but the first. It now scans all returned addresses if a particular IP
6680 value is being sought. In this situation, the contents of the
6681 $dnslist_value variable are a list of all the addresses, separated by a
6684 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6685 to accept a host name and all its aliases if the forward lookup for any of
6686 them yielded the IP address of the incoming connection. Now it accepts only
6687 those names whose forward lookup yields the correct IP address. Any other
6688 names are discarded. This closes a loophole whereby a rogue DNS
6689 administrator could create reverse DNS records to break through a
6690 wildcarded host restriction in an ACL.
6692 37. If a user filter or a system filter that ran in a subprocess used any of
6693 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6694 the wrong values were passed to the pipe command ($thisaddress had the
6695 value of $0, $0 had the value of $1, etc). This bug was introduced by
6696 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6698 38. Improved the line breaking for long SMTP error messages from ACLs.
6699 Previously, if there was no break point between 40 and 75 characters, Exim
6700 left the rest of the message alone. Two changes have been made: (a) I've
6701 reduced the minimum length to 35 characters; (b) if it can't find a break
6702 point between 35 and 75 characters, it looks ahead and uses the first one
6703 that it finds. This may give the occasional overlong line, but at least the
6704 remaining text gets split now.
6706 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6707 file descriptors might be low, and that setting 1000 would always raise it.
6708 It turns out that in some environments, the limit is already over 1000 and
6709 that lowering it causes trouble. So now Exim takes care not to decrease it.
6711 40. When delivering a message, the value of $return_path is set to $sender_
6712 address at the start of routing (routers may change the value). By an
6713 oversight, this default was not being set up when an address was tested by
6714 -bt or -bv, which affected the outcome if any router or filter referred to
6717 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6718 log only when "message" or "log_message" are set. However, if one of the
6719 conditions was an address verification, or a call to a nested ACL, the
6720 messages generated by the underlying test were being passed through. This
6721 no longer happens. The underlying message is available in $acl_verify_
6722 message for both "message" and "log_message" expansions, so it can be
6723 passed through if needed.
6725 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6726 new expansion $bh_ to give the encoded byte string without charset
6727 translation. Translation happens only if iconv() is available; HAVE_ICONV
6728 indicates this at build time. HEADERS_CHARSET gives the charset to
6729 translate to; headers_charset can change it in the configuration, and
6730 "headers charset" can change it in an individual filter file.
6732 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6733 that creates RFC 2047 encoded "words" labels them as that charset instead
6734 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6735 expansion operator; (ii) when Exim creates a From: line for a local
6736 message; (iii) when a header line is rewritten to include a "phrase" part.
6738 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6739 buggy, causing it to skip the first lines of messages whose message ID
6740 ended in 'D'. This would not have bitten before Exim release 4.14, because
6741 message IDs were unlikely to end in 'D' before then. The effect was to have
6742 incorrect size information for certain domains.
6744 45. #include "config.h" was missing at the start of the crypt16.c module. This
6745 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6748 46. If there was a timeout during a "random" callout check, Exim treated it as
6749 a failure of the random address, and carried on sending RSET and the real
6750 address. If the delay was just some slowness somewhere, the response to the
6751 original RCPT would be taken as a response to RSET and so on, causing
6752 mayhem of various kinds.
6754 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6755 when I implemented it. It didn't allow for the fact that some option values
6756 may legitimately be negative (e.g. size_addition), and it didn't even do
6757 the right test for positive values.
6759 48. Domain names in DNS records are case-independent. Exim always looks them up
6760 in lower case. Some resolvers return domain names in exactly the case they
6761 appear in the zone file, that is, they may contain uppercase letters. Not
6762 all resolvers do this - some return always lower case. Exim was treating a
6763 change of case by a resolver as a change of domain, similar to a widening
6764 of a domain abbreviation. This triggered its re-routing code and so it was
6765 trying to route what was effectively the same domain again. This normally
6766 caused routing to fail (because the router wouldn't handle the domain
6767 twice). Now Exim checks for this case specially, and just changes the
6768 casing of the domain that it ultimately uses when it transmits the message
6771 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6774 50. If a filter generated a file delivery with a non-absolute name (possible if
6775 no home directory exists for the router), the forbid_file option was not
6778 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6779 the existing equality matching.
6781 52. Exim was using ints instead of ino_t variables in some places where it was
6782 dealing with inode numbers.
6784 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6785 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6786 called TMPDIR, and if it finds it is different, it changes its value.
6788 54. The smtp_printf() function is now made available to local_scan() so
6789 additional output lines can be written before returning. There is also an
6790 smtp_fflush() function to enable the detection of a dropped connection.
6791 The variables smtp_input and smtp_batched_input are exported to
6794 55. Changed the default runtime configuration: the message "Unknown user"
6795 has been removed from the ACL, and instead placed on the localuser router,
6796 using the cannot_route_message feature. This means that any verification
6797 failures that generate their own messages won't get overridden. Similarly,
6798 the "Unrouteable address" message that was in the ACL for unverifiable
6799 relay addresses has also been removed.
6801 56. Added hosts_avoid_esmtp to the smtp transport.
6803 57. The exicyclog script was not checking for the esoteric option
6804 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6805 will work only if exicyclog is run under the appropriate euid.
6807 58. Following a discussion on the list, the rules by which Exim recognises line
6808 endings on incoming messages have been changed. The -dropcr and drop_cr
6809 options are now no-ops, retained only for backwards compatibility. The
6810 following line terminators are recognized: LF CRLF CR. However, special
6811 processing applies to CR:
6813 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6814 nor a local message in the state where . is a terminator.
6816 (ii) If a bare CR is encountered in a header line, an extra space is added
6817 after the line terminator so as not to end the header. The reasoning
6818 behind this is that bare CRs in header lines are most likely either
6819 to be mistakes, or people trying to play silly games.
6821 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6822 was being incorrectly given as 18 bytes larger than it should have been.
6823 This is a VOB (very old bug).
6825 60. This may never have affected anything current, but just in case it has:
6826 When the local host is found other than at the start of a list of hosts,
6827 the local host, those with the same MX, and any that follow, are discarded.
6828 When the list in question was part of a longer list of hosts, the following
6829 hosts (not currently being processed) were also being discarded. This no
6830 longer happens. I'm not sure if this situation could ever has previously
6833 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6836 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6837 header. This is something that is recommended in a new Internet Draft, and
6838 is something that is documented as being done by Sendmail. There are two
6839 possible values. For messages generated by the autoreply transport, Exim
6842 Auto-Submitted: auto-replied
6844 whereas for all other generated messages (e.g. bounces) it adds
6846 Auto-Submitted: auto-generated
6848 63. The "personal" condition in filters now includes a test for the
6849 Auto-Submitted: header. If it contains the string "auto-" the message it
6850 not considered personal.
6852 64. Added rcpt_include_affixes as a generic transport option.
6854 65. Added queue_only_override (default true).
6856 66. Added the syslog_duplication option.
6858 67. If what should have been the first header line of a message consisted of
6859 a space followed by a colon, Exim was mis-interpreting it as a header line.
6860 It isn't of course - it is syntactically invalid and should therefore be
6861 treated as the start of the message body. The misbehaviour could have
6862 caused a number of strange effects, including loss of data in subsequent
6863 header lines, and spool format errors.
6865 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6866 client host had authenticated. This control can now be exercised by an ACL
6867 for more flexibility.
6869 69. By default, callouts do not happen when testing with -bh. There is now a
6870 variant, -bhc, which does actually run the callout code, including
6871 consulting and updating the callout cache.
6873 70. Added support for saslauthd authentication, courtesy of Alexander
6876 71. If statvfs() failed on the spool or log directories while checking their
6877 size for availability, Exim confusingly gave the error "space shortage".
6878 Furthermore, in debugging mode it crashed with a floating point exception.
6879 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6880 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6881 serious problem, Exim now writes to the main and panic logs when this
6882 happens, with details of the failure. It then refuses to accept the
6883 incoming message, giving the message "spool directory problem" or "log
6884 directory problem" with a 421 code for SMTP messages.
6886 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6887 0, 1, and 2 exist, because some OS complain for execs without them (see
6888 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6889 descriptors. However, the code omitted to check that the open succeeded,
6890 causing mysterious errors if for some reason the permissions on /dev/null
6891 got screwed. Now Exim writes a message to the main and panic logs, and
6892 bombs out if it can't open /dev/null.
6894 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6895 interact so that it is all more flexible. It is supposed to remain
6896 backwards compatible. Also added extra_local_interfaces.
6898 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6899 to bomb out with an assertion failure - to the client this appears as a
6900 connection drop. This problem occurs in the part of the code that was taken
6901 from the Samba project. Fortunately, the assertion is in a very simple
6902 function, so I have fixed this by reproducing the function inline in the
6903 one place where it is called, and arranging for authentication to fail
6904 instead of killing the process with assert().
6906 75. The SPA client code was not working when the server requested OEM rather
6907 than Unicode encoding.
6909 76. Added code to make require_files with a specific uid setting more usable in
6910 the case where statting the file as root fails - usually a non-root-mounted
6911 NFS file system. When this happens and the failure is EACCES, Exim now
6912 forks a subprocess and does the per-uid checking as the relevant uid.
6914 77. Added process_log_path.
6916 78. If log_file_path was not explicitly set, a setting of check_log_space or
6917 check_log_inodes was ignored.
6919 79. If a space check for the spool or log partitions fails, the incident is now
6920 logged. Of course, in the latter case the data may get lost...
6922 80. Added the %p formatting code to string_format() so that it can be used to
6923 print addresses in debug_print(). Adjusted all the address printing in the
6924 debugging in store.c to use %p rather than %d.
6926 81. There was a concern that a line of code in smtp_in.c could overflow a
6927 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6928 initially expressed, the concern was not well-founded, because trailing
6929 spaces are removed early. However, if the trailing spaces were followed by
6930 a NULL, they did not get removed, so the overflow was possible. Two fixes
6933 (a) I re-wrote the offending code in a cleaner fashion.
6934 (b) If an incoming SMTP command contains a NULL character, it is rejected
6937 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6938 runs initgroups(), so that if the Exim user is in any additional groups,
6939 they will be used during message reception.
6945 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob