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
189 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
190 than -2003 did; needs libidn2 in addition to libidn.
192 JH/02 The path option on a pipe transport is now expanded before use.
194 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
195 Patch provided by "Björn", documentation fix added too.
197 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
198 missing a wire-to-host endian conversion.
200 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
201 close after a BDAT command line could be taken as a following command,
202 giving a synch failure. Fix by only checking for synch immediately
203 before acknowledging the chunk.
205 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
206 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
207 macro. Patches provided by Josh Soref.
209 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
210 Previously we did not; the RFC seems ambiguous and VRFY is not listed
211 by IANA as a service extension. However, John Klensin suggests that we
214 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
215 the dkim code may be unix-mode line endings rather than smtp wire-format
216 CRLF, so prepend a CR to any bare LF.
218 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
219 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
221 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
222 the first were themselves being wrongly included in the feed into dkim
223 processing; with most chunk sizes in use this resulted in an incorrect
224 body hash calculated value.
226 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
227 DKIM signature block, for verification. Although advised against by
228 standards it is specifically not ruled illegal.
230 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
232 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
233 missing a body hash (the bh= tag).
235 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
236 It seems that HAProxy sends the Proxy Protocol information in clear and
237 only then does a TLS startup, so do the same.
239 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
240 TCP connections (such as for Spamd) unless the daemon successfully set
241 Fast Open mode on its listening sockets. This fixes breakage seen on
242 too-old kernels or those not configured for Fast Open, at the cost of
243 requiring both directions being enabled for TFO, and TFO never being used
244 by non-daemon-related Exim processes.
246 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
247 endings, at least on the first header line. Try to canonify any that get
248 past that check, despite the cost.
250 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
251 now limited to an arbitrary five deep, while parsing addresses with the
252 strip_excess_angle_brackets option enabled.
254 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
255 instead leave the unprompted TLS handshake in socket buffer for the
256 TLS library to consume.
258 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
260 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
262 JH/16 Drop variables when they go out of scope. Memory management drops a whole
263 region in one operation, for speed, and this leaves assigned pointers
264 dangling. Add checks run only under the testsuite which checks all
265 variables at a store-reset and panics on a dangling pointer; add code
266 explicitly nulling out all the variables discovered. Fixes one known
267 bug: a transport crash, where a dangling pointer for $sending_ip_address
268 originally assigned in a verify callout, is re-used.
270 PP/06 Drop '.' from @INC in various Perl scripts.
272 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
274 PP/08 Reduce a number of compilation warnings under clang; building with
275 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
276 should be warning-free.
278 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
280 HS/01 Fix portability problems introduced by PP/08 for platforms where
281 realloc(NULL) is not equivalent to malloc() [SunOS et al].
283 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
284 chunk. This allows us to accept broken chunked messages. We need a more
285 general solution here.
287 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
288 already-broken messages in the queue.
290 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
292 JH/19 Fix reference counting bug in routing-generated-address tracking.
298 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
299 supports it and a size is available (ie. the sending peer gave us one).
301 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
302 years of being deprecated). The replacements are the ACLs
303 acl_smtp_mime and acl_not_smtp_mime.
305 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
306 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
307 or even an in-clear connection were permitted. Now, if the host lookup
308 was dnssec and dane was requested then the host is only used if the
309 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
310 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
311 if one fails this test.
312 This means that a poorly-configured remote DNS will make it incommunicado;
313 but it protects against a DNS-interception attack on it.
315 JH/04 Bug 1810: make continued-use of an open smtp transport connection
316 non-noisy when a race steals the message being considered.
318 JH/05 If main configuration option tls_certificate is unset, generate a
319 self-signed certificate for inbound TLS connections.
321 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
322 in rewrites and routers.
324 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
325 and logged a warning sing 4.83; now they are a configuration file error.
327 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
328 (lacking @domain). Apply the same qualification processing as RCPT.
330 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
332 JH/10 Support ${sha256:} applied to a string (as well as the previous
335 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
336 a cutthrough deliver is pending, as we always want to make a connection.
337 This also avoids re-routing the message when later placing the cutthrough
338 connection after a verify cache hit.
339 Do not update it with the verify result either.
341 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
342 when routing results in more than one destination address.
344 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
345 signing (which inhibits the cutthrough capability). Previously only
346 the presence of an option was tested; now an expansion evaluating as
347 empty is permissible (obviously it should depend only on data available
348 when the cutthrough connection is made).
350 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
351 the relevant preceding SMTP command did not note the pipelining mode.
353 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
354 Previously they were not counted.
356 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
357 as one having no matching records. Previously we deferred the message
358 that needed the lookup.
360 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
361 distinguished as "(=".
363 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
364 for missing MX records. Previously it only worked for missing A records.
366 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
368 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
369 after the data-go-ahead and data-ack. Patch from Jason Betts.
371 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
372 even for a "none" policy. Patch from Tony Meyer.
374 JH/22 Fix continued use of a connection for further deliveries. If a port was
375 specified by a router, it must also match for the delivery to be
378 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
379 When one of the recipients of a message was unsuitable for the connection
380 (has no matching addresses), we lost track of needing to mark it
381 deferred. As a result mail would be lost.
383 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
385 JH/25 Decoding ACL controls is now done using a binary search; the source code
386 takes up less space and should be simpler to maintain. Merge the ACL
387 condition decode tables also, with similar effect.
389 JH/26 Fix problem with one_time used on a redirect router which returned the
390 parent address unchanged. A retry would see the parent address marked as
391 delivered, so not attempt the (identical) child. As a result mail would
394 JH/27 Fix a possible security hole, wherein a process operating with the Exim
395 UID can gain a root shell. Credit to http://www.halfdog.net/ for
396 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
399 JH/28 Enable {spool,log} filesystem space and inode checks as default.
400 Main config options check_{log,spool}_{inodes,space} are now
401 100 inodes, 10MB unless set otherwise in the configuration.
403 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
404 Previously it only applied to the main-section connection policy
407 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
409 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
410 by me. Added RFC7919 DH primes as an alternative.
412 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
413 are not in the system include path.
415 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
416 GnuTLS, when a session startup failed (eg because the client disconnected)
417 Exim did stdio operations after fclose. This was exposed by a recent
418 change which nulled out the file handle after the fclose.
420 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
421 signed directly by the cert-signing cert, rather than an intermediate
422 OCSP-signing cert. This is the model used by LetsEncrypt.
424 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
426 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
427 an incoming connection.
429 HS/02 Bug 1802: Do not half-close the connection after sending a request
432 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
433 fallback to "prime256v1".
435 JH/34 SECURITY: Use proper copy of DATA command in error message.
436 Could leak key material. Remotely exploitable. CVE-2016-9963.
442 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
443 and 3.4.4 - once the server is enabled to respond to an OCSP request
444 it does even when not requested, resulting in a stapling non-aware
445 client dropping the TLS connection.
447 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
448 support variable-length bit vectors. No functional change.
450 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
451 The I= interface field on outgoing lines is now after the H= remote
452 host field, same as incoming lines. There is a separate
453 outgoing_interface log selector which allows you to disable the
456 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
457 If not running log_selector +smtp_connection the mainlog would be held
458 open indefinitely after a "too many connections" event, including to a
459 deleted file after a log rotate. Leave the per net connection logging
460 leaving it open for efficiency as that will be quickly detected by the
461 check on the next write.
463 HS/01 Bug 1671: Fix post transport crash.
464 Processing the wait-<transport> messages could crash the delivery
465 process if the message IDs didn't exist for some reason. When
466 using 'split_spool_directory=yes' the construction of the spool
467 file name failed already, exposing the same netto behaviour.
469 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
470 mime_regex ACL conditions.
472 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
473 to DSN fail messages (bounces): remote IP, remote greeting, remote response
474 to HELO, local diagnostic string.
476 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
477 log line to debug. Even when configured with a tls authenticator many
478 client connections are expected to not authenticate in this way, so
479 an authenticate fail is not an error.
481 HS/02 Add the Exim version string to the process info. This way exiwhat
482 gives some more detail about the running daemon.
484 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
485 matter for fast-change records such as DNSBLs.
487 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
488 retry record, even if constant. There may be multiple transports with
489 different interface settings and the retry behaviour needs to be kept
492 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
494 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
496 JH/10 Bug 840: fix log_defer_output option of pipe transport
498 JH/11 Bug 830: use same host for all RCPTS of a message, even under
499 hosts_randomize. This matters a lot when combined with mua_wrapper.
501 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
502 ${quote_pgsql:<string>} operator.
504 JH/13 Bug 1708: avoid misaligned access in cached lookup.
506 JH/14 Change header file name for freeradius-client. Relevant if compiling
507 with Radius support; from the Gentoo tree and checked under Fedora.
509 JH/15 Bug 1712: Introduce $prdr_requested flag variable
511 JH/16 Bug 1714: Permit an empty string as expansion result for transport
512 option transport_filter, meaning no filtering.
514 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
516 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
517 defaults to "*" (all hosts). The variable is now available when not built
518 with TLS, default unset, mainly to enable keeping the testsuite sane.
519 If a server certificate is not supplied (via tls_certificate) an error is
520 logged, and clients will find TLS connections fail on startup. Presumably
521 they will retry in-clear.
522 Packagers of Exim are strongly encouraged to create a server certificate
523 at installation time.
525 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
526 with the $config_file variable.
528 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
529 in transport context, after the attempt, and per-recipient. The latter type
530 is per host attempted. The event data is the error message, and the errno
531 information encodes the lookup type (A vs. MX) used for the (first) host,
532 and the trailing two digits of the smtp 4xx response.
534 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
535 to write to mainlog (or rejectlog, paniclog) in the window between file
536 creation and permissions/ownership being changed. Particularly affects
537 installations where exicyclog is run as root, rather than exim user;
538 result is that the running daemon panics and dies.
540 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
542 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
543 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
544 "pri" and "weight". Note that the previous implicit priority given by the
545 list order is no longer honoured.
547 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
550 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
551 by defining SUPPORT_SOCKS.
553 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
554 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
555 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
556 variables are renamed to proxy_{local,external}_{address,port}.
558 JH/27 Move Internationalisation support from Experimental to mainline, enabled
559 for a build by defining SUPPORT_I18N
561 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
562 of the query string, and make ${quote_redis:} do that quoting.
564 JH/29 Move Events support from Experimental to mainline, enabled by default
565 and removable for a build by defining DISABLE_EVENT.
567 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
569 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
570 cached by the daemon.
572 JH/32 Move Redis support from Experimental to mainline, enabled for a build
573 by defining LOOKUP_REDIS. The libhiredis library is required.
575 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
576 keys are given for lookup.
578 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
579 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
580 only supported when built with TLS support. The PolarSSL SHA routines
581 are still used when the TLS library is too old for convenient support.
583 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
584 openssl_options), for security. OpenSSL forces this from version 1.1.0
585 server-side so match that on older versions.
587 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
588 allocation for $value could be released as the expansion processing
589 concluded, but leaving the global pointer active for it.
591 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
592 and to use the domains and local_parts ACL conditions.
594 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
595 incorrectly not doubled on cutthrough transmission, hence seen as a
596 body-termination at the receiving system - resulting in truncated mails.
597 Commonly the sender saw a TCP-level error, and retransmitted the message
598 via the normal store-and-forward channel. This could result in duplicates
599 received - but deduplicating mailstores were liable to retain only the
600 initial truncated version.
602 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
604 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
606 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
607 we're in there, support oversigning also; bug 1309.
609 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
611 HS/04 Add support for keep_environment and add_environment options.
613 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
614 either intentional arithmetic overflow during PRNG, or testing config-
617 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
618 delivery resulted in actual delivery. Cancel cutthrough before DATA
621 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
622 rejects a recipient: pass the reject to the originator.
624 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
625 Many were false-positives and ignorable, but it's worth fixing the
628 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
629 for the new environment-manipulation done at startup. Move the routines
630 from being local to tls.c to being global via the os.c file.
632 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
633 an extract embedded as result-arg for a map, the first arg for extract
634 is unavailable so we cannot tell if this is a numbered or keyed
635 extraction. Accept either.
641 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
644 JH/02 The smtp transport option "multi_domain" is now expanded.
646 JH/03 The smtp transport now requests PRDR by default, if the server offers
649 JH/04 Certificate name checking on server certificates, when exim is a client,
650 is now done by default. The transport option tls_verify_cert_hostnames
651 can be used to disable this per-host. The build option
652 EXPERIMENTAL_CERTNAMES is withdrawn.
654 JH/05 The value of the tls_verify_certificates smtp transport and main options
655 default to the word "system" to access the system default CA bundle.
656 For GnuTLS, only version 3.0.20 or later.
658 JH/06 Verification of the server certificate for a TLS connection is now tried
659 (but not required) by default. The verification status is now logged by
660 default, for both outbound TLS and client-certificate supplying inbound
663 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
666 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
667 Status Notification (bounce) messages are now MIME format per RFC 3464.
668 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
669 under the control of the dsn_advertise_hosts option, and routers may
670 have a dsn_lasthop option.
672 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
673 default, modifiable by a malware= option. The list separator for
674 the options can now be changed in the usual way. Bug 68.
676 JH/10 The smtp_receive_timeout main option is now expanded before use.
678 JH/11 The incoming_interface log option now also enables logging of the
679 local interface on delivery outgoing connections.
681 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
682 if the interface and destination host and port all match.
684 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
687 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
688 Patch from Andrew Lewis.
690 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
691 now supports optional time-restrictions, weighting, and priority
692 modifiers per server. Patch originally by <rommer@active.by>.
694 JH/16 The spamd_address main option now supports a mixed list of local
695 and remote servers. Remote servers can be IPv6 addresses, and
696 specify a port-range.
698 JH/17 Bug 68: The spamd_address main option now supports an optional
699 timeout value per server.
701 JH/18 Bug 1581: Router and transport options headers_add/remove can
702 now have the list separator specified.
704 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
707 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
710 JH/21 Support for the A6 type of dns record is withdrawn.
712 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
713 rather than the verbs used.
715 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
716 from 255 to 1024 chars.
718 JH/24 Verification callouts now attempt to use TLS by default.
720 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
721 are generic router options now. The defaults didn't change.
723 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
724 Original patch from Alexander Shikoff, worked over by JH.
726 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
727 files if ZCAT_COMMAND is not executable.
729 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
731 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
733 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
734 Normally benign, it bites when the pair was led to by a CNAME;
735 modern usage is to not canonicalize the domain to a CNAME target
736 (and we were inconsistent anyway for A-only vs AAAA+A).
738 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
740 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
741 when evaluating $sender_host_dnssec.
743 JH/31 Check the HELO verification lookup for DNSSEC, adding new
744 $sender_helo_dnssec variable.
746 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
748 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
750 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
752 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
753 documented as working, but never had. Support all but $spam_report.
755 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
756 added for tls authenticator.
758 HS/03 Add perl_taintmode main config option
764 TL/01 When running the test suite, the README says that variables such as
765 no_msglog_check are global and can be placed anywhere in a specific
766 test's script, however it was observed that placement needed to be near
767 the beginning for it to behave that way. Changed the runtest perl
768 script to read through the entire script once to detect and set these
769 variables, reset to the beginning of the script, and then run through
770 the script parsing/test process like normal.
772 TL/02 The BSD's have an arc4random API. One of the functions to induce
773 adding randomness was arc4random_stir(), but it has been removed in
774 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
775 function when detected.
777 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
778 cause callback expansion.
780 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
781 syntax errors in an expansion can be treated as a string instead of
782 logging or causing an error, due to the internal use of bool_lax
783 instead of bool when processing it.
785 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
786 server certificates when making smtp deliveries.
788 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
790 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
792 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
794 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
795 Merged patch from Sebastian Wiedenroth.
797 JH/05 Fix results-pipe from transport process. Several recipients, combined
798 with certificate use, exposed issues where response data items split
799 over buffer boundaries were not parsed properly. This eventually
800 resulted in duplicates being sent. This issue only became common enough
801 to notice due to the introduction of connection certificate information,
802 the item size being so much larger. Found and fixed by Wolfgang Breyha.
804 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
805 size buffer was used, resulting in syntax errors when an expansion
808 JH/07 Add support for directories of certificates when compiled with a GnuTLS
809 version 3.3.6 or later.
811 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
812 is EXPERIMENTAL_EVENT, the main-configuration and transport options
813 both become "event_action", the variables become $event_name, $event_data
814 and $event_defer_errno. There is a new variable $verify_mode, usable in
815 routers, transports and related events. The tls:cert event is now also
816 raised for inbound connections, if the main configuration event_action
819 TL/06 In test suite, disable OCSP for old versions of openssl which contained
820 early OCSP support, but no stapling (appears to be less than 1.0.0).
822 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
823 server certificate names available under the smtp transport option
824 "tls_verify_cert_hostname" now do not permit multi-component wildcard
827 JH/10 Time-related extraction expansions from certificates now use the main
828 option "timezone" setting for output formatting, and are consistent
829 between OpenSSL and GnuTLS compilations. Bug 1541.
831 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
832 encoded parameter in the incoming message. Bug 1558.
834 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
835 include certificate info, eximon was claiming there were spoolfile
838 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
840 JH/14 Log delivery-related information more consistently, using the sequence
841 "H=<name> [<ip>]" wherever possible.
843 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
844 are problematic for Debian distribution, omit them from the release
847 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
849 JH/16 Fix string representation of time values on 64bit time_t architectures.
852 JH/17 Fix a null-indirection in certextract expansions when a nondefault
853 output list separator was used.
858 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
859 checkers that were complaining about end of non-void function with no
862 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
863 This was a regression introduced in 4.83 by another bugfix.
865 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
867 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
868 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
874 TF/01 Correctly close the server side of TLS when forking for delivery.
876 When a message was received over SMTP with TLS, Exim failed to clear up
877 the incoming connection properly after forking off the child process to
878 deliver the message. In some situations the subsequent outgoing
879 delivery connection happened to have the same fd number as the incoming
880 connection previously had. Exim would try to use TLS and fail, logging
881 a "Bad file descriptor" error.
883 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
884 utilities have not been installed.
886 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
887 temporary space as the ACL may create new global variables.
889 TL/01 LDAP support uses per connection or global context settings, depending
890 upon the detected version of the libraries at build time.
892 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
893 to extract and use the src ip:port in logging and expansions as if it
894 were a direct connection from the outside internet. PPv2 support was
895 updated based on HAProxy spec change in May 2014.
897 JH/02 Add ${listextract {number}{list}{success}{fail}}.
899 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
900 Properly escape header and check for NULL return.
902 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
905 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
907 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
908 characters in header names, implemented as a verify condition.
909 Contributed by Michael Fischer v. Mollard.
911 TL/05 Rename SPF condition results err_perm and err_temp to standardized
912 results permerror and temperror. Previous values are deprecated but
913 still accepted. In a future release, err_perm and err_temp will be
914 completely removed, which will be a backward incompatibility if the
915 ACL tests for either of these two old results. Patch contributed by
916 user bes-internal on the mailing list.
918 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
920 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
921 selectors, in both main and reject logs.
923 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
926 JH/07 Add malware type "sock" for talking to simple daemon.
928 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
930 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
931 routers/transports under cutthrough routing.
933 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
934 numbers. Touch up "bool" conditional to keep the same definition.
936 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
938 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
940 JH/12 Expand items in router/transport headers_add or headers_remove lists
941 individually rather than the list as a whole. Bug 1452.
943 Required for reasonable handling of multiple headers_ options when
944 they may be empty; requires that headers_remove items with embedded
945 colons must have them doubled (or the list-separator changed).
947 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
948 view the policy declared in the DMARC record. Currently, $dmarc_status
949 is a combined value of both the record presence and the result of the
952 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
954 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
955 dnslookup router and the smtp transport (applying to the forward
958 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
959 of ldap servers used for a specific lookup. Patch provided by Heiko
962 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
963 New variable $lookup_dnssec_authenticated for observability.
965 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
966 Patch submitted by Lars Timman.
968 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
970 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
971 Requires trusted mode and valid format message id, aborts otherwise.
972 Patch contributed by Heiko Schlichting.
974 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
975 certextract with support for various fields. Bug 1358.
977 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
978 is requested by default, modifiable by smtp transport option
981 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
982 operate on certificate variables to give certificate fingerprints
983 Also new ${sha256:cert_variable}.
985 JH/23 The PRDR feature is moved from being Experimental into the mainline.
987 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
988 Christian Aistleitner.
990 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
992 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
993 file. Patch from Wolfgang Breyha.
995 JH/25 Expand the coverage of the delivery $host and $host_address to
996 client authenticators run in verify callout. Bug 1476.
998 JH/26 Port service names are now accepted for tls_on_connect_ports, to
999 align with daemon_smtp_ports. Bug 72.
1001 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1002 support and error reporting did not work properly.
1004 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1005 and is readable. Patch from Andrew Colin Kissa.
1007 TL/14 Enhance documentation of ${run expansion and how it parses the
1008 commandline after expansion, particularly in the case when an
1009 unquoted variable expansion results in an empty value.
1011 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1013 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1014 by renaming away from it. Addresses GH issue 15, reported by
1017 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1019 TL/15 SECURITY: prevent double expansion in math comparison functions
1020 (can expand unsanitized data). Not remotely exploitable.
1027 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1029 PP/02 Make -n do something, by making it not do something.
1030 When combined with -bP, the name of an option is not output.
1032 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1035 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1036 $sender_host_name and config options to manage this, and basic check
1039 PP/05 DSCP support for outbound connections and control modifier for inbound.
1041 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1042 (Only plugin which currently uses this is kerberos4, which nobody should
1043 be using, but we should make it available and other future plugins might
1044 conceivably use it, even though it would break NAT; stuff *should* be
1045 using channel bindings instead).
1047 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1048 name; added for Sendmail compatibility; requires admin caller.
1049 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1050 just ignore it); requires trusted caller.
1051 Also parse but ignore: -Ac -Am -X<logfile>
1054 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1056 TL/02 Add +smtp_confirmation as a default logging option.
1058 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1059 Patch by Magnus Holmgren from 2007-02-20.
1061 TL/04 Bugzilla 1281 - Spec typo.
1062 Bugzilla 1283 - Spec typo.
1063 Bugzilla 1290 - Spec grammar fixes.
1065 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1067 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1069 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1070 mailing list by by Dmitry Isaikin.
1072 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1074 JH/02 Support "G" suffix to numbers in ${if comparisons.
1076 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1078 NM/01 Bugzilla 1197 - Spec typo
1079 Bugzilla 1196 - Spec examples corrections
1081 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1083 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1084 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1087 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1088 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1090 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1091 "acl {{name}{arg}...}", and optional args on acl condition
1094 JH/05 Permit multiple router/transport headers_add/remove lines.
1096 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1098 JH/07 Avoid using a waiting database for a single-message-only transport.
1099 Performance patch from Paul Fisher. Bugzilla 1262.
1101 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1104 JH/09 Add $headers_added variable, with content from use of ACL modifier
1105 add_header (but not yet added to the message). Bugzilla 199.
1107 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1108 Pulled from Bugzilla 817 by Wolfgang Breyha.
1110 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1112 (nb: this is the same fix as in Exim 4.80.1)
1114 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1117 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1120 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1122 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1123 advertise SMTP AUTH mechanism to us, instead of a generic
1124 protocol violation error. Also, make Exim more robust to bad
1125 data from the Dovecot auth socket.
1127 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1129 When a queue runner is handling a message, Exim first routes the
1130 recipient addresses, during which it prunes them based on the retry
1131 hints database. After that it attempts to deliver the message to
1132 any remaining recipients. It then updates the hints database using
1135 So if a recipient address works intermittently, it can get repeatedly
1136 deferred at routing time. The retry hints record remains fresh so the
1137 address never reaches the final cutoff time.
1139 This is a fairly common occurrence when a user is bumping up against
1140 their storage quota. Exim had some logic in its local delivery code
1141 to deal with this. However it did not apply to per-recipient defers
1142 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1144 This change adds a proper retry rule check during routing so that the
1145 final cutoff time is checked against the message's age. We only do
1146 this check if there is an address retry record and there is not a
1147 domain retry record; this implies that previous attempts to handle
1148 the address had the retry_use_local_parts option turned on. We use
1149 this as an approximation for the destination being like a local
1150 delivery, as in LMTP.
1152 I suspect this new check makes the old local delivery cutoff check
1153 redundant, but I have not verified this so I left the code in place.
1155 TF/02 Correct gecos expansion when From: is a prefix of the username.
1157 Test 0254 submits a message to Exim with the header
1161 When I ran the test suite under the user fanf2, Exim expanded
1162 the header to contain my full name, whereas it should have added
1163 a Resent-Sender: header. It erroneously treats any prefix of the
1164 username as equal to the username.
1166 This change corrects that bug.
1168 GF/01 DCC debug and logging tidyup
1169 Error conditions log to paniclog rather than rejectlog.
1170 Debug lines prefixed by "DCC: " to remove any ambiguity.
1172 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1174 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1175 Bug spotted by Jeremy Harris; was flawed since initial commit.
1176 Would have resulted in OCSP responses post-SNI triggering an Exim
1177 NULL dereference and crash.
1179 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1181 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1182 Bug detection, analysis and fix by Samuel Thibault.
1183 Bugzilla 1331, Debian bug #698092.
1185 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1187 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1188 Server implementation by Todd Lyons, client by JH.
1189 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1190 config variable "prdr_enable" controls whether the server
1191 advertises the facility. If the client requests PRDR a new
1192 acl_data_smtp_prdr ACL is called once for each recipient, after
1193 the body content is received and before the acl_smtp_data ACL.
1194 The client is controlled by both of: a hosts_try_prdr option
1195 on the smtp transport, and the server advertisement.
1196 Default client logging of deliveries and rejections involving
1197 PRDR are flagged with the string "PRDR".
1199 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1200 fclose(). Diagnosis by Todd Lyons.
1202 PP/17 Update configure.default to handle IPv6 localhost better.
1203 Patch by Alain Williams (plus minor tweaks).
1206 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1207 This is now consistent with GnuTLS, and is now documented: the
1208 previous undocumented portable approach to treating the option as
1209 unset was to force an expansion failure. That still works, and
1210 an empty string is now equivalent.
1212 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1213 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1214 not performing validation itself.
1216 PP/20 Added force_command boolean option to pipe transport.
1217 Patch from Nick Koston, of cPanel Inc.
1219 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1222 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1224 PP/21 Fix eximon continuous updating with timestamped log-files.
1225 Broken in a format-string cleanup in 4.80, missed when I repaired the
1226 other false fix of the same issue.
1227 Report and fix from Heiko Schlichting.
1230 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1231 Report from Prashanth Katuri.
1233 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1234 It's SecureTransport, so affects any MacOS clients which use the
1235 system-integrated TLS libraries, including email clients.
1237 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1238 using a MIME ACL for non-SMTP local injection.
1239 Report and assistance in diagnosis by Warren Baker.
1241 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1243 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1245 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1246 last id that failed so it may be referenced in subsequent ACL's.
1248 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1251 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1252 ldap library initialization, allowing self-signed CA's to be
1253 used. Also properly sets require_cert option later in code by
1254 using NULL (global ldap config) instead of ldap handle (per
1255 session). Bug diagnosis and testing by alxgomz.
1257 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1258 the src/util/ subdirectory.
1260 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1261 renamed to Transport Post Delivery Action by Jeremy Harris, as
1264 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1265 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1266 redis_servers = needs to be configured which will be used by the redis
1267 lookup. Patch from Warren Baker, of The Packet Hub.
1269 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1271 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1272 hostname or reverse DNS when processing a host list. Used suggestions
1273 from multiple comments on this bug.
1275 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1277 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1278 Missed a few lines, added it to make the runtest require no keyboard
1281 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1282 contains upper case chars. Make router use caseful_local_part.
1284 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1285 support when GnuTLS has been built with p11-kit.
1291 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1293 This, or similar/improved, will also be change PP/11 of 4.82.
1299 PP/01 Handle short writes when writing local log-files.
1300 In practice, only affects FreeBSD (8 onwards).
1301 Bugzilla 1053, with thanks to Dmitry Isaikin.
1303 NM/01 Bugzilla 949 - Documentation tweak
1305 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1308 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1310 PP/02 Implemented gsasl authenticator.
1312 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1314 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1315 `pkg-config foo` for cflags/libs.
1317 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1318 with rest of GSASL and with heimdal_gssapi.
1320 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1321 `pkg-config foo` for cflags/libs for the TLS implementation.
1323 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1324 properties get this fed in as external SSF. A number of robustness
1325 and debugging improvements to the cyrus_sasl authenticator.
1327 PP/08 cyrus_sasl server now expands the server_realm option.
1329 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1330 Patch by Jeremy Harris.
1332 PP/10 Added dbmjz lookup type.
1334 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1336 PP/12 MAIL args handles TAB as well as SP, for better interop with
1337 non-compliant senders.
1338 Analysis and variant patch by Todd Lyons.
1340 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1341 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1342 Patch from Dirk Mueller <dmueller@suse.com>
1344 PP/13 tls_peerdn now print-escaped for spool files.
1345 Observed some $tls_peerdn in wild which contained \n, which resulted
1346 in spool file corruption.
1348 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1349 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1350 or write after TLS renegotiation, which otherwise led to messages
1353 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1354 as a tracking header (ie: a signed header comes before the signature).
1355 Patch from Wolfgang Breyha.
1357 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1358 comma-sep list; embedded commas doubled.
1360 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1362 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1364 Report and patch from Dmitry Banschikov.
1366 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1367 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1368 failures. We appear to now support TLS1.1+ with Exim.
1370 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1371 lets Exim select keys and certificates based upon TLS SNI from client.
1372 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1373 before an outbound SMTP session. New log_selector, +tls_sni.
1375 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1376 NULL dereference. Report and patch from Alun Jones.
1378 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1379 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1380 Not seeing resolver debug output on NetBSD, but suspect this is a
1381 resolver implementation change.
1383 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1384 Left warnings. Added "eximon gdb" invocation mode.
1386 PP/21 Defaulting "accept_8bitmime" to true, not false.
1388 PP/22 Added -bw for inetd wait mode support.
1390 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1391 locate the relevant includes and libraries. Made this the default.
1393 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1394 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1396 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1397 This may cause build issues on older platforms.
1399 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1400 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1401 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1402 Added SNI support via GnuTLS too.
1403 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1405 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1407 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1408 Applied second patch from Janne, implementing suggestion to default
1409 multiple-strings-in-record handling to match SPF spec.
1411 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1413 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1414 read-only, out of scope).
1415 Patch from Wolfgang Breyha, report from Stuart Northfield.
1417 PP/29 Fix three issues highlighted by clang analyser static analysis.
1418 Only crash-plausible issue would require the Cambridge-specific
1419 iplookup router and a misconfiguration.
1420 Report from Marcin Mirosław.
1422 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1424 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1425 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1426 As part of this, removing so much warning spew let me fix some minor
1427 real issues in debug logging.
1429 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1430 assignment on my part. Fixed.
1432 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1433 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1434 Janne Snabb (who went above and beyond: thank you).
1436 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1437 string otherwise requires a connection and a bunch more work and it's
1438 relatively easy to get wrong. Should also expose TLS library linkage
1441 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1442 64-bit ${eval} (JH/03).
1444 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1445 GNU libc to support some of the 64-bit stuff, should not lead to
1446 conflicts. Defined before os.h is pulled in, so if a given platform
1447 needs to override this, it can.
1449 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1450 protection layer was required, which is not implemented.
1451 Bugzilla 1254, patch from Wolfgang Breyha.
1453 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1454 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1455 tls_dhparam take prime identifiers. Also unbreak combination of
1456 OpenSSL+DH_params+TLSSNI.
1458 PP/39 Disable SSLv2 by default in OpenSSL support.
1464 PP/01 Solaris build fix for Oracle's LDAP libraries.
1465 Bugzilla 1109, patch from Stephen Usher.
1467 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1469 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1472 TF/02 Fix a couple more cases where we did not log the error message
1473 when unlink() failed. See also change 4.74-TF/03.
1475 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1476 lock up or crash if it happened to be inside a call to libc when it
1477 got a SIGUSR1 from exiwhat.
1479 The SIGUSR1 handler appends the current process status to the process
1480 log which is later printed by exiwhat. It used to use the general
1481 purpose logging code to do this, but several functions it calls are
1482 not safe for signals.
1484 The new output code in the SIGUSR1 handler is specific to the process
1485 log, and simple enough that it's easy to inspect for signal safety.
1486 Removing some special cases also simplifies the general logging code.
1487 Removing the spurious timestamps from the process log simplifies
1490 TF/04 Improved ratelimit ACL condition.
1492 The /noupdate option has been deprecated in favour of /readonly which
1493 has clearer semantics. The /leaky, /strict, and /readonly update modes
1494 are mutually exclusive. The update mode is no longer included in the
1495 database key; it just determines when the database is updated. (This
1496 means that when you upgrade Exim will forget old rate measurements.)
1498 Exim now checks that the per_* options are used with an update mode that
1499 makes sense for the current ACL. For example, when Exim is processing a
1500 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1501 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1502 must specify per_mail/readonly. If you omit the update mode it defaults to
1503 /leaky where that makes sense (as before) or /readonly where required.
1505 The /noupdate option is now undocumented but still supported for
1506 backwards compatibility. It is equivalent to /readonly except that in
1507 ACLs where /readonly is required you may specify /leaky/noupdate or
1508 /strict/noupdate which are treated the same as /readonly.
1510 A useful new feature is the /count= option. This is a generalization
1511 of the per_byte option, so that you can measure the throughput of other
1512 aggregate values. For example, the per_byte option is now equivalent
1513 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1515 The per_rcpt option has been generalized using the /count= mechanism
1516 (though it's more complicated than the per_byte equivalence). When it is
1517 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1518 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1519 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1520 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1521 Note that using per_rcpt with a non-readonly update mode in more than
1522 one ACL will cause the recipients to be double-counted. (The per_mail
1523 and per_byte options don't have this problem.)
1525 The handling of very low rates has changed slightly. If the computed rate
1526 is less than the event's count (usually one) then this event is the first
1527 after a long gap. In this case the rate is set to the same as this event's
1528 count, so that the first message of a spam run is counted properly.
1530 The major new feature is a mechanism for counting the rate of unique
1531 events. The new per_addr option counts the number of different
1532 recipients that someone has sent messages to in the last time period. It
1533 behaves like per_rcpt if all the recipient addresses are different, but
1534 duplicate recipient addresses do not increase the measured rate. Like
1535 the /count= option this is a general mechanism, so the per_addr option
1536 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1537 example, measure the rate that a client uses different sender addresses
1538 with the options per_mail/unique=$sender_address. There are further
1539 details in the main documentation.
1541 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1543 TF/06 Removed a few PCRE remnants.
1545 TF/07 Automatically extract Exim's version number from tags in the git
1546 repository when doing development or release builds.
1548 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1549 Bugzilla 879. Patch from Paul Fisher.
1551 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1552 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1555 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1557 PP/05 Variable $av_failed, true if the AV scanner deferred.
1558 Bugzilla 1078. Patch from John Horne.
1560 PP/06 Stop make process more reliably on build failure.
1561 Bugzilla 1087. Patch from Heiko Schlittermann.
1563 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1564 Bugzilla 1089. Patch from Heiko Schlittermann.
1566 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1567 Bugzilla 1131. Patch from Holger Weiß.
1569 PP/09 Handle IPv6 addresses with SPF.
1570 Bugzilla 860. Patch from Wolfgang Breyha.
1572 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1574 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1577 PP/11 match_* no longer expand right-hand-side by default.
1578 New compile-time build option, EXPAND_LISTMATCH_RHS.
1579 New expansion conditions, "inlist", "inlisti".
1581 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1583 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1585 PP/14 fix log_write() format string regression from TF/03.
1586 Bugzilla 1152. Patch from Dmitry Isaikin.
1592 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1594 PP/02 Harmonised TLS library version reporting; only show if debugging.
1595 Layout now matches that introduced for other libraries in 4.74 PP/03.
1597 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1599 PP/04 New "dns_use_edns0" global option.
1601 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1604 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1605 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1607 TK/01 Updated PolarSSL code to 0.14.2.
1608 Bugzilla 1097. Patch from Andreas Metzler.
1610 PP/07 Catch divide-by-zero in ${eval:...}.
1611 Fixes bugzilla 1102.
1613 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1616 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1617 format-string attack -- SECURITY: remote arbitrary code execution.
1619 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1620 time unintentionally subject to list matching rules, letting the header
1621 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1622 arbitrary string expansion). This allowed for information disclosure.
1624 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1625 INT_MIN/-1 -- value coerced to INT_MAX.
1631 NM/01 Workaround for PCRE version dependency in version reporting
1634 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1635 This fixes portability to compilers other than gcc, notably
1636 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1638 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1639 makefiles for portability to HP-UX and POSIX correctness.
1641 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1642 Also via indented variable definition in the Makefile.
1643 (Debugging by Oliver Heesakkers).
1645 PP/02 Restore caching of spamd results with expanded spamd_address.
1646 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1648 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1649 Improves build reliability. Fix from: Frank Elsner
1651 NM/02 Fix wide character breakage in the rfc2047 coding
1652 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1654 NM/03 Allow underscore in dnslist lookups
1655 Fixes bug 1026. Patch from Graeme Fowler
1657 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1658 Code patches from Adam Ciarcinski of NetBSD.
1660 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1663 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1664 is logged, to avoid truncation. Patch from John Horne.
1666 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1667 Patch from Jakob Hirsch.
1669 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1670 SQL string expansion failure details.
1671 Patch from Andrey Oktyabrski.
1673 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1674 Patch from Simon Arlott.
1676 PP/09 New lookups functionality failed to compile on old gcc which rejects
1677 extern declarations in function scope.
1678 Patch from Oliver Fleischmann
1680 PP/10 Use sig_atomic_t for flags set from signal handlers.
1681 Check getgroups() return and improve debugging.
1682 Fixed developed for diagnosis in bug 927 (which turned out to be
1685 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1686 Patch from Mark Zealey.
1688 PP/12 Bugzilla 1056: Improved spamd server selection.
1689 Patch from Mark Zealey.
1691 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1692 Based on patch from Heiko Schlittermann.
1694 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1695 Patch from Uwe Doering, sign-off by Michael Haardt.
1697 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1698 variable declaration deep within a block. Bug and patch from
1701 PP/15 lookups-Makefile IRIX compatibility coercion.
1703 PP/16 Make DISABLE_DKIM build knob functional.
1705 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1706 Patch by Simon Arlott
1708 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1709 variable argument macros. Our copy now differs from upstream.
1715 TF/01 Failure to get a lock on a hints database can have serious
1716 consequences so log it to the panic log.
1718 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1719 controlled using the smtp_confirmation log selector.
1721 TF/03 Include the error message when we fail to unlink a spool file.
1723 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1724 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1725 for maintaining out-of-tree patches for some time.
1727 PP/01 Bugzilla 139: Documentation and portability issues.
1728 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1729 Handle per-OS dynamic-module compilation flags.
1731 PP/02 Let /dev/null have normal permissions.
1732 The 4.73 fixes were a little too stringent and complained about the
1733 permissions on /dev/null. Exempt it from some checks.
1734 Reported by Andreas M. Kirchwitz.
1736 PP/03 Report version information for many libraries, including
1737 Exim version information for dynamically loaded libraries. Created
1738 version.h, now support a version extension string for distributors
1739 who patch heavily. Dynamic module ABI change.
1741 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1742 privilege escalation vulnerability whereby the Exim run-time user
1743 can cause root to append content of the attacker's choosing to
1746 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1749 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1750 If dropping privileges for untrusted macros, we disabled normal logging
1751 on the basis that it would fail; for the Exim run-time user, this is not
1752 the case, and it resulted in successful deliveries going unlogged.
1753 Fixed. Reported by Andreas Metzler.
1759 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1760 only prepend for the Resent-* case. Fixes regression introduced in
1761 Exim 4.70 by NM/22 for Bugzilla 607.
1763 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1764 increasing numbers of administrators be bitten by this.
1766 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1768 PP/03 Bugzilla 994: added openssl_options main configuration option.
1770 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1772 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1774 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1776 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1777 without a peer certificate, leading to a segfault because of an
1778 assumption that peers always have certificates. Be a little more
1779 paranoid. Problem reported by Martin Tscholak.
1781 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1782 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1783 NB: ClamAV planning to remove STREAM in "middle of 2010".
1784 CL also introduces -bmalware, various -d+acl logging additions and
1785 more caution in buffer sizes.
1787 PP/09 Implemented reverse_ip expansion operator.
1789 PP/10 Bugzilla 937: provide a "debug" ACL control.
1791 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1793 PP/12 Bugzilla 973: Implement --version.
1795 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1797 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1799 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1801 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1802 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1803 ignore trailing whitespace.
1805 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1807 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1810 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1811 Notification from Dr Andrew Aitchison.
1813 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1814 ExtendedDetectionInfo response format.
1815 Notification from John Horne.
1817 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1820 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1821 XSL and documented dependency on system catalogs, with examples of how
1824 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1827 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1828 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1829 configuration file which is writeable by the Exim user or group.
1831 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1832 of configuration files to cover files specified with the -C option if
1833 they are going to be used with root privileges, not just the default
1836 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1837 option (effectively making it always true).
1839 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1840 files to be used while preserving root privileges.
1842 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1843 that rogue child processes cannot use them.
1845 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1846 run-time user, instead of root.
1848 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1849 Exim run-time user without dropping privileges.
1851 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1852 result string, instead of calling string_vformat() twice with the same
1855 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1856 for other users. Others should always drop root privileges if they use
1857 -C on the command line, even for a whitelisted configure file.
1859 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1861 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1867 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1868 $data_path, and $header_path variables; fixed documentation bugs and
1871 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1872 exipick to access non-standard spools, including the "frozen" queue
1875 NM/01 Bugzilla 965: Support mysql stored procedures.
1876 Patch from Alain Williams
1878 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1880 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1881 Patch from Andreas Metzler
1883 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1884 Patch from Kirill Miazine
1886 NM/05 Bugzilla 671: Added umask to procmail example.
1888 JJ/03 installed exipick 20100323.0, fixing doc bug
1890 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1891 directory. Notification and patch from Dan Rosenberg.
1893 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1895 TK/02 Improve log output when DKIM signing operation fails.
1897 MH/01 Treat the transport option dkim_domain as a colon separated
1898 list, not as a single string, and sign the message with each element,
1899 omitting multiple occurences of the same signer.
1901 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1902 Bugzilla 985, 986. Patch by Simon Arlott
1904 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1905 Patch by Simon Arlott
1907 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1908 MBX locking. Notification from Dan Rosenberg.
1914 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1916 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1918 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1920 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1922 NM/04 Bugzilla 915: Flag broken perl installation during build.
1928 TK/01 Added patch by Johannes Berg that expands the main option
1929 "spamd_address" if it starts with a dollar sign.
1931 TK/02 Write list of recipients to X-Envelope-Sender header when building
1932 the mbox-format spool file for content scanning (suggested by Jakob
1935 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1936 (http://www.dcc-servers.net/) support via dccifd. Activated by
1937 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1939 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1940 by Mark Daniel Reidel <mr@df.eu>.
1942 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1943 When building exim an external PCRE library is now needed -
1944 PCRE is a system library on the majority of modern systems.
1945 See entry on PCRE_LIBS in EDITME file.
1947 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1948 conversation. Added nologin parameter to request.
1949 Patch contributed by Kirill Miazine.
1951 TF/01 Do not log submission mode rewrites if they do not change the address.
1953 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1955 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1956 log files in place. Contributed by Roberto Lima.
1958 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1960 TF/03 Bugzilla 615: When checking the local_parts router precondition
1961 after a local_part_suffix or local_part_prefix option, Exim now
1962 does not use the address's named list lookup cache, since this
1963 contains cached lookups for the whole local part.
1965 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1966 Robert Millan. Documentation is in experimental-spec.txt.
1968 TF/04 Bugzilla 668: Fix parallel build (make -j).
1970 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1972 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1973 Patch provided by Jan Srzednicki.
1975 TF/05 Leading white space used to be stripped from $spam_report which
1976 wrecked the formatting. Now it is preserved.
1978 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1979 that they are available at delivery time.
1981 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1983 TF/08 TLS error reporting now respects the incoming_interface and
1984 incoming_port log selectors.
1986 TF/09 Produce a more useful error message if an SMTP transport's hosts
1987 setting expands to an empty string.
1989 NM/06 Bugzilla 744: EXPN did not work under TLS.
1990 Patch provided by Phil Pennock.
1992 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1993 Patch provided by Richard Godbee.
1995 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1996 acl_smtp_notquit, added index entry.
1998 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1999 Patch provided by Eugene Bujak.
2001 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2002 accept(). Patch provided by Maxim Dounin.
2004 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2005 Patch provided by Phil Pennock.
2007 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2009 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2010 Patch provided by Brad "anomie" Jorsch.
2012 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2013 Patch provided by Dean Brooks.
2015 TK/05 Add native DKIM support (does not depend on external libraries).
2017 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2018 Patch provided by Graeme Fowler.
2020 NM/16 Bugzilla 851: Documentation example syntax fix.
2022 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2024 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2027 NM/19 Bugzilla 745: TLS version reporting.
2028 Patch provided by Phil Pennock.
2030 NM/20 Bugzilla 167: bool: condition support.
2031 Patch provided by Phil Pennock.
2033 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2034 clients. Patch provided by Phil Pennock.
2036 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2037 Patch provided by Brad "anomie" Jorsch.
2039 NM/23 Bugzilla 687: Fix misparses in eximstats.
2040 Patch provided by Heiko Schlittermann.
2042 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2043 Patch provided by Heiko Schlittermann.
2045 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2046 plus update to original patch.
2048 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2050 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2051 Patch provided by David Brownlee.
2053 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2055 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2057 NM/30 Bugzilla 888: TLS documentation bugfixes.
2059 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2061 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2062 Unlike the original bugzilla I have changed all shell scripts in src tree.
2064 NM/33 Bugzilla 898: Transport filter timeout fix.
2065 Patch by Todd Rinaldo.
2067 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2068 Patch by Serge Demonchaux.
2070 NM/35 Bugzilla 39: Base64 decode bug fixes.
2071 Patch by Jakob Hirsch.
2073 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2075 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2077 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2079 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2085 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2086 ALT-N's libdkim that I have put here:
2087 http://duncanthrax.net/exim-experimental/
2089 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2090 were called 'true' and it seems that C99 defines that as a reserved
2091 keyword to be used with 'bool' variable types. That means you could
2092 not include C99-style headers which use bools without triggering
2093 build errors in sieve.c.
2095 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2096 as mailq or other aliases. Changed the --help handling significantly
2097 to do whats expected. exim_usage() emits usage/help information.
2099 SC/01 Added the -bylocaldomain option to eximstats.
2101 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2103 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2105 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2111 PH/01 Another patch from the Sieve maintainer.
2113 PH/02 When an IPv6 address is converted to a string for single-key lookup
2114 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2115 dots are used instead of colons so that keys in lsearch files need not
2116 contain colons. This was done some time before quoting was made available
2117 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2118 (notated using the quote facility) so as to distinguish them from IPv4
2119 keys. This meant that lookups for IP addresses in host lists did not work
2120 for iplsearch lookups.
2122 This has been fixed by arranging for IPv6 addresses to be expressed with
2123 colons if the lookup type is iplsearch. This is not incompatible, because
2124 previously such lookups could never work.
2126 The situation is now rather anomalous, since one *can* have colons in
2127 ordinary lsearch keys. However, making the change in all cases is
2128 incompatible and would probably break a number of configurations.
2130 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2133 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2134 conversion specification without a maximum field width, thereby enabling
2135 a rogue spamd server to cause a buffer overflow. While nobody in their
2136 right mind would setup Exim to query an untrusted spamd server, an
2137 attacker that gains access to a server running spamd could potentially
2138 exploit this vulnerability to run arbitrary code as the Exim user.
2140 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2141 $primary_hostname instead of what libspf2 thinks the hosts name is.
2143 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2144 a directory entry by the name of the lookup key. Previously, if a
2145 symlink pointed to a non-existing file or a file in a directory that
2146 Exim lacked permissions to read, a lookup for a key matching that
2147 symlink would fail. Now it is enough that a matching directory entry
2148 exists, symlink or not. (Bugzilla 503.)
2150 PH/03 The body_linecount and body_zerocount variables are now exported in the
2153 PH/04 Added the $dnslist_matched variable.
2155 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2156 This means they are set thereafter only if the connection becomes
2159 PH/06 Added the client_condition to authenticators so that some can be skipped
2160 by clients under certain conditions.
2162 PH/07 The error message for a badly-placed control=no_multiline_responses left
2163 "_responses" off the end of the name.
2165 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2167 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2168 (without spaces) instead of just copying the configuration text.
2170 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2172 PH/11 Added $max_received_linelength.
2174 PH/12 Added +ignore_defer and +include_defer to host lists.
2176 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2177 way in which PCRE > 7.0 is built.
2179 PH/14 Implemented queue_only_load_latch.
2181 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2182 MAIL command. The effect was to mangle the value on 64-bit systems.
2184 PH/16 Another patch from the Sieve maintainer.
2186 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2188 PH/18 If a system quota error occurred while trying to create the file for
2189 a maildir delivery, the message "Mailbox is full" was not appended to the
2190 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2191 only to a quota excession during the actual writing of the file.
2193 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2194 characters?) which causes problems in log lines. The DN values are now
2195 passed through string_printing() before being added to log lines.
2197 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2198 and InterBase are left for another time.)
2200 PH/21 Added message_body_newlines option.
2202 PH/22 Guard against possible overflow in moan_check_errorcopy().
2204 PH/23 POSIX allows open() to be a macro; guard against that.
2206 PH/24 If the recipient of an error message contained an @ in the local part
2207 (suitably quoted, of course), incorrect values were put in $domain and
2208 $local_part during the evaluation of errors_copy.
2214 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2215 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2218 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2219 issue a MAIL command.
2221 PH/02 In an ACL statement such as
2223 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2225 if a client was not listed at all, or was listed with a value other than
2226 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2227 the condition was not true (as it should be), so access was not denied.
2228 The bug was that the ! inversion was incorrectly passed on to the second
2229 item. This has been fixed.
2231 PH/03 Added additional dnslists conditions == and =& which are different from
2232 = and & when the dns lookup returns more than one IP address.
2234 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2235 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2237 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2238 FSYNC, which compiles an option called disable_fsync that allows for
2239 bypassing fsync(). The documentation is heavily laced with warnings.
2241 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2243 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2244 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2245 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2246 including adding "make clean"; (3) Added -fPIC when compiling the test
2247 dynamically loaded module, to get rid of a warning.
2249 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2250 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2251 = 0s. The bug is otherwise harmless.
2253 PH/07 There was a bug in the dovecot authenticator such that the value of
2254 $auth1 could be overwritten, and so not correctly preserved, after a
2255 successful authentication. This usually meant that the value preserved by
2256 the server_setid option was incorrect.
2258 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2260 PH/09 Installed PCRE release 7.0.
2262 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2263 run for batched SMTP input. It is now run at the start of every message
2264 in the batch. While fixing this I discovered that the process information
2265 (output by running exiwhat) was not always getting set for -bs and -bS
2266 input. This is fixed, and it now also says "batched" for BSMTP.
2268 PH/11 Added control=no_pipelining.
2270 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2271 patch, slightly modified), and move the expansion of helo_data till after
2272 the connection is made in the smtp transport (so it can use these
2275 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2277 PH/14 Added log_selector = +pid.
2279 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2281 PH/16 Add ${if forany and ${if forall.
2283 PH/17 Added dsn_from option to vary the From: line in DSNs.
2285 PH/18 Flush SMTP output before performing a callout, unless control =
2286 no_callout_flush is set.
2288 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2289 was true (the default) a successful delivery failed to delete the retry
2290 item, thus causing premature timeout of the address. The bug is now
2293 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2295 PH/21 Long custom messages for fakedefer and fakereject are now split up
2296 into multiline responses in the same way that messages for "deny" and
2297 other ACL rejections are.
2299 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2300 with slight modification.
2302 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2303 draft, changing the syntax and factoring some duplicate code".
2305 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2306 for deliveries of the second and subsequent messages over the same SMTP
2309 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2310 ${reduce, with only minor "tidies".
2312 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2314 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2315 expansion side effects.
2317 PH/27 When a message times out after an over-quota error from an Exim-imposed
2318 quota, the bounce message says "mailbox is full". This message was not
2319 being given when it was a system quota that was exceeded. It now should
2322 MH/03 Made $recipients available in local_scan(). local_scan() already has
2323 better access to the recipient list through recipients_list[], but
2324 $recipients can be useful in postmaster-provided expansion strings.
2326 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2327 in the case of a MAIL command with additional options following the
2328 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2329 were accidentally chopped off.
2331 PH/29 SMTP synchronization checks are implemented when a command is read -
2332 there is a check that no more input is waiting when there shouldn't be
2333 any. However, for some commands, a delay in an ACL can mean that it is
2334 some time before the response is written. In this time, more input might
2335 arrive, invalidly. So now there are extra checks after an ACL has run for
2336 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2337 pipelining has not been advertised.
2339 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2341 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2342 returned in the SMTP response when smtp_return_error_details was false.
2343 This has been fixed.
2345 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2346 instead of the C I/O that was originally supplied, because problems were
2347 reported on Solaris.
2349 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2350 Exim which did not show up earlier: it was assuming that a call to
2351 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2352 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2353 was a macro that became an assignment, so it seemed to work. This has
2354 changed to a proper function call with a void return, hence the compile
2355 error. Exim's code has been fixed.
2357 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2360 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2362 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2364 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2365 because it turns out to be too big or there is a timeout) there is no
2366 "Completed" line in the log. When some messages of this type were
2367 selected by exigrep, they were listed as "not completed". Others were
2368 picked up by some special patterns. I have improved the selection
2369 criteria to be more general.
2371 PH/38 The host_find_failed option in the manualroute router can now be set
2372 to "ignore", to completely ignore a host whose IP address cannot be
2373 found. If all hosts are ignored, the behaviour is controlled by the new
2374 host_all_ignored option.
2376 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2377 homing or because of multiple MX records with /mx) generated more than
2378 one IP address, and the following item turned out to be the local host,
2379 all the secondary addresses of the first item were incorrectly removed
2380 from the list, along with the local host and any following hosts (which
2381 is what is supposed to happen).
2383 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2384 whoever called Exim into the -H file. In the case of the daemon it was
2385 behaving confusingly. When first started, it used values for whoever
2386 started the daemon, but after a SIGHUP it used the Exim user (because it
2387 calls itself on a restart). I have changed the code so that it now always
2390 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2391 message are rejected with the same error (e.g. no authentication or bad
2392 sender address), and a DATA command is nevertheless sent (as can happen
2393 with PIPELINING or a stupid MUA), the error message that was given to the
2394 RCPT commands is included in the rejection of the DATA command. This is
2395 intended to be helpful for MUAs that show only the final error to their
2398 PH/42 Another patch from the Sieve maintainer.
2400 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2401 Eximstats - Fixed some broken HTML links and added missing column headers
2403 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2404 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2406 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2408 PH/43 Yet another patch from the Sieve maintainer.
2410 PH/44 I found a way to check for a TCP/IP connection going away before sending
2411 the response to the final '.' that terminates a message, but only in the
2412 case where the client has not sent further data following the '.'
2413 (unfortunately, this is allowed). However, in many cases there won't be
2414 any further data because there won't be any more messages to send. A call
2415 to select() can be used: if it shows that the input is "ready", there is
2416 either input waiting, or the socket has been closed. An attempt to read
2417 the next input character can distinguish the two cases. Previously, Exim
2418 would have sent an OK response which the client would never have see.
2419 This could lead to message repetition. This fix should cure that, at
2420 least in a lot of common cases.
2422 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2423 advertised in response to EHLO.
2429 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2430 fixed by 4.65/MH/01 (is this a record?) are fixed:
2432 (i) An empty string was always treated as zero by the numeric comparison
2433 operators. This behaviour has been restored.
2435 (ii) It is documented that the numeric comparison operators always treat
2436 their arguments as decimal numbers. This was broken in that numbers
2437 starting with 0 were being interpreted as octal.
2439 While fixing these problems I realized that there was another issue that
2440 hadn't been noticed. Values of message_size_limit (both the global option
2441 and the transport option) were treated as octal if they started with 0.
2442 The documentation was vague. These values are now always treated as
2443 decimal, and I will make that clear in the documentation.
2449 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2450 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2453 MH/01 Don't check that the operands of numeric comparison operators are
2454 integers when their expansion is in "skipping" mode (fixes bug
2455 introduced by 4.64-PH/07).
2457 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2458 child addresses, Exim now panics and dies. Previously, because the count
2459 is held in a short int, deliveries were likely to be lost. As such a
2460 large number of recipients for a single message is ridiculous
2461 (performance will be very, very poor), I have chosen to impose a limit
2462 rather than extend the field.
2468 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2469 leftover -K file (the existence of which was triggered by #402).
2470 While we were at it, introduced process PID as part of the -K
2471 filename. This should rule out race conditions when creating
2474 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2475 processing considerably. Previous code took too long for large mails,
2476 triggering a timeout which in turn triggers #401.
2478 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2479 in the DK code in transports.c. sendfile() is not really portable,
2480 hence the _LINUX specificness.
2482 TF/01 In the add_headers option to the mail command in an Exim filter,
2483 there was a bug that Exim would claim a syntax error in any
2484 header after the first one which had an odd number of characters
2487 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2488 callout verification, Exim cached a "reject" for the entire domain. This
2489 is correct for most verifications, but it is not correct for a recipient
2490 verification with use_sender or use_postmaster set, because in that case
2491 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2492 case of MAIL FROM:<> rejection from other early rejections (e.g.
2493 rejection of HELO). When verifying a recipient using a non-null MAIL
2494 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2495 Whatever the result of the callout, the value of the domain cache is
2496 left unchanged (for any other kind of callout, getting as far as trying
2497 RCPT means that the domain itself is ok).
2499 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2502 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2503 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2504 session, the daemon ignores ECONNECT errors and logs others; it now
2505 ignores EPIPE as well.
2507 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2508 (quoted-printable decoding).
2510 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2511 later the small subsequent patch to fix an introduced bug.
2513 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2515 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2517 PH/08 An error is now given if message_size_limit is specified negative.
2519 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2520 to be given (somewhat) arbitrary names.
2522 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2525 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2526 miscellaneous code fixes
2528 PH/10 Added the log_reject_target ACL modifier to specify where to log
2531 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2532 hostname. This is wrong, because it relates to the incoming message (and
2533 probably the interface on which it is arriving) and not to the outgoing
2534 callout (which could be using a different interface). This has been
2535 changed to use the value of the helo_data option from the smtp transport
2536 instead - this is what is used when a message is actually being sent. If
2537 there is no remote transport (possible with a router that sets up host
2538 addresses), $smtp_active_hostname is used.
2540 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2541 tweaks were necessary in order to get it to work (see also 21 below):
2542 (a) The code assumed that strncpy() returns a negative number on buffer
2543 overflow, which isn't the case. Replaced with Exim's string_format()
2545 (b) There were several signed/unsigned issues. I just did the minimum
2546 hacking in of casts. There is scope for a larger refactoring.
2547 (c) The code used strcasecmp() which is not a standard C function.
2548 Replaced with Exim's strcmpic() function.
2549 (d) The code set only $1; it now sets $auth1 as well.
2550 (e) A simple test gave the error "authentication client didn't specify
2551 service in request". It would seem that Dovecot has changed its
2552 interface. Fortunately there's a specification; I followed it and
2553 changed what the client sends and it appears to be working now.
2555 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2558 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2559 address A is aliased to B and C, where B exists and C does not. Without
2560 -v the output is "A verified" because verification stops after a
2561 successful redirection if more than one address is generated. However,
2562 with -v the child addresses are also verified. Exim was outputting "A
2563 failed to verify" and then showing the successful verification for C,
2564 with its parentage. It now outputs "B failed to verify", showing B's
2565 parentage before showing the successful verification of C.
2567 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2568 look up a TXT record in a specific list after matching in a combined
2571 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2572 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2573 they consult the DNS. I had assumed they would set it the way they
2574 wanted; and indeed my experiments on Linux seem to show that in some
2575 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2576 To be on the safe side, however, I have now made the interface to
2577 host_find_byname() similar to host_find_bydns(), with an argument
2578 containing the DNS resolver options. The host_find_byname() function now
2579 sets these options at its start, just as host_find_bydns() does. The smtp
2580 transport options dns_qualify_single and dns_search_parents are passed to
2581 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2582 of host_find_byname() use the default settings of RES_DEFNAMES
2583 (qualify_single) but not RES_DNSRCH (search_parents).
2585 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2586 spool_read_header() do less string testing, by means of a preliminary
2587 switch on the second character of optional "-foo" lines. (This is
2588 overdue, caused by the large number of possibilities that now exist.
2589 Originally there were few.) While I was there, I also converted the
2590 str(n)cmp tests so they don't re-test the leading "-" and the first
2591 character, in the hope this might squeeze out yet more improvement.
2593 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2594 flag allowing group syntax was set by the header_syntax check but not
2595 turned off, possible causing trouble later; (2) The flag was not being
2596 set at all for the header_verify test, causing "group"-style headers to
2597 be rejected. I have now set it in this case, and also caused header_
2598 verify to ignore an empty address taken from a group. While doing this, I
2599 came across some other cases where the code for allowing group syntax
2600 while scanning a header line wasn't quite right (mostly, not resetting
2601 the flag correctly in the right place). These bugs could have caused
2602 trouble for malformed header lines. I hope it is now all correct.
2604 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2605 with the "reply" argument non-NULL. The code, however (which originally
2606 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2607 but it didn't always do it. This confused somebody who was copying the
2608 code for some other use. I have removed all the tests.
2610 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2611 feature that was used to support insecure browsers during the U.S. crypto
2612 embargo. It requires special client support, and Exim is probably the
2613 only MTA that supported it -- and would never use it because real RSA is
2614 always available. This code has been removed, because it had the bad
2615 effect of slowing Exim down by computing (never used) parameters for the
2616 RSA_EXPORT functionality.
2618 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2619 authenticator to fail if there's a tab character in the incoming data
2620 (there should never be unless someone is messing about, as it's supposed
2621 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2622 if the connection is using TLS or if the remote IP is the same as the
2623 local IP, and the "valid-client-cert option" if a client certificate has
2626 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2627 authenticators. This can be used for authorization after authentication
2628 succeeds. (In the case of plaintext, it servers for both authentication
2631 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2632 if any retry times were supplied.
2634 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2635 connection, where there is no client IP address to check. In this
2636 situation, the verify now always succeeds.
2638 PH/25 Applied John Jetmore's -Mset patch.
2640 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2642 PH/27 In a string expansion for a processed (not raw) header when multiple
2643 headers of the same name were present, leading whitespace was being
2644 removed from all of them, but trailing whitespace was being removed only
2645 from the last one. Now trailing whitespace is removed from each header
2646 before concatenation. Completely empty headers in a concatenation (as
2647 before) are ignored.
2649 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2650 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2652 PH/29 [Removed. This was a change that I later backed out, and forgot to
2653 correct the ChangeLog entry (that I had efficiently created) before
2654 committing the later change.]
2656 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2657 address errors (4xx response to RCPT) over the same connection as other
2658 messages routed to the same hosts. Such deliveries are always "forced",
2659 so retry times are not inspected. This resulted in far too many retries
2660 for the affected addresses. The effect occurred only when there were more
2661 hosts than the hosts_max_try setting in the smtp transport when it had
2662 the 4xx errors. Those hosts that it had tried were not added to the list
2663 of hosts for which the message was waiting, so if all were tried, there
2664 was no problem. Two fixes have been applied:
2666 (i) If there are any address or message errors in an SMTP delivery, none
2667 of the hosts (tried or untried) are now added to the list of hosts
2668 for which the message is waiting, so the message should not be a
2669 candidate for sending over the same connection that was used for a
2670 successful delivery of some other message. This seems entirely
2671 reasonable: after all the message is NOT "waiting for some host".
2672 This is so "obvious" that I'm not sure why it wasn't done
2673 previously. Hope I haven't missed anything, but it can't do any
2674 harm, as the worst effect is to miss an optimization.
2676 (ii) If, despite (i), such a delivery is accidentally attempted, the
2677 routing retry time is respected, so at least it doesn't keep
2678 hammering the server.
2680 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2681 in ${readsocket because some servers need this prod.
2683 PH/32 Added some extra debug output when updating a wait-xxx database.
2685 PH/33 The hint "could be header name not terminated by colon", which has been
2686 given for certain expansion errors for a long time, was not being given
2687 for the ${if def:h_colon_omitted{... case.
2689 PH/34 The spec says: "With one important exception, whenever a domain list is
2690 being scanned, $domain contains the subject domain." There was at least
2691 one case where this was not true.
2693 PH/35 The error "getsockname() failed: connection reset by peer" was being
2694 written to the panic log as well as the main log, but it isn't really
2695 panic-worthy as it just means the connection died rather early on. I have
2696 removed the panic log writing for the ECONNRESET error when getsockname()
2699 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2700 runs only) independently of the message's sender address. This meant
2701 that, if the 4xx error was in fact related to the sender, a different
2702 message to the same recipient with a different sender could confuse
2703 things. In particular, this can happen when sending to a greylisting
2704 server, but other circumstances could also provoke similar problems.
2705 I have changed the default so that the retry time for these errors is now
2706 based a combination of the sender and recipient addresses. This change
2707 can be overridden by setting address_retry_include_sender=false in the
2710 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2711 remote server are returned as part of bounce messages. This was not
2712 happening for LMTP over a pipe (the lmtp transport), but now it is the
2713 same for both kinds of LMTP.
2715 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2716 in header lines that were in fact CNAMEs. This is no longer the case.
2718 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2719 and queue runs started by the daemon processed all messages. This has
2720 been fixed so that -R and -S can now usefully be given with -q<time>.
2722 PH/40 Import PCRE release 6.7 (fixes some bugs).
2724 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2726 PH/42 Give an error if -q is specified more than once.
2728 PH/43 Renamed the variables $interface_address and $interface_port as
2729 $received_ip_address and $received_port, to make it clear that these
2730 values apply to message reception, and not to the outgoing interface when
2731 a message is delivered. (The old names remain recognized, of course.)
2733 PH/44 There was no timeout on the connect() call when using a Unix domain
2734 socket in the ${readsocket expansion. There now is.
2736 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2737 be meaningful with "accept".
2739 SC/01 Eximstats V1.43
2740 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2742 SC/02 Eximstats V1.44
2743 Use a glob alias rather than an array ref in the generated
2744 parser. This improves both readability and performance.
2746 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2747 Collect SpamAssassin and rejection statistics.
2748 Don't display local sender or destination tables unless
2749 there is data to show.
2750 Added average volumes into the top table text output.
2752 SC/04 Eximstats V1.46
2753 Collect data on the number of addresses (recipients)
2754 as well as the number of messages.
2756 SC/05 Eximstats V1.47
2757 Added 'Message too big' to the list of mail rejection
2758 reasons (thanks to Marco Gaiarin).
2760 SC/06 Eximstats V1.48
2761 Mainlog lines which have GMT offsets and are too short to
2762 have a flag are now skipped.
2764 SC/07 Eximstats V1.49 (Alain Williams)
2765 Added the -emptyok flag.
2767 SC/08 Eximstats V1.50
2768 Fixes for obtaining the IP address from reject messages.
2770 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2771 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2772 whitespace changes from 4.64-PH/27
2774 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2777 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2778 are found, allow negative numbers in numeric criteria)
2780 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2782 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2783 to match changes made in 4.64-PH/43
2785 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2787 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2788 "domain" is actually an IP address, and force a failure. This locks out
2789 those revolvers/nameservers that support "A-for-A" lookups, in
2790 contravention of the specifications.
2792 PH/48 When a host name was looked up from an IP address, and the subsequent
2793 forward lookup of the name timed out, the host name was left in
2794 $sender_host_name, contrary to the specification.
2796 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2797 restricted to single-key lookups, Exim was not diagnosing an error if
2798 * or *@ was used with a query-style lookup.
2800 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2802 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2803 long ago, but noone interested enough thought of it. Let's just say that
2804 the "1.1" means that there are some new functions that weren't there at
2805 some point in the past.
2807 PH/51 Error processing for expansion failure of helo_data from an smtp
2808 transport during callout processing was broken.
2810 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2811 tested/used via the -bh/-bhc/-bs options.
2813 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2814 bug, fixed in subsequent PCRE releases).
2816 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2817 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2819 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2825 SC/01 Use a glob alias rather than an array ref in eximstats generated
2826 parser. This improves both readability and performance.
2828 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2829 Don't display local sender or destination tables in eximstats unless
2830 there is data to show.
2831 Added average volumes into the eximstats top table text output.
2833 SC/03 Collect data on the number of addresses (recipients) as well
2834 as the number of messages in eximstats.
2836 TF/01 Correct an error in the documentation for the redirect router. Exim
2837 does (usually) call initgroups() when daemonizing.
2839 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2840 with consistent privilege compared to when running as a daemon.
2842 TF/03 Note in the spec that $authenticated_id is not set for local
2843 submissions from trusted users.
2845 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2846 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2848 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2849 by adding some example configuration directives to the default
2850 configuration file. A little bit of work is required to uncomment the
2851 directives and define how usernames and passwords are checked, but
2852 there is now a framework to start from.
2854 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2855 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2856 without this. I don't know how relevant this is to other LDAP libraries.
2858 PH/02 Add the verb name to the "unknown ACL verb" error.
2860 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2862 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2864 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2865 directory not expanded when it should be if an expanded home directory
2866 was set for the address (which is overridden by the transport).
2868 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2871 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2872 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2873 because it is too late at that time, and has no effect.
2875 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2876 security issue with \' (bugzilla #107). I could not use the
2877 PQescapeStringConn() function, because it needs a PGconn value as one of
2880 PH/08 When testing addresses using -bt, indicate those final addresses that
2881 are duplicates that would not cause an additional delivery. At least one
2882 person was confused, thinking that -bt output corresponded to deliveries.
2883 (Suppressing duplicates isn't a good idea as you lose the information
2884 about possibly different redirections that led to the duplicates.)
2886 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2887 systems where poll() doesn't work, in particular OS X.
2889 PH/10 Added more information to debugging output for retry time not reached.
2891 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2892 operations in malware.c.
2894 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2897 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2898 syslog_duplication set false, log lines that would normally be written
2899 both the the main log and to the reject log were not written to syslog at
2902 PH/14 In the default configuration, change the use of "message" in ACL warn
2903 statements to "add_header".
2905 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2906 not followed by a command (e.g. "seen endif").
2908 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2909 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2912 PH/17 Added extra conditions to the default value of delay_warning_condition
2916 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2917 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2918 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2921 The Auto-Submitted: and various List- headers are standardised, whereas I
2922 don't think Precedence: ever was.
2924 PH/18 Refactored debugging code in route_finduser() to show more information,
2925 in particular, the error code if getpwnam() issues one.
2927 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2928 This is apparently needed in addition to the PH/07 change above to avoid
2929 any possible encoding problems.
2931 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2932 but not after initializing Perl.
2934 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2935 output them only if debugging. By default they are written stderr,
2936 apparently, which is not desirable.
2938 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2941 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2944 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2946 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2947 authenticated or an ident call has been made. Suppress the default
2948 values for $authenticated_id and $authenticated_sender (but permit -oMai
2949 and -oMas) when testing with -bh.
2951 PH/24 Re-jigged the order of the tests in the default configuration so that the
2952 tests for valid domains and recipients precede the DNS black list and CSA
2953 tests, on the grounds that those ones are more expensive.
2955 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2956 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2957 This bug exists in every version of Exim that I still have, right back to
2960 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2961 However, an attempt to turn on case-sensitivity in a regex key by
2962 including (?-i) didn't work because the subject string was already
2963 lowercased, and the effects were non-intuitive. It turns out that a
2964 one-line patch can be used to allow (?-i) to work as expected.
2970 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2971 other effects) broke the use of negated acl sub-conditions.
2973 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2976 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2977 "Deny" causes Exim to reject the incoming connection with a 554 error.
2978 Unfortunately, if there is a major crisis, such as a disk failure,
2979 tcp-wrappers gives "deny", whereas what one would like would be some
2980 kind of temporary error. A kludge has been added to help with this.
2981 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2982 554 error is used if errno is still zero or contains ENOENT (which occurs
2983 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2986 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2988 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2989 errors. Otherwise a message that provokes a temporary error (when other
2990 messages do not) can cause a whole host to time out.
2992 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2993 addresses were routed directly to files or pipes from a redirect router.
2994 File deliveries just didn't batch; pipe deliveries might have suffered
2997 PH/06 A failure to get a lock for a hints database would erroneously always say
2998 "Failed to get write lock", even when it was really a read lock.
3000 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3001 of 0600. This has been changed to use the value of the lockfile_mode
3002 option (which defaults to 0600).
3004 PH/08 Applied small patch from the Sieve maintainer.
3006 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3007 folder from quota calculations, a direct delivery into this folder messed
3008 up the contents of the maildirsize file. This was because the regex was
3009 used only to exclude .Trash (or whatever) when the size of the mailbox
3010 was calculated. There was no check that a delivery was happening into an
3011 excluded directory. This bug has been fixed by ignoring all quota
3012 processing for deliveries into excluded directories.
3014 PH/10 Added the maildirfolder_create_regex option to appendfile.
3020 PH/01 The code for finding all the local interface addresses on a FreeBSD
3021 system running IPv6 was broken. This may well have applied to all BSD
3022 systems, as well as to others that have similar system calls. The broken
3023 code found IPv4 interfaces correctly, but gave incorrect values for the
3024 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3025 that it would not match correctly against @[] and not recognize the IPv6
3028 PH/02 The ipliteral router was not recognizing addresses of the form user@
3029 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3031 PH/03 Added disable_ipv6.
3033 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3034 decoded form, because it is most often used to construct To: headers
3035 lines in autoreplies, and the decoded form may well be syntactically
3036 invalid. However, $reply_address has leading white space removed, and all
3037 newlines turned into spaces so that the autoreply transport does not
3040 PH/05 If group was specified without a user on a router, and no group or user
3041 was specified on a transport, the group from the router was ignored.
3043 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3044 for visible compile-time settings that can be used to change these
3045 numbers, for those that want even more. Backwards compatibility with old
3046 spool files has been maintained. However, going back to a previous Exim
3047 release will lost any variables that are in spool files.
3049 PH/07 Two small changes when running in the test harness: increase delay when
3050 passing a TCP/IP connection to a new process, in case the original
3051 process has to generate a bounce, and remove special handling of
3052 127.0.0.2 (sic), which is no longer necessary.
3054 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3055 be the same on different OS.
3057 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3060 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3061 whether --show-vars was specified or not
3063 JJ/02 exipick: Added support for new ACL variable spool format introduced
3066 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3067 syntactically invalid From: or Reply-to: line, and a filter used this to
3068 generate an autoreply, and therefore failed to obtain an address for the
3069 autoreply, Exim could try to deliver to a non-existent relative file
3070 name, causing unrelated and misleading errors. What now happens is that
3071 it logs this as a hard delivery error, but does not attempt to create a
3074 PH/11 The exinext utility has a -C option for testing purposes, but although
3075 the given file was scanned by exinext itself; it wasn't being passed on
3076 when Exim was called.
3078 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3079 an end-of-file indication when reading a command response.
3081 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3082 compiled. In many other places in Exim, IPv6 addresses are always
3083 recognized, so I have changed this. It also means that IPv4 domain
3084 literals of the form [IPV4:n.n.n.n] are now always recognized.
3086 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3087 used if the router is not running as root, for example, when verifying at
3088 ACL time, or when using -bh. The debugging output from this situation was
3089 non-existent - all you got was a failure to exec. I have made two
3092 (a) Failures to set uid/gid, the current directory, or a process leader
3093 in a subprocess such as that created by queryprogram now generate
3094 suitable debugging output when -d is set.
3096 (b) The queryprogram router detects when it is not running as root,
3097 outputs suitable debugging information if -d is set, and then runs
3098 the subprocess without attempting to change uid/gid.
3100 PH/15 Minor change to Makefile for building test_host (undocumented testing
3103 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3104 additional section of a DNS packet that returns MX or SRV records.
3105 Instead, it always explicitly searches for A/AAAA records. This avoids
3106 major problems that occur when a DNS server includes only records of one
3107 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3108 fixed another bug: if SRV records were looked up and the corresponding
3109 address records were *not* found in the additional section, the port
3110 values from the SRV records were lost.
3112 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3113 using the correct key (the original address) when searching the retry
3114 rules in order to find which one to use for generating the retry hint.
3116 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3117 adding the default one. Similarly, if it contains a Reply-To: header, the
3118 errors_reply_to option, if set, is not used.
3120 PH/19 When calculating a retry time, Exim used to measure the "time since
3121 failure" by looking at the "first failed" field in the retry record. Now
3122 it does not use this if it is later than than the arrival time of the
3123 message. Instead it uses the arrival time. This makes for better
3124 behaviour in cases where some deliveries succeed, thus re-setting the
3125 "first failed" field. An example is a quota failure for a huge message
3126 when small messages continue to be delivered. Without this change, the
3127 "time since failure" will always be short, possible causing more frequent
3128 delivery attempts for the huge message than are intended.
3129 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3131 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3132 $1, $2, $3) because the numerical variables can be reset during some
3133 expansion items (e.g. "match"), thereby losing the authentication data.
3135 PH/21 Make -bV show the size of off_t variables so that the test suite can
3136 decide whether to run tests for quotas > 2G.
3138 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3139 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3140 filecount value is greater than 2G or if a quota value is greater than 2G
3141 on a system where the size of off_t is not greater than 4, a panic error
3144 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3145 never match. The debug and -bh output now contains an explicit error
3146 message indicating a malformed IPv4 address or mask.
3148 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3149 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3150 PH/23 above applies.
3152 PH/25 Do not write to syslog when running in the test harness. The only
3153 occasion when this arises is a failure to open the main or panic logs
3154 (for which there is an explicit test).
3156 PH/26 Added the /no_tell option to "control=freeze".
3158 PH/27 If a host name lookup failed very early in a connection, for example, if
3159 the IP address matched host_lookup and the reverse lookup yielded a name
3160 that did not have a forward lookup, an error message of the form "no IP
3161 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3162 could be logged. Now it outputs the IP address instead of "NULL".
3164 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3165 allows the sender and the authenticated sender to be set when
3166 submitting a message from within Exim. Since child_open_exim() is
3167 documented for local_scan(), the new function should be too.
3169 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3170 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3171 results in an empty string is now treated as unset.
3173 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3175 PH/31 Added sender_verify_fail logging option.
3177 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3178 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3179 tidied the source and removed it altogether.
3181 PH/33 When a queue run was abandoned because the load average was too high, a
3182 log line was always written; now it is written only if the queue_run log
3183 selector is set. In addition, the log line for abandonment now contains
3184 information about the queue run such as the pid. This is always present
3185 in "start" and "stop" lines but was omitted from the "abandon" line.
3187 PH/34 Omit spaces between a header name and the colon in the error message that
3188 is given when verify = headers_syntax fails (if there are lots of them,
3189 the message gets confusing).
3191 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3192 names, as there are now some PTR records that contain slashes. This check
3193 is only to protect against broken name servers that fall over on strange
3194 characters, so the fact that it applies to all lookups doesn't matter.
3196 PH/36 Now that the new test suite is complete, we can remove some of the
3197 special code in Exim that was needed for the old test suite. For example,
3198 sorting DNS records because real resolvers return them in an arbitrary
3199 order. The new test suite's fake resolver always returns records in the
3202 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3203 bounces) except when queue_only is set, to avoid logging races between
3204 the different processes.
3206 PH/38 Panic-die if .include specifies a non-absolute path.
3208 PH/39 A tweak to the "H" retry rule from its user.
3210 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3211 a label. They prevented compilation on older perls.
3213 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3214 a warning to be raised on newish perls.
3216 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3217 on queue. Changes to match documented behaviour of showing count of
3218 messages matching specified criteria.
3220 PH/40 Changed the default ident timeout from 30s to 5s.
3222 PH/41 Added support for the use of login_cap features, on those BSD systems
3223 that have them, for controlling the resources used by pipe deliveries.
3225 PH/42 The content-scanning code uses fopen() to create files in which to put
3226 message data. Previously it was not paying any attention to the mode of
3227 the files. Exim runs with umask(0) because the rest of the code creates
3228 files with open(), and sets the required mode explicitly. Thus, these
3229 files were ending up world-writeable. This was not a big issue, because,
3230 being within the spool directory, they were not world-accessible. I have
3231 created a function called modefopen, which takes an additional mode
3232 argument. It sets umask(777), creates the file, chmods it to the required
3233 mode, then resets the umask. All the relevant calls to fopen() in the
3234 content scanning code have been changed to use this function.
3236 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3237 to 24 hours. This avoids potential overflow problems when processing G
3238 and H retry rules. I suspect nobody ever tinkers with this value.
3240 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3242 PH/45 When the plaintext authenticator is running as a client, the server's
3243 challenges are checked to ensure they are valid base64 strings. By
3244 default, the authentication attempt is cancelled if an invalid string is
3245 received. Setting client_ignore_invalid_base64 true ignores these errors.
3246 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3247 they are received. Thus, the responses can be made to depend on the
3248 challenges. If an invalid string is ignored, an empty string is placed in
3251 PH/46 Messages that are created by the autoreply transport now contains a
3252 References: header, in accordance with RFCs 2822 and 3834.
3254 PH/47 Added authenticated_sender_force to the smtp transport.
3256 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3258 PH/49 Installed latest patch from the Sieve maintainer.
3260 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3261 was also set, the appendfile transport was unnecessarily scanning a
3262 directory of message files (e.g. for maildir delivery) to find the count
3263 of files (along with the size), even though it did not need this
3264 information. It now does the scan only if it needs to find either the
3265 size of the count of files.
3267 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3269 PH/52 Two bugs concerned with error handling when the smtp transport is
3272 (i) Exim was not creating retry information for temporary errors given
3273 for individual recipients after the DATA command when the smtp transport
3274 was used in LMTP mode. This meant that they could be retried too
3275 frequently, and not timed out correctly.
3277 (ii) Exim was setting the flag that allows error details to be returned
3278 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3279 recipients that were returned after the DATA command.
3281 PH/53 This is related to PH/52, but is more general: for any failing address,
3282 when detailed error information was permitted to be returned to the
3283 sender, but the error was temporary, then after the final timeout, only
3284 "retry timeout exceeded" was returned. Now it returns the full error as
3285 well as "retry timeout exceeded".
3287 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3288 do this, and (what is worse) MTAs that accept it.
3290 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3291 will now be deprecated.
3293 PH/56 New os.c-cygwin from the Cygwin maintainer.
3295 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3296 formats (previously only available in exim formats via -bpr, -bpru,
3297 and -bpra. Now also available in native and exiqgrep formats)
3299 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3300 with very large, slow to parse queues
3302 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3304 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3306 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3307 responses to authentication challenges, though it was showing the
3308 challenges; (ii) I've removed the CR characters from the debug output for
3311 PH/58 Allow for the insertion of a newline as well as a space when a string
3312 is turned into more than one encoded-word during RFC 2047 encoding. The
3313 Sieve code now uses this.
3315 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3316 data_4xx, lost_connection, tls_required.
3318 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3319 message was being sent as an SMTP response.
3321 PH/61 Add -l and -k options to exicyclog.
3323 PH/62 When verifying, if an address was redirected to one new address, so that
3324 verification continued, and the new address failed or deferred after
3325 having set something in $address_data, the value of $address_data was not
3326 passed back to the ACL. This was different to the case when no
3327 redirection occurred. The value is now passed back in both cases.
3329 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3330 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3331 use login_cap.h, so on its own it isn't the distinguishing feature. The
3332 new name refers directly to the setclassresources() function.
3334 PH/65 Added configuration files for NetBSD3.
3336 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3338 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3339 is preferred over IPv4.
3341 PH/68 The bounce_return_message and bounce_return_body options were not being
3342 honoured for bounces generated during the reception of non-SMTP messages.
3343 In particular, this applied to messages rejected by the ACL. This bug has
3344 been fixed. However, if bounce_return_message is true and bounce_return_
3345 body is false, the headers that are returned for a non-SMTP message
3346 include only those that have been read before the error was detected.
3347 (In the case of an ACL rejection, they have all been read.)
3349 PH/69 The HTML version of the specification is now built in a directory called
3350 spec_html instead of spec.html, because the latter looks like a path with
3351 a MIME-type, and this confuses some software.
3353 PH/70 Catch two compiler warnings in sieve.c.
3355 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3356 function verify_get_ident() calls ip_connect() to connect a socket, but
3357 if the "connect()" function timed out, ip_connect() used to close the
3358 socket. However, verify_get_ident() also closes the socket later, and in
3359 between Exim writes to the log, which may get opened at this point. When
3360 the socket was closed in ip_connect(), the log could get the same file
3361 descriptor number as the socket. This naturally causes chaos. The fix is
3362 not to close the socket in ip_connect(); the socket should be closed by
3363 the function that creates it. There was only one place in the code where
3364 this was missing, in the iplookup router, which I don't think anybody now
3365 uses, but I've fixed it anyway.
3367 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3368 well as to direct DNS lookups. Otherwise the handling of names in host
3369 lists is inconsistent and therefore confusing.
3375 PH/01 Two changes to the default runtime configuration:
3377 (1) Move the checks for relay_from_hosts and authenticated clients from
3378 after to before the (commented out) DNS black list checks.
3380 (2) Add control=submission to the relay_from_hosts and authenticated
3381 clients checks, on the grounds that messages accepted by these
3382 statements are most likely to be submissions.
3384 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3386 (1) Generate an error if the third argument for the ${prvs expansion is
3389 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3392 (3) Reset the variables that are obtained from the first argument of
3393 ${prvscheck and used in the second argument before leaving the code,
3394 because their memory is reclaimed, so using them afterwards may do
3397 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3398 one (it's much easier than Tom thought :-).
3400 (5) Because of (4), we can now allow for the use of $prvscheck_result
3401 inside the third argument.
3403 PH/03 For some reason, the default setting of PATH when running a command from
3404 a pipe transport was just "/usr/bin". I have changed it to
3407 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3408 anything to be listed in the output from -bV.
3410 PH/05 When a filter generated an autoreply, the entire To: header line was
3411 quoted in the delivery log line, like this:
3413 => >A.N.Other <ano@some.domain> <original@ddress> ...
3415 This has been changed so that it extracts the operative address. There
3416 may be more than one such address. If so, they are comma-separated, like
3419 => >ano@some.domain,ona@other.domain <original@ddress> ...
3421 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3422 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3423 not being looked up in the rDNS to get a host name, Exim was showing the
3424 IP address twice in Received: lines, even though the IP addresses were
3425 identical. For example:
3427 Received: from [1.2.3.4] (helo=[1.2.3.4])
3429 However, if the real host name was known, it was omitting the HELO data
3430 if it matched the actual IP address. This has been tidied up so that it
3431 doesn't show the same IP address twice.
3433 PH/07 When both +timestamp and +memory debugging was on, the value given by
3434 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3435 called by the debug printing, thereby overwriting the timestamp buffer.
3436 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3438 PH/08 When the original message was included in an autoreply transport, it
3439 always said "this is a copy of the message, including all the headers",
3440 even if body_only or headers_only was set. It now gives an appropriate
3443 PH/09 Applied a patch from the Sieve maintainer which:
3445 o fixes some comments
3446 o adds the (disabled) notify extension core
3447 o adds some debug output for the result of if/elsif tests
3448 o points to the current vacation draft in the documentation
3449 and documents the missing references header update
3453 o fixes a bug in processing the envelope test (when testing
3454 multiple envelope elements, the last element determined the
3457 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3458 Electronic Mail") by including:
3460 Auto-submitted: auto-generated
3462 in the messages that it generates (bounce messages and others, such as
3463 warnings). In the case of bounce messages for non-SMTP messages, there was
3464 also a typo: it was using "Auto_submitted" (underscore instead of
3465 hyphen). Since every message generated by Exim is necessarily in response
3466 to another message, thes have all been changed to:
3468 Auto-Submitted: auto-replied
3470 in accordance with these statements in the RFC:
3472 The auto-replied keyword:
3474 - SHOULD be used on messages sent in direct response to another
3475 message by an automatic process,
3477 - MUST NOT be used on manually-generated messages,
3479 - MAY be used on Delivery Status Notifications (DSNs) and Message
3480 Disposition Notifications (MDNs),
3482 - MUST NOT be used on messages generated by automatic or periodic
3483 processes, except for messages which are automatic responses to
3486 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3487 to the default Received: header definition.
3489 PH/12 Added log selector acl_warn_skipped (default on).
3491 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3492 variables because (a) they are in the wrong storage pool and (b) even if
3493 they were copied, it wouldn't work properly because of the caching.
3495 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3496 checking when decoding. Apparently there are clients that generate
3497 overlong encoded strings. Why am I not surprised?
3499 PH/15 If the first argument of "${if match_address" was not empty, but did not
3500 contain an "@" character, Exim crashed. Now it writes a panic log message
3501 and treats the condition as false.
3503 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3505 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3506 extension "envelope-auth". The code is finished and in agreement with
3507 other implementations, but there is no documentation so far and in fact,
3508 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3509 not changing the active code.
3511 Print executed "if" and "elsif" statements when debugging is used. This
3512 helps a great deal to understand what a filter does.
3514 Document more things not specified clearly in RFC3028. I had all this
3515 sorted out, when out of a sudden new issues came to my mind. Oops."
3517 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3520 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3521 canonical form (as documented). However, after a host name lookup from
3522 the IP address, check_host() was doing a simple string comparison with
3523 addresses acquired from the DNS when checking that the found name did
3524 have the original IP as one of its addresses. Since any found IPv6
3525 addresses are likely to be in abbreviated form, the comparison could
3526 fail. Luckily, there already exists a function for doing the comparison
3527 by converting both addresses to binary, so now that is used instead of
3528 the text comparison.
3530 PH/20 There was another similar case to PH/19, when a complete host name was
3531 given in a host list; looking up its IP address could give an abbreviated
3532 form, whereas the current host's name might or might not be abbreviated.
3533 The same fix has been applied.
3539 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3540 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3543 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3544 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3546 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3548 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3549 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3550 Also, for "Auto-Submitted", treat anything other than "no" as
3551 non-personal, in accordance with RFC 3834. (Previously it treated
3552 anything starting "auto-" as non-personal.)
3554 TF/01 The control=submission/name=... option had a problem with syntax
3555 errors if the name included a slash character. The /name= option
3556 now slurps the rest of the string, so it can include any characters
3557 but it must come last in the list of options (after /sender_retain
3560 PH/05 Some modifications to the interface to the fake nameserver for the new
3568 TK/01 Added the "success_on_redirect" address verification option. See
3569 NewStuff for rationale and an example.
3571 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3573 PH/02 Patch to exigrep to allow it to work on syslog lines.
3575 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3576 fread() to skip over the body file's header line, because in Cygwin the
3577 header line is locked and is inaccessible.
3579 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3580 co-exist for some time) to make it clear that it is the Exim ID that is
3581 referenced, not the Message-ID: header line.
3583 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3584 string_format() function, because snprintf() does not exist on all
3587 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3588 ${stat: expansion item.
3590 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3591 protocol synchronization error", to keep the pedants happy.
3593 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3594 well as for IRIX systems, when gcc is being used. See the host.c source
3597 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3599 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3602 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3603 transport and to the smtp transport in LMTP mode.
3605 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3607 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3608 run a filter in a subprocess. This could lead to confusion in subsequent
3609 lookups in the parent process. There should also be a search_tidyup() at
3610 the end of the subprocess.
3612 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3613 only if the host matched helo_try_verify_hosts, which caused the
3614 verification to occur when the EHLO/HELO command was issued. The ACL just
3615 tested the remembered result. Now, if a previous verification attempt has
3616 not happened, "verify = helo" does it there and then.
3618 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3620 TK/03 Fix log output including CR from clamd.
3622 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3623 exist provoked a memory error which could cause a segfault.
3625 PH/15 Installed PCRE 6.2
3627 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3629 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3630 of the problem. Specifically, suggested +O2 rather than +O1 for the
3633 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3635 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3636 record, the port was not correctly logged when the outgoing_port log
3637 selector was set (it logged the transort's default port).
3639 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3640 fallback_hosts, and "hosts" in the smtp transport.
3642 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3643 host errors such as "Connection refused".
3645 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3646 authentication with radiusclient 0.4.9:
3648 - Error returned from rc_read_config was caught wrongly
3649 - Username/password not passed on to radius server due to wrong length.
3651 The presumption is that some radiusclient API changes for 4.51/PH/17
3652 were not taken care of correctly. The code is still untested by me (my
3653 Linux distribution still has 0.3.2 of radiusclient), but it was
3654 contributed by a Radius user.
3656 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3657 expanding the "port" option of the smtp transport.
3659 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3660 while reading a MIME header. Thanks to Tom Hughes for a patch.
3662 PH/24 Include config.h inside local_scan.h so that configuration settings are
3665 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3666 that in an ACL for RCPT (for example), you can examine exactly what was
3669 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3670 commands, but it was not correctly comparing the address with the actual
3671 client host address. Thus, it would show the EHLO address in Received:
3672 header lines when this was not necessary.
3674 PH/27 Added the % operator to ${eval:}.
3676 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3677 it should be ignoring failures (because with -C, for example, it has lost
3678 privilege). It wasn't ignoring creation failures other than "already
3681 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3684 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3685 because an input line was too long, either on its own, or by virtue of
3686 too many continuations, the temporary file was not being removed, and the
3687 return code was incorrect.
3689 PH/31 Missing "BOOL" in function definition in filtertest.c.
3691 PH/32 Applied Sieve patches from the maintainer.
3693 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3695 PH/33 Added "verify = not_blind".
3697 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3698 Local/Makefile (with some defaults set). These are used in built scripts
3699 such as exicyclog, but they have never been used in the exim_install
3700 script (though there are many overriding facilities there). I have
3701 arranged that the exim_install script now takes note of these two
3704 PH/35 Installed configuration files for Dragonfly.
3706 PH/36 When a locally submitted message by a trusted user did not contain a
3707 From: header, and the sender address was obtained from -f or from an SMTP
3708 MAIL command, and the trusted user did not use -F to supply a sender
3709 name, $originator_name was incorrectly used when constructing a From:
3710 header. Furthermore, $originator_name was used for submission mode
3711 messages from external hosts without From: headers in a similar way,
3712 which is clearly wrong.
3714 PH/37 Added control=suppress_local_fixups.
3716 PH/38 When log_selector = +received_sender was set, and the addition of the
3717 sender made the log line's construction buffer exactly full, or one byte
3718 less than full, an overflow happened when the terminating "\n" was
3721 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3722 when the result of a list match is failure because a DNS lookup failed.
3724 PH/40 RM_COMMAND is now used in the building process.
3726 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3727 the "build-* directories that it finds.
3729 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3730 address in a domain literal was a prefix of an interface address.
3732 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3733 when verifying a sender address, unless rewrite_headers is false.
3735 PH/44 Wrote a long comment about why errors_to addresses are verified as
3736 recipients, not senders.
3738 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3739 the ratelimit ACL was added.
3741 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3743 PH/46 Added extra information about PostgreSQL errors to the error string.
3745 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3746 avoiding the need to install special zones in a real server. This is
3747 backwards compatible; if it can't find the fake resolver, it drops back.
3748 Thus, both old and new test suites can be run.
3750 TF/02 Added util/ratelimit.pl
3752 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3753 clock is set back in time.
3755 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3756 Candler <B.Candler@pobox.com>.
3758 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3759 OK for addresses that are the result of redirections.
3761 PH/48 A number of further additions for the benefit of the new test suite,
3762 including a fake gethostbyname() that interfaces to the fake DNS resolver
3765 TF/06 The fix for widen_domains has also been applied to qualify_single and
3766 search_parents which are the other dnslookup options that can cause
3769 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3772 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3774 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3775 service" if it encountered an MX record with an empty target hostname.
3776 The message is now "an MX or SRV record indicated no SMTP service".
3778 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3779 if the DNS of the sending site is misconfigured. This is quite a
3780 common situation. This change restores the behaviour of treating a
3781 helo verification defer as a failure.
3783 PH/51 If self=fail was set on a router, the bounce message did not include the
3784 actual error message.
3790 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3792 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3793 command itself ended in error, the underlying message about the transport
3794 filter timeout was being overwritten with the pipe command error. Now the
3795 underlying error message should be appended to the second error message.
3797 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3799 PH/02 Reduce the amount of output that "make" produces by default. Full output
3800 can still be requested.
3802 PH/03 The warning log line about a condition test deferring for a "warn" verb
3803 was being output only once per connection, rather than after each
3804 occurrence (because it was using the same function as for successful
3805 "warn" verbs). This seems wrong, so I have changed it.
3807 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3808 it should not have, which might have caused a crash in the right
3809 circumstances, but probably never did.
3811 PH/04 Installed a modified version of Tony Finch's patch to make submission
3812 mode fix the return path as well as the Sender: header line, and to
3813 add a /name= option so that you can make the user's friendly name appear
3816 TF/03 Added the control = fakedefer ACL modifier.
3818 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3819 Mark Lowes for thorough testing.
3821 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3823 TK/03 Merged latest SRS patch from Miles Wilton.
3825 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3826 with the definition in sysexits.h (which is #included earlier).
3827 Fortunately, Exim does not actually use EX_OK. The code used to try to
3828 preserve the sysexits.h value, by assuming that macro definitions were
3829 scanned for macro replacements. I have been disabused of this notion,
3830 so now the code just undefines EX_OK before #including unistd.h.
3832 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3833 in the smtp transport. When a block could not be written in a single
3834 write() function, the timeout was being re-applied to each part-write.
3835 This seems wrong - if the receiver was accepting one byte at a time it
3836 would take for ever. The timeout is now adjusted when this happens. It
3837 doesn't have to be particularly precise.
3839 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3840 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3842 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3843 without a domain if the check to <postmaster@domain> fails.
3845 SC/01 Eximstats: added -xls and the ability to specify output files
3846 (patch written by Frank Heydlauf).
3848 SC/02 Eximstats: use FileHandles for outputting results.
3850 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3852 SC/04 Eximstats: fixed display of large numbers with -nvr option
3854 SC/05 Eximstats: fixed merging of reports with empty tables.
3856 SC/06 Eximstats: added the -include_original_destination flag
3858 SC/07 Eximstats: removed tabs and trailing whitespace.
3860 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3862 TK/06 MBOX spool code: Add real "From " MBOX separator line
3863 so the .eml file is really in mbox format (even though
3864 most programs do not really care). Patch from Alex Miller.
3866 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3867 The latter is generated from $received_to and is only set if the
3868 message has one envelope recipient. SA can use these headers,
3869 obviously out-of-the-box. Patch from Alex Miller.
3871 PH/08 The ${def test on a variable was returning false if the variable's
3872 value was "0", contrary to what the specification has always said!
3873 The result should be true unless the variable is empty.
3875 PH/09 The syntax error of a character other than { following "${if
3876 def:variable_name" (after optional whitespace) was not being diagnosed.
3877 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3878 accidental colon was present, for example, could give incorrect results.
3880 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3881 result is used (not including appendfile, where other changes are about
3884 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3885 This involved changing a lot of size variables from int to off_t. It
3886 should work with maildirs and everything.
3888 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3889 spamd dying while we are connected to it.
3891 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3894 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3895 function for BDB 4.3.
3897 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3899 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3900 the sources for PCRE have been re-arranged and more files are now
3903 PH/15 The code I had for printing potentially long long variables in PH/11
3904 above was not the best (it lost precision). The length of off_t variables
3905 is now inspected at build time, and an appropriate printing format (%ld
3906 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3907 to be "long long int" or "long int". This is needed for the internal
3908 formatting function string_vformat().
3910 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3911 the configuration file to be ":syslog", then the script "guesses" where
3912 the logs files are, rather than using the compiled in default. In our
3913 case the guess is not the same as the compiled default, so the script
3914 suddenly stopped working when I started to use syslog. The patch checks
3915 to see if log_file_path is "". If so, it attempts to read it from exim
3916 with no configuration file to get the compiled in version, before it
3917 falls back to the previous guessing code."
3919 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3920 implementing BATV in an Exim configuration. See NewStuff for the gory
3923 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3924 Makefile that are specific to HP-UX.
3926 PH/18 If the "use_postmaster" option was set for a recipient callout together
3927 with the "random" option, the postmaster address was used as the MAIL
3928 FROM address for the random test, but not for the subsequent recipient
3929 test. It is now used for both.
3931 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3932 patch removes a few documentation additions to RFC 3028, because the
3933 latest draft now contains them. It adds the new en;ascii-case comparator
3934 and a new error check for 8bit text in MIME parts. Comparator and
3935 require names are now matched exactly. I enabled the subaddress
3936 extension, but it is not well tested yet (read: it works for me)."
3938 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3939 rework some of the code of TK/09 above to avoid the hardwired use of
3940 "%lld" and "long long". Replaced the call to snprintf() with a call to
3943 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3944 records point to non-existent hosts", "retry timeout exceeded", and
3945 "retry time not reached for any host after a long failure period".
3947 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3948 experimental DomainKeys support:
3950 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3951 (2) On an error such as an illegally used "control", the wrong name for
3952 the control was given.
3954 These problems did NOT occur unless DomainKeys support was compiled.
3956 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3958 PH/24 Added ${if match_ip condition.
3960 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3961 databases so that it will be absolutely obvious if a crash occurs in the
3962 DB library. This is a regular occurrence (often caused by mis-matched
3965 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3966 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3967 picked up on a user's system that detects such things. There doesn't seem
3968 to be a gcc warning option for this - only an attribute that has to be
3969 put on the function's prototype. It seems that in Fedora Core 4 they have
3970 set this on a number of new functions. No doubt there will be more in due
3973 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3974 specify a transport. However, if an address that was verified by such a
3975 router was the subject of a callout, Exim crashed because it tried to
3976 read the rcpt_include_affixes from the non-existent transport. Now it
3977 just assumes that the setting of that option is false. This bug was
3978 introduced by 4.51/PH/31.
3980 PH/28 Changed -d+all to exclude +memory, because that information is very
3981 rarely of interest, but it makes the output a lot bigger. People tend to
3982 do -d+all out of habit.
3984 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3985 code in os-type was giving problems when libc.so lives in lib64, like on
3988 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3989 aren't the modern standard, and it seems that some systems' include files
3990 don't always have them. Exim was already checking for some of the newer
3991 ones like T_AAAA, and defining it itself. I've added checks for all the
3992 record types that Exim uses.
3994 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3995 not automatically generating a new one, as it is supposed to. This
3996 prevented TLS from working. If the file did exist, but contained invalid
3997 data, a new version was generated, as expected. It was only the case of a
3998 non-existent file that was broken.
4000 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4001 with a change in libdomainkeys > 0.64.
4003 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4004 from DNS. If the selector record carries the flag, it now has
4005 precedence over the domain-wide flag.
4007 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4009 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4010 the use of a port name that isn't defined in /etc/services) occurred, the
4011 message was deferred as in a normal delivery, and thus remained on the
4012 spool, instead of being failed because of the mua_wrapper setting. This
4013 is now fixed, and I tidied up some of the mua_wrapper messages at the
4016 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4017 the messages in a hash of arrays rather than using individual hashes.
4018 This is a bit cleaner and results in dramatic memory savings, albeit
4019 at a slight CPU cost.
4021 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4022 as requested by Marc Sherman.
4024 SC/10 Eximstats: added histograms for user specified patterns as requested
4027 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4029 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4030 fopen() in the content-scanning modules that did not already have it.
4036 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4037 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4039 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4041 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4043 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4044 to test Sieve filters that use "vacation".
4046 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4047 that changes the way the GnuTLS parameters are stored in the cache file.
4048 The new format can be generated externally. For backward compatibility,
4049 if the data in the cache doesn't make sense, Exim assumes it has read an
4050 old-format file, and it generates new data and writes a new file. This
4051 means that you can't go back to an older release without removing the
4054 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4055 work if there are any delivery delays because "one_time" forces the
4056 parent to be marked "delivered", so its unseen clone is never tried
4057 again. For this reason, Exim now forbids the simultaneous setting of
4060 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4061 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4062 entry if you want to know the details. The fix, however, neglected to
4063 consider the case when local delivery batching is involved. The test for
4064 "previously delivered" was not happening when checking to see if an
4065 address could be batched with a previous (undelivered) one; under
4066 certain circumstances this could lead to multiple deliveries to the same
4069 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4070 in its include files, and this causes problems building Exim.
4072 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4073 header_syntax) but Exim was just ignoring anything given after a slash.
4074 In particular, this caused confusion with an attempt to use "verify =
4075 reverse_host_lookup/defer_ok". An error is now given when options are
4076 supplied for verify items that do not have them. (Maybe reverse_host_
4077 lookup should have a defer_ok option, but that's a different point.)
4079 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4080 defined by RFC 821) to 2048, because there were problems with some AUTH
4081 commands, and RFC 1869 says the size should be increased for extended
4082 SMTP commands that take arguments.
4084 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4087 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4088 "unknown" error; now it says that the functionality isn't in the binary.
4090 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4091 an address' error message when a string expansion fails (syntax or
4092 whatever). Otherwise the password may appear in the log. Following change
4093 PH/42 below, there is no longer a chance of it appearing in a bounce
4096 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4098 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4099 of its addresses was ever tried. (Bugzilla bug #2.)
4101 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4102 the result incorrectly in the debug output. (It correctly added a newline
4103 to what was transported.)
4105 TF/01 Added $received_time.
4107 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4108 commented out examples of how to interface to a virus scanner and to
4109 SpamAssassin. Also added commented examples of av_scanner and
4110 spamd_address settings.
4112 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4113 and controls are allowed in which ACLs. There were a couple of minor
4114 errors. Some of the entries in the conditions table (which is a table of
4115 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4116 negation of where the condition IS allowed.
4118 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4120 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4121 header file does not have a version number, so I've had to invent a new
4122 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4123 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4124 radiusclient), but it was contributed by a Radius user.
4126 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4127 files or directories, for OpenSSL.
4129 PH/19 When an Exim process that is running as root has to create an Exim log
4130 file, it does so in a subprocess that runs as exim:exim so as to get the
4131 ownership right at creation (otherwise, other Exim processes might see
4132 the file with the wrong ownership). There was no test for failure of this
4133 fork() call, which would lead to the process getting stuck as it waited
4134 for a non-existent subprocess. Forks do occasionally fail when resources
4135 run out. I reviewed all the other calls to fork(); they all seem to check
4138 PH/20 When checking for unexpected SMTP input at connect time (before writing
4139 the banner), Exim was not dealing correctly with a non-positive return
4140 from the read() function. If the client had disconnected by this time,
4141 the result was a log entry for a synchronization error with an empty
4142 string after "input=" when read() returned zero. If read() returned -1
4143 (an event I could not check), uninitialized data bytes were printed.
4144 There were reports of junk text (parts of files, etc) appearing after
4147 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4149 PH/22 Added support for macro redefinition, and (re)definition in between
4150 driver and ACL definitions.
4152 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4153 forgetting to use the resulting value; it was using the unexpanded value.
4155 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4156 hadn't been configured. The fix is from Juergen Kreileder, who
4157 understands it better than I do:
4159 "Here's what I see happening with three configured cyrus_sasl
4160 authenticators configured (plain, login, cram-md5):
4162 On startup auth_cyrus_sasl_init() gets called for each of these.
4163 This means three calls to sasl_listmech() without a specified mech_list.
4164 => SASL tests which mechs of all available mechs actually work
4165 => three warnings about OTP not working
4166 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4168 With the patch, sasl_listmech() also gets called three times. But now
4169 SASL's mech_list option is set to the server_mech specified in the the
4170 authenticator. Or in other words, the answer from sasl_listmech()
4171 gets limited to just the mech you're testing for (which is different
4173 => the return list contains just 'plain' or 'login', 'cram-md5' or
4174 nothing depending on the value of ob->server_mech.
4176 I've just tested the patch: Authentication still works fine,
4177 unavailable mechs specified in the exim configuration are still
4178 caught, and the auth.log warnings about OTP are gone."
4180 PH/25 When debugging is enabled, the contents of the command line are added
4181 to the debugging output, even when log_selector=+arguments is not
4184 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4185 answer is "GNU", and only if the return is "GNU/something" is the answer
4188 PH/27 $acl_verify_message is now set immediately after the failure of a
4189 verification in an ACL, and so is available in subsequent modifiers. In
4190 particular, the message can be preserved by coding like this:
4192 warn !verify = sender
4193 set acl_m0 = $acl_verify_message
4195 Previously, $acl_verify_message was set only while expanding "message"
4196 and "log_message" when a very denied access.
4198 PH/28 Modified OS/os.c-Linux with
4200 -#ifndef OS_LOAD_AVERAGE
4201 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4203 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4204 nomenclature these days.)
4206 PH/29 Installed patch from the Sieve maintainer that adds the options
4207 sieve_useraddress and sieve_subaddress to the redirect router.
4209 PH/30 In these circumstances:
4210 . Two addresses routed to the same list of hosts;
4211 . First host does not offer TLS;
4212 . First host accepts first address;
4213 . First host gives temporary error to second address;
4214 . Second host offers TLS and a TLS session is established;
4215 . Second host accepts second address.
4216 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4217 and peerdn, if requested) that were in fact used only for the second
4220 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4221 attention to any local part prefix or suffix that was matched by the
4222 router that accepted the address. It now behaves in the same way as it
4223 does for delivery: the affixes are removed from the local part unless
4224 rcpt_include_affixes is set on the transport.
4226 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4227 timeout during the DATA phase of an incoming message.
4229 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4230 applied a patch sanctioned by the Sieve maintainer.
4232 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4233 the uid or gid is negative. A case of a negative gid caused this to be
4234 noticed. The fix allows for either to be negative.
4236 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4237 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4240 PH/36 The change PH/12 above was broken. Fixed it.
4242 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4243 the grounds that routing the same address twice would always produce the
4244 same answer. This might have been true once, but it is certainly no
4245 longer true now. Routing a child address may depend on the previous
4246 routing that produced that child. Some complicated redirection strategies
4247 went wrong when messages had multiple recipients, and made Exim's
4248 behaviour dependent on the order in which the addresses were given.
4250 I have moved the duplicate checking until after the routing is complete.
4251 Exim scans the addresses that are assigned to local and remote
4252 transports, and removes any duplicates. This means that more work will be
4253 done, as duplicates will always all be routed, but duplicates are
4254 presumably rare, so I don't expect this is of any significance.
4256 For deliveries to pipes, files, and autoreplies, the duplicate checking
4257 still happens during the routing process, since they are not going to be
4260 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4261 It corrects a timeout issue with spamd. This is Ian's comment: "The
4262 background is that sometimes spamd either never reads data from a
4263 connection it has accepted, or it never writes response data. The exiscan
4264 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4265 blindly assumes that writes won't block so it may never time out."
4267 PH/39 Allow G after quota size as well as K and M.
4269 PH/40 The value set for $authenticated_id in an authenticator may not contain
4270 binary zeroes or newlines because the value is written to log lines and
4271 to spool files. There was no check on this. Now the value is run through
4272 the string_printing() function so that such characters are converted to
4273 printable escape sequences.
4275 PH/41 $message_linecount is a new variable that contains the total number of
4276 lines in the message. Compare $body_linecount, which is the count for the
4279 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4280 bounce and delay warning messages, except in certain special cases, which
4283 (a) An SMTP error message from a remote host;
4284 (b) A message specified in a :fail: redirection;
4285 (c) A message specified in a "fail" command in a system filter;
4286 (d) A message specified in a FAIL return from the queryprogram router;
4287 (e) A message specified by the cannot_route_message router option.
4289 In these cases only, Exim does include the error details in bounce and
4290 warning messages. There are also a few cases where bland messages such
4291 as "unrouteable address" or "local delivery error" are given.
4293 PH/43 $value is now also set for the "else" part of a ${run expansion.
4295 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4296 being worked on, but at least Exim now implements the latest version to
4299 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4300 process to complete was treated as a delivery failure, a timeout while
4301 writing the message to the pipe was logged, but erroneously treated as a
4302 successful delivery. Such timeouts include transport filter timeouts. For
4303 consistency with the overall process timeout, these timeouts are now
4304 treated as errors, giving rise to delivery failures by default. However,
4305 there is now a new Boolean option for the pipe transport called
4306 timeout_defer, which, if set TRUE, converts the failures into defers for
4307 both kinds of timeout. A transport filter timeout is now identified in
4310 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4311 systems where "make" and "gmake" are different, calling "gmake" at top
4312 level broke things. I've arranged for the value of $(MAKE) to be passed
4313 from the Makefile to this script so that it can call the same version of
4317 A note about Exim versions 4.44 and 4.50
4318 ----------------------------------------
4320 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4321 changes of various kinds. As a consequence, a big documentation update was
4322 needed. This delayed the release for rather longer than seemed good, especially
4323 in the light of a couple of (minor) security issues. Therefore, the changes
4324 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4325 release. So 4.44 and 4.50 are in effect two different branches that both start
4328 I have left the 4.50 change log unchanged; it contains all the changes since
4329 4.43. The change log for 4.44 is below; many of its items are identical to
4330 those for 4.50. This seems to be the most sensible way to preserve the
4331 historical information.
4337 1. Minor wording change to the doc/README.SIEVE file.
4339 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4340 computation of the current number of files was incorrect.
4342 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4343 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4346 4. Give more explanation in the error message when the command for a transport
4347 filter fails to execute.
4349 5. There are several places where Exim runs a non-Exim command in a
4350 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4351 was being done only for the command run by the queryprogram router. It is
4352 now done for all such subprocesses. The other cases are: ${run, transport
4353 filters, and the commands run by the lmtp and pipe transports.
4355 6. Added CONFIGURE_GROUP build-time option.
4357 7. Some older OS have a limit of 256 on the maximum number of file
4358 descriptors. Exim was using setrlimit() to set 1000 as a large value
4359 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4360 systems. I've change it so that if it can't get 1000, it tries for 256.
4362 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4363 was an oversight, and furthermore, ever since the addition of extra
4364 controls (e.g. 4.43/32), the checks on when to allow different forms of
4365 "control" were broken. There should now be diagnostics for all cases when a
4366 control that does not make sense is encountered.
4368 9. Added the /retain_sender option to "control=submission".
4370 10. $recipients is now available in the predata ACL (oversight).
4372 11. Tidy the search cache before the fork to do a delivery from a message
4373 received from the command line. Otherwise the child will trigger a lookup
4374 failure and thereby defer the delivery if it tries to use (for example) a
4375 cached ldap connection that the parent has called unbind on.
4377 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4378 of $address_data from the recipient verification was clobbered by the
4379 sender verification.
4381 13. The value of address_data from a sender verification is now available in
4382 $sender_address_data in subsequent conditions in the ACL statement.
4384 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4386 15. Added a new option "connect=<time>" to callout options, to set a different
4389 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4390 was its contents. (It was OK if the option was not defined at all.)
4392 17. A "Completed" log line is now written for messages that are removed from
4393 the spool by the -Mrm option.
4395 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4396 information about exactly what failed.
4398 19. Added -dd to debug only the daemon process.
4400 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4401 handles timeouts, both on the server side and network timeouts. Renamed the
4402 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4404 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4405 It is now set to "smtps".
4407 22. $host_address is now set to the target address during the checking of
4408 ignore_target_hosts.
4410 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4411 being passed; this would have caused $sender_host_name to have been used if
4412 matching the list had actually called for a host name (not very likely,
4413 since this list is usually IP addresses). A host name is now passed as
4416 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4417 code that specifies a non-restarting handler (typically sigaction() in
4418 modern systems) in an attempt to fix a rare and obscure crash bug.
4420 25. Narrowed the window for a race in the daemon that could cause it to ignore
4421 SIGCHLD signals. This is not a major problem, because they are used only to
4422 wake it up if nothing else does.
4424 26. A malformed maildirsize file could cause Exim to calculate negative values
4425 for the mailbox size or file count. Odd effects could occur as a result.
4426 The maildirsize information is now recalculated if the size or filecount
4429 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4430 support for a long time. Removed HAVE_SYS_VFS_H.
4432 28. Installed the latest version of exipick from John Jetmore.
4434 29. In an address list, if the pattern was not a regular expression, an empty
4435 subject address (from a bounce message) matched only if the pattern was an
4436 empty string. Non-empty patterns were not even tested. This was the wrong
4437 because it is perfectly reasonable to use an empty address as part of a
4438 database query. An empty address is now tested by patterns that are
4439 lookups. However, all the other forms of pattern expect the subject to
4440 contain a local part and a domain, and therefore, for them, an empty
4441 address still always fails if the pattern is not itself empty.
4443 30. Exim went into a mad DNS loop when attempting to do a callout where the
4444 host was specified on an smtp transport, and looking it up yielded more
4445 than one IP address.
4447 31. Re-factored the code for checking spool and log partition space into a
4448 function that finds that data and another that does the check. The former
4449 is then used to implement four new variables: $spool_space, $log_space,
4450 $spool_inodes, and $log_inodes.
4452 32. The RFC2047 encoding function was originally intended for short strings
4453 such as real names; it was not keeping to the 75-character limit for
4454 encoded words that the RFC imposes. It now respects the limit, and
4455 generates multiple encoded words if necessary. To be on the safe side, I
4456 have increased the buffer size for the ${rfc2047: expansion operator from
4459 33. It is now permitted to omit both strings after an "if" condition; if the
4460 condition is true, the result is "true". As before, when the second string
4461 is omitted, a false condition yields an empty string. This makes it less
4462 cumbersome to write custom ACL and router conditions.
4464 34. Failure to deliver a bounce message always caused it to be frozen, even if
4465 there was an errors_to setting on the router. The errors_to setting is now
4468 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4469 canonical form (fully expanded) before being placed in
4470 $sender_host_address.
4472 36. The table in the code that translates DNS record types into text (T_A to
4473 "A" for instance) was missing entries for NS and CNAME. It is just possible
4474 that this could have caused confusion if both these types were looked up
4475 for the same domain, because the text type is used as part of Exim's
4476 per-process caching. But the chance of anyone hitting this buglet seems
4479 37. The dnsdb lookup has been extended in a number of ways.
4481 (1) There is a new type, "zns", which walks up the domain tree until it
4482 finds some nameserver records. It should be used with care.
4484 (2) There is a new type, "mxh", which is like "mx" except that it returns
4485 just the host names, not the priorities.
4487 (3) It is now possible to give a list of domains (or IP addresses) to be
4488 looked up. The behaviour when one of the lookups defers can be
4489 controlled by a keyword.
4491 (4) It is now possible to specify the separator character for use when
4492 multiple records are returned.
4494 38. The dnslists ACL condition has been extended: it is now possible to supply
4495 a list of IP addresses and/or domains to be looked up in a particular DNS
4498 39. Added log_selector=+queue_time_overall.
4500 40. When running the queue in the test harness, wait just a tad after forking a
4501 delivery process, to get repeatability of debugging output.
4503 41. Include certificate and key file names in error message when GnuTLS fails
4504 to set them up, because the GnuTLS error message doesn't include the name
4505 of the failing file when there is a problem reading it.
4507 42. Allow both -bf and -bF in the same test run.
4509 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4511 44. The "Exiscan patch" is now merged into the mainline Exim source.
4513 45. Sometimes the final signoff response after QUIT could fail to get
4514 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4515 < 0 before doing a fflush(). This bug looks as though it goes back to the
4516 introduction of TLS in release 3.20, but "sometimes" must have been rare
4517 because the tests only now provoked it.
4519 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4520 (this can affect the format of dates).
4522 47. exim_tidydb, when checking for the continued existence of a message for
4523 which it has found a message-specific retry record, was not finding
4524 messages that were in split spool directories. Consequently, it was
4525 deleting retry records that should have stayed in existence.
4527 48. Steve fixed some bugs in eximstats.
4529 49. The SPA authentication driver was not abandoning authentication and moving
4530 on to the next authenticator when an expansion was forced to fail,
4531 contradicting the general specification for all authenticators. Instead it
4532 was generating a temporary error. It now behaves as specified.
4534 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4535 (the order specifies the preference for clients). The order is now AES256,
4536 AES128, 3DES, ARCFOUR128.
4538 51. Small patch to Sieve code - explicitly set From: when generating an
4541 52. Exim crashed if a remote delivery caused a very long error message to be
4542 recorded - for instance if somebody sent an entire SpamAssassin report back
4543 as a large number of 550 error lines. This bug was coincidentally fixed by
4544 increasing the size of one of Exim's internal buffers (big_buffer) that
4545 happened as part of the Exiscan merge. However, to be on the safe side, I
4546 have made the code more robust (and fixed the comments that describe what
4549 53. Now that there can be additional text after "Completed" in log lines (if
4550 the queue_time_overall log selector is set), a one-byte patch to exigrep
4551 was needed to allow it to recognize "Completed" as not the last thing in
4554 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4555 patch that reportedly fixes this has been added. I am not expert enough to
4556 create a test for it. This is what the patch creator wrote:
4558 "I found a little strange behaviour of ldap code when working with
4559 Windows 2003 AD Domain, where users was placed in more than one
4560 Organization Units. When I tried to give exim partial DN, the exit code
4561 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4562 But simultaneously result of request was absolutely normal ldap result,
4563 so I produce this patch..."
4565 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4566 REFERENCE, so I have modified the code to exclude the patch when that macro
4569 55. Some experimental protocols are using DNS PTR records for new purposes. The
4570 keys for these records are domain names, not reversed IP addresses. The
4571 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4572 leaves it alone. Component reversal etc. now happens only for IP addresses.
4575 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4577 57. Double the size of the debug message buffer (to 2048) so that more of very
4578 long debug lines gets shown.
4580 58. The exicyclog utility now does better if the number of log files to keep
4581 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4583 59. Two changes related to the smtp_active_hostname option:
4585 (1) $smtp_active_hostname is now available as a variable.
4586 (2) The default for smtp_banner uses $smtp_active_hostname instead
4587 of $primary_hostname.
4589 60. The host_aton() function is supposed to be passed a string that is known
4590 to be a valid IP address. However, in the case of IPv6 addresses, it was
4591 not checking this. This is a hostage to fortune. Exim now panics and dies
4592 if the condition is not met. A case was found where this could be provoked
4593 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4594 components; fortuitously, this particular loophole had already been fixed
4595 by change 4.50/55 above.
4597 If there are any other similar loopholes, the new check in host_aton()
4598 itself should stop them being exploited. The report I received stated that
4599 data on the command line could provoke the exploit when Exim was running as
4600 exim, but did not say which command line option was involved. All I could
4601 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4602 running as the user.
4605 61. There was a buffer overflow vulnerability in the SPA authentication code
4606 (which came originally from the Samba project). I have added a test to the
4607 spa_base64_to_bits() function which I hope fixes it.
4610 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4611 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4613 63. The daemon start-up calls getloadavg() while still root for those OS that
4614 need the first call to be done as root, but it missed one case: when
4615 deliver_queue_load_max is set with deliver_drop_privilege. This is
4616 necessary for the benefit of the queue runner, because there is no re-exec
4617 when deliver_drop_privilege is set.
4619 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4620 This has been fixed.
4622 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4623 was in use, was not putting the data itself into the right store pool;
4624 consequently, it could be overwritten for a subsequent message in the same
4625 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4628 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4630 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4631 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4632 the function were treating the return as a boolean value, which happened to
4633 work because 0=false and not-0=true, but is not correct code.
4635 68. The host_aton() function was not handling scoped IPv6 addresses (those
4636 with, for example, "%eth0" on the end) correctly.
4638 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4639 negated items (that is, ~something) in unsigned ints. Some compilers
4640 apparently mutter when there is no cast.
4642 70. If an address verification called from an ACL failed, and did not produce a
4643 user-specific message (i.e. there was only a "system" message), nothing was
4644 put in $acl_verify_message. In this situation, it now puts the system
4647 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4648 session; change 4.31/43 added the unwanted input to the log line - except
4649 that it did not do this in the start of session case. It now does.
4651 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4652 This is wrong and can cause the other end to generate a synchronization
4653 error if it is another Exim or anything else that does the synchronization
4654 check. A QUIT command is no longer sent after a timeout.
4656 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4657 during host lookups.
4659 74. The defer_ok option of callout verification was not working if it was used
4660 when verifying addresses in header lines, that is, for this case:
4662 verify = header_sender/callout=defer_ok
4664 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4665 those file descriptors could be used for SMTP connections. If anything
4666 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4667 could be sent to the SMTP client, causing chaos. The daemon now opens
4668 stdin, stdout, and stderr to /dev/null when it puts itself into the
4671 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4672 log by default. The user can override this with suitable Perl magic.
4674 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4675 the log message when discard triggers, was not working for the DATA ACL or
4676 for the non-SMTP ACL.
4678 78. Error message wording change in sieve.c.
4680 79. If smtp_accept_max_per_host was set, the number of connections could be
4681 restricted to fewer than expected, because the daemon was trying to set up
4682 a new connection before checking whether the processes handling previous
4683 connections had finished. The check for completed processes is now done
4684 earlier. On busy systems, this bug wouldn't be noticed because something
4685 else would have woken the daemon, and it would have reaped the completed
4688 80. If a message was submitted locally by a user whose login name contained one
4689 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4690 It caused a spool format error. I have fixed the spool reading code. A
4691 related problem was that the "from" clause in the Received: line became
4692 illegal because of the space(s). It is now covered by ${quote_local_part.
4694 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4697 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4699 83. Patch from Sieve maintainer to fix unterminated string problem in
4700 "vacation" handling.
4702 84. Some minor changes to the Linux configuration files to help with other
4703 OS variants using glibc.
4705 85. One more patch for Sieve to update vacation handling to latest spec.
4708 ----------------------------------------------------
4709 See the note above about the 4.44 and 4.50 releases.
4710 ----------------------------------------------------
4716 1. Change 4.43/35 introduced a bug that caused file counts to be
4717 incorrectly computed when quota_filecount was set in an appendfile
4720 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4721 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4724 3. Give more explanation in the error message when the command for a transport
4725 filter fails to execute.
4727 4. There are several places where Exim runs a non-Exim command in a
4728 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4729 was being done only for the command run by the queryprogram router. It is
4730 now done for all such subprocesses. The other cases are: ${run, transport
4731 filters, and the commands run by the lmtp and pipe transports.
4733 5. Some older OS have a limit of 256 on the maximum number of file
4734 descriptors. Exim was using setrlimit() to set 1000 as a large value
4735 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4736 systems. I've change it so that if it can't get 1000, it tries for 256.
4738 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4739 was an oversight, and furthermore, ever since the addition of extra
4740 controls (e.g. 4.43/32), the checks on when to allow different forms of
4741 "control" were broken. There should now be diagnostics for all cases when a
4742 control that does not make sense is encountered.
4744 7. $recipients is now available in the predata ACL (oversight).
4746 8. Tidy the search cache before the fork to do a delivery from a message
4747 received from the command line. Otherwise the child will trigger a lookup
4748 failure and thereby defer the delivery if it tries to use (for example) a
4749 cached ldap connection that the parent has called unbind on.
4751 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4752 of $address_data from the recipient verification was clobbered by the
4753 sender verification.
4755 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4756 was its contents. (It was OK if the option was not defined at all.)
4758 11. A "Completed" log line is now written for messages that are removed from
4759 the spool by the -Mrm option.
4761 12. $host_address is now set to the target address during the checking of
4762 ignore_target_hosts.
4764 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4765 being passed; this would have caused $sender_host_name to have been used if
4766 matching the list had actually called for a host name (not very likely,
4767 since this list is usually IP addresses). A host name is now passed as
4770 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4771 code that specifies a non-restarting handler (typically sigaction() in
4772 modern systems) in an attempt to fix a rare and obscure crash bug.
4774 15. Narrowed the window for a race in the daemon that could cause it to ignore
4775 SIGCHLD signals. This is not a major problem, because they are used only to
4776 wake it up if nothing else does.
4778 16. A malformed maildirsize file could cause Exim to calculate negative values
4779 for the mailbox size or file count. Odd effects could occur as a result.
4780 The maildirsize information is now recalculated if the size or filecount
4783 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4784 support for a long time. Removed HAVE_SYS_VFS_H.
4786 18. Updated exipick to current release from John Jetmore.
4788 19. Allow an empty sender to be matched against a lookup in an address list.
4789 Previously the only cases considered were a regular expression, or an
4792 20. Exim went into a mad DNS lookup loop when doing a callout where the
4793 host was specified on the transport, if the DNS lookup yielded more than
4796 21. The RFC2047 encoding function was originally intended for short strings
4797 such as real names; it was not keeping to the 75-character limit for
4798 encoded words that the RFC imposes. It now respects the limit, and
4799 generates multiple encoded words if necessary. To be on the safe side, I
4800 have increased the buffer size for the ${rfc2047: expansion operator from
4803 22. Failure to deliver a bounce message always caused it to be frozen, even if
4804 there was an errors_to setting on the router. The errors_to setting is now
4807 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4808 canonical form (fully expanded) before being placed in
4809 $sender_host_address.
4811 24. Updated eximstats to version 1.33
4813 25. Include certificate and key file names in error message when GnuTLS fails
4814 to set them up, because the GnuTLS error message doesn't include the name
4815 of the failing file when there is a problem reading it.
4817 26. Expand error message when OpenSSL has problems setting up cert/key files.
4820 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4821 (this can affect the format of dates).
4823 28. exim_tidydb, when checking for the continued existence of a message for
4824 which it has found a message-specific retry record, was not finding
4825 messages that were in split spool directories. Consequently, it was
4826 deleting retry records that should have stayed in existence.
4828 29. eximstats updated to version 1.35
4829 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4830 1.35 - bugfix such that pie charts by volume are generated correctly
4832 30. The SPA authentication driver was not abandoning authentication and moving
4833 on to the next authenticator when an expansion was forced to fail,
4834 contradicting the general specification for all authenticators. Instead it
4835 was generating a temporary error. It now behaves as specified.
4837 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4838 (the order specifies the preference for clients). The order is now AES256,
4839 AES128, 3DES, ARCFOUR128.
4841 31. Small patch to Sieve code - explicitly set From: when generating an
4844 32. Exim crashed if a remote delivery caused a very long error message to be
4845 recorded - for instance if somebody sent an entire SpamAssassin report back
4846 as a large number of 550 error lines. This bug was coincidentally fixed by
4847 increasing the size of one of Exim's internal buffers (big_buffer) that
4848 happened as part of the Exiscan merge. However, to be on the safe side, I
4849 have made the code more robust (and fixed the comments that describe what
4852 33. Some experimental protocols are using DNS PTR records for new purposes. The
4853 keys for these records are domain names, not reversed IP addresses. The
4854 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4855 leaves it alone. Component reversal etc. now happens only for IP addresses.
4858 34. The host_aton() function is supposed to be passed a string that is known
4859 to be a valid IP address. However, in the case of IPv6 addresses, it was
4860 not checking this. This is a hostage to fortune. Exim now panics and dies
4861 if the condition is not met. A case was found where this could be provoked
4862 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4863 components; fortuitously, this particular loophole had already been fixed
4864 by change 4.50/55 or 4.44/33 above.
4866 If there are any other similar loopholes, the new check in host_aton()
4867 itself should stop them being exploited. The report I received stated that
4868 data on the command line could provoke the exploit when Exim was running as
4869 exim, but did not say which command line option was involved. All I could
4870 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4871 running as the user.
4874 35. There was a buffer overflow vulnerability in the SPA authentication code
4875 (which came originally from the Samba project). I have added a test to the
4876 spa_base64_to_bits() function which I hope fixes it.
4879 36. The daemon start-up calls getloadavg() while still root for those OS that
4880 need the first call to be done as root, but it missed one case: when
4881 deliver_queue_load_max is set with deliver_drop_privilege. This is
4882 necessary for the benefit of the queue runner, because there is no re-exec
4883 when deliver_drop_privilege is set.
4885 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4886 was in use, was not putting the data itself into the right store pool;
4887 consequently, it could be overwritten for a subsequent message in the same
4888 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4891 38. Sometimes the final signoff response after QUIT could fail to get
4892 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4893 < 0 before doing a fflush(). This bug looks as though it goes back to the
4894 introduction of TLS in release 3.20, but "sometimes" must have been rare
4895 because the tests only now provoked it.
4901 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4902 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4903 Now it returns an integer. A number of calls were still expecting a T/F
4904 return. Fortuitously, in all cases, the tests worked in OK situations,
4905 which is the norm. However, things would have gone wrong on any write
4906 failures on the smtp file descriptor. This function is used when sending
4907 messages over SMTP and also when doing verify callouts.
4909 2. When Exim is called to do synchronous delivery of a locally submitted
4910 message (the -odf or -odi options), it no longer closes stderr before doing
4913 3. Implemented the mua_wrapper option.
4915 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4917 5. Implemented the functions header_remove(), header_testname(),
4918 header_add_at_position(), and receive_remove_recipient(), and exported them
4921 6. If an ACL "warn" statement specified the addition of headers, Exim already
4922 inserted X-ACL-Warn: at the start if there was no header name. However, it
4923 was not making this test for the second and subsequent header lines if
4924 there were newlines in the string. This meant that an invalid header could
4925 be inserted if Exim was badly configured.
4927 7. Allow an ACL "warn" statement to add header lines at the start or after all
4928 the Received: headers, as well as at the end.
4930 8. Added the rcpt_4xx retry error code.
4932 9. Added postmaster_mailfrom=xxx to callout verification option.
4934 10. Added mailfrom=xxxx to the callout verification option, for verify=
4937 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4938 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4940 12. Inserted some casts to stop certain compilers warning when using pointer
4941 differences as field lengths or precisions in printf-type calls (mostly
4942 affecting debugging statements).
4944 13. Added optional readline() support for -be (dynamically loaded).
4946 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4947 same clock tick as a message's arrival, so that its received time was the
4948 same as the "first fail" time on the retry record, and that message
4949 remained on the queue past the ultimate address timeout, every queue runner
4950 would try a delivery (because it was past the ultimate address timeout) but
4951 after another failure, the ultimate address timeout, which should have then
4952 bounced the address, did not kick in. This was a "< instead of <=" error;
4953 in most cases the first failure would have been in the next clock tick
4954 after the received time, and all would be well.
4956 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4957 being recognized when the domain list was tested by the match_domain
4958 condition in an expansion string.
4960 16. Added the ${str2b64: operator.
4962 17. Exim was always calling setrlimit() to set a large limit for the number of
4963 processes, without checking whether the existing limit was already
4964 adequate. (It did check for the limit on file descriptors.) Furthermore,
4965 errors from getrlimit() and setrlimit() were being ignored. Now they are
4966 logged to the main and panic logs, but Exim does carry on, to try to do its
4967 job under whatever limits there are.
4969 18. Imported PCRE 5.0.
4971 19. Trivial typo in log message " temporarily refused connection" (the leading
4974 20. If the log selector return_path_on_delivery was set and an address was
4975 redirected to /dev/null, the delivery process crashed because it assumed
4976 that a return path would always be set for a "successful" delivery. In this
4977 case, the whole delivery is bypassed as an optimization, and therefore no
4980 21. Internal re-arrangement: the function for sending a challenge and reading
4981 a response while authentication was assuming a zero-terminated challenge
4982 string. It's now changed to take a pointer and a length, to allow for
4983 binary data in such strings.
4985 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4987 23. Exim was not respecting finduser_retries when seeking the login of the
4988 uid under which it was called; it was always trying 10 times. (The default
4989 setting of finduser_retries is zero.) Also, it was sleeping after the final
4990 failure, which is pointless.
4992 24. Implemented tls_on_connect_ports.
4994 25. Implemented acl_smtp_predata.
4996 26. If the domain in control=submission is set empty, Exim assumes that the
4997 authenticated id is a complete email address when it generates From: or
4998 Sender: header lines.
5000 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5001 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5002 chown and chgrp in /bin and hostname in /usr/bin.
5004 28. Exim was keeping the "process log" file open after each use, just as it
5005 does for the main log. This opens the possibility of it remaining open for
5006 long periods when the USR1 signal hits a daemon. Occasional processlog
5007 errors were reported, that could have been caused by this. Anyway, it seems
5008 much more sensible not to leave this file open at all, so that is what now
5011 29. The long-running daemon process does not normally write to the log once it
5012 has entered its main loop, and it closes the log before doing so. This is
5013 so that log files can straightforwardly be renamed and moved. However,
5014 there are a couple of unusual error situations where the daemon does write
5015 log entries, and I had neglected to close the log afterwards.
5017 30. The text of an SMTP error response that was received during a remote
5018 delivery was being truncated at 512 bytes. This is too short for some of
5019 the long messages that one sometimes sees. I've increased the limit to
5022 31. It is now possible to make retry rules that apply only when a message has a
5023 specific sender, in particular, an empty sender.
5025 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5026 it possible to be selective about when SMTP synchronization is enforced.
5028 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5030 32. Implemented hosts_connection_nolog.
5032 33. Added an ACL for QUIT.
5034 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5037 35. Added mailbox_size and mailbox_filecount to appendfile.
5039 36. Added control = no_multiline_responses to ACLs.
5041 37. There was a bug in the logic of the code that waits for the clock to tick
5042 in the case where the clock went backwards by a substantial amount such
5043 that the microsecond fraction of "now" was more than the microsecond
5044 fraction of "then" (but the whole seconds number was less).
5046 38. Added support for the libradius Radius client library this is found on
5047 FreeBSD (previously only the radiusclient library was supported).
5053 1. When certain lookups returned multiple values in the form name=value, the
5054 quoting of the values was not always being done properly. Specifically:
5055 (a) If the value started with a double quote, but contained no whitespace,
5057 (b) If the value contained whitespace other than a space character (i.e.
5058 tabs or newlines or carriage returns) it was not quoted.
5059 This fix has been applied to the mysql and pgsql lookups by writing a
5060 separate quoting function and calling it from the lookup code. The fix
5061 should probably also be applied to nisplus, ibase and oracle lookups, but
5062 since I cannot test any of those, I have not disturbed their existing code.
5064 2. A hit in the callout cache for a specific address caused a log line with no
5065 reason for rejecting RCPT. Now it says "Previous (cached) callout
5066 verification failure".
5068 3. There was an off-by-one bug in the queryprogram router. An over-long
5069 return line was truncated at 256 instead of 255 characters, thereby
5070 overflowing its buffer with the terminating zero. As well as fixing this, I
5071 have increased the buffer size to 1024 (and made a note to document this).
5073 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5074 when Exim is waiting for an SMTP response from a remote server, Exim
5075 restarts its select() call on the socket, thereby resetting its timeout.
5076 This is not a problem when such interrupts are rare. Somebody set up a cron
5077 job to run exiwhat every 2 minutes, which is less than the normal select()
5078 timeout (5 or 10 minutes). This meant that the select() timeout never
5079 kicked in because it was always reset. I have fixed this by comparing the
5080 time when an interrupt arrives with the time at the start of the first call
5081 to select(). If more time than the timeout has elapsed, the interrupt is
5082 treated as a timeout.
5084 5. Some internal re-factoring in preparation for the addition of Sieve
5085 extensions (by MH). In particular, the "personal" test is moved to a
5086 separate function, and given an option for scanning Cc: and Bcc: (which is
5087 not set for Exim filters).
5089 6. When Exim created an email address using the login of the caller as the
5090 local part (e.g. when creating a From: or Sender: header line), it was not
5091 quoting the local part when it contained special characters such as @.
5093 7. Installed new OpenBSD configuration files.
5095 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5096 try to make them clearer.
5098 9. Callout options, other than the timeout value, were being ignored when
5099 verifying sender addresses in header lines. For example, when using
5101 verify = header_sender/callout=no_cache
5103 the cache was (incorrectly) being used.
5105 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5106 only the Cygwin environment.
5108 11. When return_path_on_delivery was set as a log selector, if different remote
5109 addresses in the same message used different return paths and parallel
5110 remote delivery occurred, the wrong values would sometimes be logged.
5111 (Whenever a remote delivery process finished, the return path value from
5112 the most recently started remote delivery process was logged.)
5114 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5115 lines when AUTH and/or TLS are in use. This is the "received protocol"
5116 field. Exim used to use "asmtp" for authenticated SMTP, without any
5117 indication (in the protocol name) for TLS use. Now it follows the RFC and
5118 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5119 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5120 names appear in log lines as well as in Received: header lines.
5122 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5123 extensions, and comparison tests, and to fix some bugs.
5125 14. Changes to the "personal" filter test:
5127 (1) The test was buggy in that it was just doing the equivalent of
5128 "contains" tests on header lines. For example, if a user's address was
5129 anne@some.where, the "personal" test would incorrectly be true for
5131 To: susanne@some.where
5133 This test is now done by extracting each address from the header in turn,
5134 and checking the entire address. Other tests that are part of "personal"
5135 are now done using regular expressions (for example, to check local parts
5136 of addresses in From: header lines).
5138 (2) The list of non-personal local parts in From: addresses has been
5139 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5140 taken from the Sieve specification recommendations.
5142 (3) If the message contains any header line starting with "List-" it is
5143 treated as non-personal.
5145 (4) The test for "circular" in the Subject: header line has been removed
5146 because it now seems ill-conceived.
5148 15. Minor typos in src/EDITME comments corrected.
5150 16. Installed latest exipick from John Jetmore.
5152 17. If headers_add on a router specified a text string that was too long for
5153 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5154 of string_sprintf() is now avoided.
5156 18. $message_body_size was not set (it was always zero) when running the DATA
5157 ACL and the local_scan() function.
5159 19. For the "mail" command in an Exim filter, no default was being set for
5160 the once_repeat time, causing a random time value to be used if "once" was
5161 specified. (If the value happened to be <= 0, no repeat happened.) The
5162 default is now 0s, meaning "never repeat". The "vacation" command was OK
5163 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5164 (I found it when inspecting the code).
5166 20. There is now an overall timeout for performing a callout verification. It
5167 defaults to 4 times the callout timeout, which applies to individual SMTP
5168 commands during the callout. The overall timeout applies when there is more
5169 than one host that can be tried. The timeout is checked before trying the
5170 next host. This prevents very long delays if there are a large number of
5171 hosts and all are timing out (e.g. when the network connections are timing
5172 out). The value of the overall timeout can be changed by specifying an
5173 additional sub-option for "callout", called "maxwait". For example:
5175 verify = sender/callout=5s,maxwait=20s
5177 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5178 to the end before writing, but this should make it even safer).
5180 22. Exim was forgetting that it had advertised PIPELINING for the second and
5181 subsequent messages on an SMTP connection. It was also not resetting its
5182 memory on STARTTLS and an internal HELO.
5184 23. When Exim logs an SMTP synchronization error within a session, it now
5185 records whether PIPELINING has been advertised or not.
5187 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5188 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5189 rather than long int.
5191 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5193 26. Added the never_mail option to autoreply.
5199 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5200 crash if the getsockname() call failed; this can happen if a connection is
5201 closed very soon after it is established. The problem was simply in the
5202 order in which certain operations were done, causing Exim to try to write
5203 to the SMTP stream before it had set up the file descriptor. The bug has
5204 been fixed by making things happen in the correct order.
5210 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5211 before the connection was closed, thus losing the rejection response.
5213 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5214 some early Solaris releases, but causes trouble in current releases where
5215 socklen_t is defined.
5217 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5220 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5223 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5224 that releasing the top of it at the end releases what was used for sub-
5225 expansions (unless the block got too big). However, discard this block if
5226 the first thing is a variable or header, so that we can use its block when
5227 it is dynamic (useful for very large $message_headers, for example).
5229 6. Lookups now cache *every* query, not just the most recent. A new, separate
5230 store pool is used for this. It can be recovered when all lookup caches are
5231 flushed. Lookups now release memory at the end of their result strings.
5232 This has involved some general refactoring of the lookup sources.
5234 7. Some code has been added to the store_xxx() functions to reduce the amount
5235 of flapping under certain conditions.
5237 8. log_incoming_interface used to affect only the <= reception log lines. Now
5238 it causes the local interface and port to be added to several more SMTP log
5239 lines, for example "SMTP connection from", and rejection lines.
5241 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5243 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5245 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5246 could be overwritten at the end of the current message (or the start of a
5247 new message if it was set in a HELO ACL). The value is now preserved for
5248 the duration of the SMTP connection.
5250 12. If a transport had a headers_rewrite setting, and a matching header line
5251 contained an unqualified address, that address was qualified, even if it
5252 did not match any rewriting rules. The underlying bug was that the values
5253 of the flags that permit the existence of unqualified sender and recipient
5254 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5255 non-local messages, and by -bnq for local messages) were not being
5256 preserved with the message after it was received.
5258 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5259 "next input=" as part of the text comprising the host identity instead of
5260 the correct text. The code was using the same buffer for two different
5261 strings. However, depending on which order the printing function evaluated
5262 its arguments, the bug did not always show up. Under Linux, for example, my
5263 test suite worked just fine.
5265 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5266 doesn't work with some older versions of Perl. It has been changed to "my",
5267 which in any case is probably the better facility to use.
5269 15. A really picky compiler found some instances of statements for creating
5270 error messages that either had too many or two few arguments for the format
5273 16. The size of the buffer for calls to the DNS resolver has been increased
5274 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5275 for addresses that have a lot of PTR records. This alleviates a problem; it
5276 does not fully solve it.
5278 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5279 buffer now truncates the list and logs the incident, which is the same
5280 action as happens when Exim is looking up a host name and its aliases.
5281 Previously in this situation something unpredictable would happen;
5282 sometimes it was "internal error: store_reset failed".
5284 18. If a server dropped the connection unexpectedly when an Exim client was
5285 using GnuTLS and trying to read a response, the client delivery process
5286 crashed while trying to generate an error log message.
5288 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5289 string, for example:
5291 warn message = H1: something\nH2: something
5293 the text was added as a single header line from Exim's point of view
5294 though it ended up OK in the delivered message. However, searching for the
5295 second and subsequent header lines using $h_h2: did not work. This has been
5296 fixed. Similarly, if a system filter added multiple headers in this way,
5297 the routers could not see them.
5299 20. Expanded the error message when iplsearch is called with an invalid key to
5300 suggest using net-iplsearch in a host list.
5302 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5303 ACLs are no longer actually imposed (and a message to that effect is
5306 22. If a "gecos" field in a passwd entry contained escaped characters, in
5307 particular, if it contained a \" sequence, Exim got it wrong when building
5308 a From: or a Sender: header from that name. A second bug also caused
5309 incorrect handling when an unquoted " was present following a character
5310 that needed quoting.
5312 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5313 was not being matched caselessly.
5315 24. Arranged for all hyphens in the exim.8 source to be escaped with
5318 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5319 properly. Recipient callout cache records were still being keyed to include
5320 the sender, even when use_sender was set false. This led to far more
5321 callouts that were necessary. The sender is no longer included in the key
5322 when use_sender is false.
5324 26. Added "control = submission" modifier to ACLs.
5326 27. Added the ${base62d: operator to decode base 62 numbers.
5328 28. dnsdb lookups can now access SRV records.
5330 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5331 the configuration file.
5333 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5334 (-v) mode. This makes the output for a verbose queue run more intelligible.
5336 31. Added a use_postmaster feature to recipient callouts.
5338 32. Added the $body_zerocount variable, containing the number of binary zero
5339 bytes in the message body.
5341 33. The time of last modification of the "new" subdirectory is now used as the
5342 "mailbox time last read" when there is a quota error for a maildir
5345 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5347 35. Added +ignore_unknown as a special item in host lists.
5349 36. Code for decoding IPv6 addresses in host lists is now included, even if
5350 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5351 address was recognized as an IP address, but was then not correctly decoded
5352 into binary, causing unexpected and incorrect effects when compared with
5359 1. Very minor rewording of debugging text in manualroute to say "list of
5360 hosts" instead of "hostlist".
5362 2. If verify=header_syntax was set, and a header line with an unqualified
5363 address (no domain) and a large number of spaces between the end of the
5364 name and the colon was received, the reception process suffered a buffer
5365 overflow, and (when I tested it) crashed. This was caused by some obsolete
5366 code that should have been removed. The fix is to remove it!
5368 3. When running in the test harness, delay a bit after writing a bounce
5369 message to get a bit more predictability in the log output.
5371 4. Added a call to search_tidyup() just before forking a reception process. In
5372 theory, someone could use a lookup in the expansion of smtp_accept_max_
5373 per_host which, without the tidyup, could leave open a database connection.
5375 5. Added the variables $recipient_data and $sender_data which get set from a
5376 lookup success in an ACL "recipients" or "senders" condition, or a router
5377 "senders" option, similar to $domain_data and $local_part_data.
5379 6. Moved the writing of debug_print from before to after the "senders" test
5382 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5383 problems for message scanning, either using a data ACL, or using
5384 local_scan() because the Received: header was not generated till after they
5385 were called (in order to set the time as the time of reception completion).
5386 I have revised the way this works. The header is now generated after the
5387 body is received, but before the ACL or local_scan() are called. After they
5388 are run, the timestamp in the header is updated.
5394 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5395 before starting a queue runner without re-exec. This happened only when
5396 deliver_drop_privilege was set or when the Exim user was set to root. The
5397 effect of the bug was that timeouts during subsequent deliveries caused
5398 crashes instead of being properly handled. The handler is now left at its
5399 default (and expected) setting.
5401 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5402 message, again when deliver_drop_privilege is set or Exim is run as root.
5403 The bug described in (1) was not present in this case, but the tidying up
5404 of the other signals was missing. I have made the two cases consistent.
5406 3. The ignore_target_hosts setting on a manualroute router was being ignored
5407 for hosts that were looked up using the /MX notation.
5409 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5412 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5413 operated on the sender address. After changing the $sender_address to <>
5414 for the sender address verify, Exim was re-instated it as the original
5415 (before rewriting) address, but remembering that it had rewritten it, so it
5416 wasn't rewriting it again. This bug also had the effect of breaking the
5417 sender address verification caching when the sender address was rewritten.
5419 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5420 This has been changed so that if the ip literal address matches
5421 ignore_target_hosts, the router declines.
5423 7. Added expansion conditions match_domain, match_address, and match_local_
5424 part (NOT match_host).
5426 8. The placeholder for the Received: header didn't have a length field set.
5428 9. Added code to Exim itself and to exim_lock to test for a specific race
5429 condition that could lead to file corruption when using MBX delivery. The
5430 issue is with the lockfile that is created in /tmp. If this file is removed
5431 after a process has opened it but before that process has acquired a lock,
5432 there is the potential for a second process to recreate the file and also
5433 acquire a lock. This could lead to two Exim processes writing to the file
5434 at the same time. The added code performs the same test as UW imapd; it
5435 checks after acquiring the lock that its file descriptor still refers to
5436 the same named file.
5438 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5439 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5442 11. Added the smtp_active_hostname option. If used, this will typically be made
5443 to depend on the incoming interface address. Because $interface_address is
5444 not set up until the daemon has forked a reception process, error responses
5445 that can happen earlier (such as "too many connections") no longer contain
5448 12. If an expansion in a condition on a "warn" statement fails because a lookup
5449 defers, the "warn" statement is abandoned, and the next ACL statement is
5450 processed. Previously this caused the whole ACL to be aborted.
5452 13. Added the iplsearch lookup type.
5454 14. Added ident_timeout as a log selector.
5456 15. Added tls_certificate_verified as a log selector.
5458 16. Added a global option tls_require_ciphers (compare the smtp transport
5459 option of the same name). This controls incoming TLS connections.
5461 17. I finally figured out how to make tls_require_ciphers do a similar thing
5462 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5463 before starting the TLS session.
5465 18. Tabs are now shown as \t in -bP output.
5467 19. If the log selector return_path_on_delivery was set, Exim crashed when
5468 bouncing a message because it had too many Received: header lines.
5470 20. If two routers both had headers_remove settings, and the first one included
5471 a superfluous trailing colon, the final name in the first list and the
5472 first name in the second list were incorrectly joined into one item (with a
5473 colon in the middle).
5479 1. Added -C and -D options to the exinext utility, mainly to make it easier
5480 to include in the automated testing, but these could be helpful when
5481 multiple configurations are in use.
5483 2. The exinext utility was not formatting the output nicely when there was
5484 an alternate port involved in the retry record key, nor when there was a
5485 message id as well (for retries that were specific to a specific message
5486 and a specific host). It was also confused by IPv6 addresses, because of
5487 the additional colons they contain. I have fixed the IPv4 problem, and
5488 patched it up to do a reasonable job for IPv6.
5490 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5491 delivery, the log line now contains "pipelined" if PIPELINING was used.
5493 4. An SMTP transport process used to panic and die if the bind() call to set
5494 an explicit outgoing interface failed. This has been changed; it is now
5495 treated in the same way as a connect() failure.
5497 5. A reference to $sender_host_name in the part of a conditional expansion
5498 that was being skipped was still causing a DNS lookup. This no longer
5501 6. The def: expansion condition was not recognizing references to header lines
5502 that used bh_ and bheader_.
5504 7. Added the _cache feature to named lists.
5506 8. The code for checking quota_filecount in the appendfile transport was
5507 allowing one more file than it should have been.
5509 9. For compatibility with Sendmail, the command line option
5517 and sets the incoming protocol and host name (for trusted callers). The
5518 host name and its colon can be omitted when only the protocol is to be set.
5519 Note the Exim already has two private options, -pd and -ps, that refer to
5520 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5521 "s", but I don't think that's a major issue.
5523 10. A number of refactoring changes to the code, none of which should affect
5526 (a) The number of logging options was getting close to filling up the
5527 32-bit word that was used as a bit map. I have split them into two classes:
5528 those that are passed in the argument to log_write(), and those that are
5529 only ever tested independently outside of that function. These are now in
5530 separate 32-bit words, so there is plenty of room for expansion again.
5531 There is no change in the user interface or the logging behaviour.
5533 (b) When building, for example, log lines, the code previously used a
5534 macro that called string_cat() twice, in order to add two strings. This is
5535 not really sufficiently general. Furthermore, there was one instance where
5536 it was actually wrong because one of the argument was used twice, and in
5537 one call a function was used. (As it happened, calling the function twice
5538 did not affect the overall behaviour.) The macro has been replaced by a
5539 function that can join an arbitrary number of extra strings onto a growing
5542 (c) The code for expansion conditions now uses a table and a binary chop
5543 instead of a serial search (which was left over from when there were very
5544 few conditions). Also, it now recognizes conditions like "pam" even when
5545 the relevant support is not compiled in: a suitably worded error message is
5546 given if an attempt is made to use such a condition.
5548 11. Added ${time_interval:xxxxx}.
5550 12. A bug was causing one of the ddress fields not to be passed back correctly
5551 from remote delivery subprocesses. The field in question was not being
5552 subsequently used, so this caused to problems in practice.
5554 13. Added new log selectors queue_time and deliver_time.
5556 14. Might have fixed a bug in maildirsizefile handling that threw up
5557 "unexpected character" debug warnings, and recalculated the data
5558 unnecessarily. In any case, I expanded the warning message to give more
5561 15. Added the message "Restricted characters in address" to the statements in
5562 the default ACL that block characters like @ and % in local parts.
5564 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5565 Three changes have been made:
5567 (a) There was a serious bug; a negative response to MAIL caused the whole
5568 recipient domain to be cached as invalid, thereby blocking all messages
5569 to all local parts at the same domain, from all senders. This bug has
5570 been fixed. The domain is no longer cached after a negative response to
5571 MAIL if the sender used is not empty.
5573 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5576 (c) A new callout option, "use_sender" has been added for people who want
5577 the modified behaviour.
5583 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5586 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5587 indeed breaks things for older releases.
5589 3. Added additional logging to the case where there is a problem reading data
5590 from a filter that is running in a subprocess using a pipe, in order to
5591 try to track down a specific problem.
5593 4. Testing facility fudge: when running in the test harness and attempting
5594 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5595 getting "No route to host". Convert this to a timeout.
5597 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5600 6. Some OS don't have socklen_t but use size_t instead. This affects the
5601 fifth argument of getsockopt() amongst other things. This is now
5602 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5603 can be set for individual OS. I have set it for SunOS5, OSF1, and
5604 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5605 some earlier ones do not.
5607 7. Change 4.30/15 was not doing the test caselessly.
5609 8. The standard form for an IPv6 address literal was being rejected by address
5610 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5611 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5612 this, as well as the form without the "IPv6" on the front (but only when
5613 address literals are enabled, of course).
5615 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5617 10. Exim crashed if a message with an empty sender address specified by -f
5618 encountered a router with an errors_to setting. This could be provoked only
5619 by a command such as
5623 where an empty string was supplied; "<>" did not hit this bug.
5625 11. Installed PCRE release 4.5.
5627 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5628 remained set. It is now erased.
5630 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5631 times from message ids (which are base 36 rather than the normal 62).
5633 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5634 were being counted as actual protocol errors, and logged if the log
5635 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5636 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5637 and DATA following a set of rejected RCPTs do not count as protocol errors.
5638 In other words, Exim assumes they were pipelined, though this may not
5639 actually be the case. Of course, in all cases the client gets an
5640 appropriate error code.
5642 15. If a lookup fails in an ACL condition, a message about the failure may
5643 be available; it is used if testing the ACL cannot continue, because most
5644 such messages specify what the cause of the deferral is. However, some
5645 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5646 that caused an old message to be retained and used if a later statement
5647 caused a defer, replacing the real cause of the deferral.
5649 16. If an IP address had so many PTR records that the DNS lookup buffer
5650 was not large enough to hold them, Exim could crash while trying to process
5651 the truncated data. It now detects and logs this case.
5653 17. Further to 4.21/58, another change has been made: if (and only if) the
5654 first line of a message (the first header line) ends with CRLF, a bare LF
5655 in a subsequent header line has a space inserted after it, so as not to
5656 terminate the header.
5658 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5659 unnecessarily, used atoi() instead of strtol(), and didn't check the
5660 termination when getting file sizes from file names by regex.
5662 19. Completely re-implemented the support for maildirsize files, in the light
5663 of a number of problems with the previous contributed implementation
5664 (4.30/29). In particular:
5666 . If the quota is zero, the maildirsize file is maintained, but no quota is
5669 . If the maildir directory does not exist, it is created before any attempt
5670 to write a maildirsize file.
5672 . The quota value in the file is just a cache; if the quota is changed in
5673 the transport, the new value overrides.
5675 . A regular expression is available for excluding directories from the
5678 20. The autoreply transport checks the characters in options that define the
5679 message's headers; it allows continued headers, but it was checking with
5680 isspace() after an embedded newline instead of explicitly looking for a
5683 21. If all the "regular" hosts to which an address was routed had passed their
5684 expiry times, and had not reached their retry times, the address was
5685 bounced, even if fallback hosts were defined. Now Exim should go on to try
5688 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5689 equivalent code in the SMTP transport. Some hosts send humungous responses
5690 to HELO/EHLO, more than 1024 it seems.
5692 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5693 gives compiler warnings in some environments. I've now done it "properly",
5696 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5697 (because of problems with the built-in one) was declared to return uschar *
5698 instead of char *, causing compiler failure.
5700 25. Fixed a file descriptor leak when processing alias/forward files.
5702 26. Fixed a minor format string issue in dbfn.c.
5704 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5706 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5707 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5710 29. When a sender address is verified, it is cached, to save repeating the test
5711 when there is more than one recipient in a message. However, when the
5712 verification involves a callout, it is possible for different callout
5713 options to be set for different recipients. It is too complicated to keep
5714 track of this in the cache, so now Exim always runs a verification when a
5715 callout is required, relying on the callout cache for the optimization.
5716 The overhead is duplication of the address routing, but this should not be
5719 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5720 to be verified with callout=postmaster, and the main callout worked but the
5721 postmaster check failed, the verification correctly failed. However, if a
5722 subsequent RCPT command asked for sender verification *without* the
5723 postmaster check, incorrect caching caused this verification also to fail,
5726 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5727 it was not caching the DNS options (qualify_single, search_parents) that
5728 were used when the lookup failed. A subsequent lookup with different
5729 options therefore always gave the same answer, though there were cases
5730 where it should not have. (Example: a "domains = !$mx_any" option on a
5731 dnslookup router: the "domains" option is always processed without any
5732 widening, but the router might have qualify_single set.) Now Exim uses the
5733 cached value only when the same options are set.
5735 32. Added John Jetmore's "exipick" utility to the distribution.
5737 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5738 than a timeout (e.g. a certificate is required, and is not provided), an
5739 Exim server now closes the connection immediately. Previously it waited for
5740 the client to close - but if the client is SSL, it seems that they each
5741 wait for each other, leading to a delay before one of them times out.
5743 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5744 maintained 0.8.x compatibility because I don't think many are using it, and
5745 it is clearly obsolete.
5747 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5750 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5751 client certificate was expired. A simple patch fixes this, though I don't
5752 understand the full logic of why the verify callback is called multiple
5755 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5756 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5757 which causes problems with some clients (such as the Certicom SSL Plus
5758 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5759 disables the coutermeasure allowing Eudora to connect."
5761 38. Exim was not checking that a write() to a log file succeeded. This could
5762 lead to Bad Things if a log got too big, in particular if it hit a file
5763 size limit. Exim now panics and dies if it cannot write to a log file, just
5764 as it does if it cannot open a log file.
5766 39. Modified OS/Makefile-Linux so that it now contains
5768 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5770 The two -D definitions ensure that Exim is compiled with large file
5771 support, which makes it possible to handle log files that are bigger than
5774 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5775 instance) a domain was checked against a named list that involved a lookup,
5776 causing $domain_data to be set, then another domain was checked against the
5777 same list, then the first domain was re-checked, the value of $domain_data
5778 after the final check could be wrong. In particular, if the second check
5779 failed, it could be set empty. This bug probably also applied to
5782 41. The strip_trailing_dot option was not being applied to the address given
5783 with the -f command-line option.
5785 42. The code for reading a message's header from the spool was incrementing
5786 $received_count, but never initializing it. This meant that the value was
5787 incorrect (doubled) while delivering a message in the same process in which
5788 it was received. In the most common configuration of Exim, this never
5789 happens - a fresh exec is done - but it can happen when
5790 deliver_drop_privilege is set.
5792 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5793 it now includes up to 150 characters of the unexpected data in the log
5796 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5797 and building data strings. The size of both of these buffers was 10 000
5798 bytes - far larger than anybody would *ever* want, thought I. Needless to
5799 say, somebody hit the limit. I have increased the maximum line length to
5800 20 000 and the maximum data length of concatenated lines to 100 000. I have
5801 also fixed two bugs, because there was no checking on these buffers. Tsk,
5802 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5803 buffer is too small.
5805 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5806 lsearch lookups. Now it does.
5808 46. When parsing a route_list item in a manualroute router, a fixed-length
5809 buffer was used for the list of hosts. I made this 1024 bytes long,
5810 thinking that nobody would ever have a list of hosts that long. Wrong.
5811 Somebody had a whole pile of complicated expansion conditions, and the
5812 string was silently truncated, leading to an expansion error. It turns out
5813 that it is easier to change to an unlimited length (owing to other changes
5814 that have happened since this code was originally written) than to build
5815 structure for giving a limitation error. The length of the item that
5816 expands into the list of hosts is now unlimited.
5818 47. The lsearch lookup could not handle data where the length of text line was
5819 more than 4095 characters. Such lines were truncated, leading to shortened
5820 data being returned. It should now handle lines of any length.
5822 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5823 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5826 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5827 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5828 no longer happens in comment lines. A list of replacements is now placed
5829 at the head of all of the source files, except those whose only change is
5830 to replace PERL_COMMAND in the very first #! line.
5832 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5833 messages on the queue, with a bottom-up merge sort, using code contributed
5834 by Michael Haardt. This should make operations like -bp somewhat faster on
5835 large queues. It won't affect queue runners, except when queue_run_in_order
5838 51. Installed eximstats 1.31 in the distribution.
5840 52. Added support for SRV lookups to the dnslookup router.
5842 53. If an ACL referred to $message_body or $message_body_end, the value was not
5843 reset for any messages that followed in the same SMTP session.
5845 54. The store-handling optimization for building very long strings was not
5846 differentiating between the different store pools. I don't think this
5847 actually made any difference in practice, but I've tidied it.
5849 55. While running the routers to verify a sender address, $sender_address
5850 was still set to the sender address. This is wrong, because when routing to
5851 send a bounce to the sender, it would be empty. Therefore, I have changed
5852 it so that, while verifying a sender address, $sender_address is set to <>.
5853 (There is no change to what happens when verifying a recipient address.)
5855 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5856 target A or AAAA records (if not already returned) without resetting the
5857 qualify_single or search_parents options of the DNS resolver. These are
5858 inappropriate in this case because the targets of MX and SRV records must
5859 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5860 target that, when qualified, matched something in the local domain. These
5861 two options are now turned off when doing these lookups.
5863 57. It seems that at least some releases of Reiserfs (which does not have the
5864 concept of a fixed number of inodes) returns zero and not -1 for the
5865 number of available inodes. This interacted badly with check_spool_inodes,
5866 which assumed that -1 was the "no such thing" setting. What I have done is
5867 to check that the total number of inodes is greater than zero before doing
5868 the test of how many are available.
5870 58. When a "warn" ACL statement has a log_message modifier, the message is
5871 remembered, and not repeated. This is to avoid a lot of repetition when a
5872 message has many recipients that cause the same warning to be written.
5873 However, Exim was preserving the list of already written lines for an
5874 entire SMTP session, which doesn't seem right. The memory is now reset if a
5875 new message is started.
5877 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5878 debugging output unless log_rewrite was also set.
5880 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5881 of (int)(handle) when we know that handle contains (void *)(-1).
5883 61. The Exim daemon panic-logs an error return when it closes the incoming
5884 connection. However "connection reset by peer" seems to be common, and
5885 isn't really an error worthy of noting specially, so that particular error
5888 62. When Exim is trying to find all the local interfaces, it used to panic and
5889 die if the ioctl to get the interface flags failed. However, it seems that
5890 on at least one OS (Solaris 9) it is possible to have an interface that is
5891 included in the list of interfaces, but for which you get a failure error
5892 for this call. This happens when the interface is not "plumbed" into a
5893 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5894 failure of the "get flags" call assumes that the interface is down.
5896 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5897 makes life easier for people who are doing arithmetic on fields extracted
5898 from dates, where you often get leading zeros that should not be
5899 interpreted as octal.
5901 64. Added qualify_domain to the redirect router, to override the global
5904 65. If a pathologically long header line contained very many addresses (the
5905 report of this problem mentioned 10 000) and each of them was rewritten,
5906 Exim could use up a very large amount of memory. (It kept on making new
5907 copies of the header line as it rewrote, and never released the old ones.)
5908 At the expense of a bit more processing, the header rewriting function has
5909 been changed so that it no longer eats memory in this way.
5911 66. The generation of the Received: header has been moved from the time that a
5912 message starts to be received, to the time that it finishes. The timestamp
5913 in the Received: header should now be very close to that of the <= log
5914 line. There are two side-effects of this change:
5916 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5917 logged header lines no longer include the local Received: line, because
5918 it has not yet been created. The same applies to a copy of the message
5919 that is returned to a non-SMTP sender when a message is rejected.
5921 (b) When a filter file is tested using -bf, no additional Received: header
5922 is added to the test message. After some thought, I decided that this
5925 This change does not affect the value of $received_for. It is still set
5926 after address rewriting, but before local_scan() is called.
5928 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5930 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5931 gave an unhelpful panic error message, and a defer error. I have managed to
5932 change this behaviour so that it now rejects any supplied certificate,
5933 which seems right, as the list of acceptable certificates is empty.
5935 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5936 gave an unhelpful defer error. I have not managed to make this reject any
5937 supplied certificates, but the error message it gives is "no certificate
5938 supplied", which is not helpful.
5940 70. exigrep's output now also includes lines that are not associated with any
5941 message, but which match the given pattern. Implemented by a patch from
5942 Martin Sluka, which also tidied up the Perl a bit.
5944 71. Recipient callout verification, like sender verification, was using <> in
5945 the MAIL FROM command. This isn't really the right thing, since the actual
5946 sender may affect whether the remote host accepts the recipient or not. I
5947 have changed it to use the actual sender in the callout; this means that
5948 the cache record is now keyed on a recipient/sender pair, not just the
5949 recipient address. There doesn't seem to be a real danger of callout loops,
5950 since a callout by the remote host to check the sender would use <>.
5951 [SEE ABOVE: changed after hitting problems.]
5953 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5954 temporary errors. However, in the case of such a code being given after
5955 the end of a data transmission (i.e. after ".") Exim was failing to write
5956 a retry record for the message. (Yes, there was some broken host that was
5957 actually sending 8xx at this point.)
5959 73. An unknown lookup type in a host list could cause Exim to panic-die when
5960 the list was checked. (An example that provoked this was putting <; in the
5961 middle of a list instead of at the start.) If this happened during a DATA
5962 ACL check, a -D file could be left lying around. This kind of configuration
5963 error no longer causes Exim to die; instead it causes a defer error. The
5964 incident is still logged to the main and panic logs.
5966 74. Buglet left over from Exim 3 conversion. The message "too many messages
5967 in one connection" was written to the rejectlog but not the mainlog, except
5968 when address rewriting (yes!) was being logged.
5970 75. Added write_rejectlog option.
5972 76. When a system filter was run not as root (that is, when system_filter_user
5973 was set), the values of the $n variables were not being returned to the
5974 main process; thus, they were not subsequently available in the $sn
5977 77. Added +return_path_on_delivery log selector.
5979 78. A connection timeout was being treated differently from recipients deferred
5980 when testing hosts_max_try with a message that was older than the host's
5981 retry timeout. (The host should not be counted, thus allowing all hosts to
5982 be tried at least once before bouncing.) This may have been the cause of an
5983 occasionally reported bug whereby a message would remain on the queue
5984 longer than the retry timeout, but would be bounced if a delivery was
5985 forced. I say "may" because I never totally pinned down the problem;
5986 setting up timeout/retry tests is difficult. See also the next item.
5988 79. The ultimate address timeout was not being applied to errors that involved
5989 a combination of host plus message (for example, a timeout on a MAIL
5990 command). When an address resolved to a number of possible hosts, and they
5991 were not all tried for each delivery (e.g. because of hosts_max_try), a
5992 message could remain on the queue longer than the retry timeout.
5994 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5997 81. Fixed an obscure SMTP outgoing bug which required at least the following
5998 conditions: (a) there was another message waiting for the same server;
5999 (b) the server returned 5xx to all RCPT commands in the first message so
6000 that the message was not completed; (c) the server dropped the connection
6001 or gave a negative response to the RSET that Exim sends to abort the
6002 transaction. The observed case was a dropped connection after DATA that had
6003 been sent in pipelining mode. That is, the server had advertised PIPELINING
6004 but was not implementing it correctly. The effect of the bug was incorrect
6005 behaviour, such as trying another host, and this could lead to a crash.
6011 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6012 and daemon.c were passed as pointers to ints; they should have been
6013 pointers to socklen_t variables (which are typically unsigned ints).
6015 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6018 3. Fixed a really odd bug that affected only the testing scheme; patching a
6019 certain fixed string in the binary changed the value of another string that
6020 happened to be identical to the end of the original first string.
6022 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6023 name", it returns that address as the IP address. On some operating
6024 systems (e.g. Solaris), it also passes back the IP address string as the
6025 "host name". However, on others (e.g. Linux), it passes back an empty
6026 string. Exim wasn't checking for this, and was changing the host name to an
6027 empty string, assuming it had been canonicalized.
6029 5. Although rare, it is permitted to have more than one PTR record for a given
6030 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6031 all the names associated with an address, because they do in Solaris.
6032 However, it seems that they do not in Linux for data that comes from the
6033 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6034 I found this out when I moved to a new Linux workstation and tried to run
6035 the Exim test suite.
6037 To get round this problem I have changed the code so that it now does its
6038 own call to the DNS to look up PTR records when searching for a host name.
6039 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6040 addresses that are only in /etc/hosts are still found.
6042 This behaviour is, however, controlled by an option called host_lookup_
6043 order, which defaults to "bydns:byaddr". If people want to use the other
6044 order, or indeed, just use one or the other means of lookup, they can
6045 specify it in this variable.
6047 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6048 some operating systems, this comes back from gethostbyaddr() as an empty
6049 string, and this is what Exim used to test for. However, it seems that in
6050 other systems, "." is yielded. Exim now tests for this case too.
6052 7. The values of check_spool_space and check_log_space are now held internally
6053 as a number of kilobytes instead of an absolute number of bytes. If a
6054 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6055 kilobyte. This means that much larger values can be stored.
6057 8. Exim monitor: an attempt to get the action menu when not actually pointing
6058 at a message produces an empty menu entitled "No message selected". This
6059 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6060 no entries in it ("Shell widget menu has zero width and/or height"). So I
6061 have added a single, blank menu entry in this case.
6063 9. Added ${quote_local_part.
6065 10. MIME decoding is now applied to the contents of Subject: header lines when
6068 11. Now that a reference to $sender_host_address automatically causes a reverse
6069 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6070 host lookup before query-style lookups in lists that might use this
6071 variable. This has therefore been abolished, and the "net-" prefix is no
6072 longer necessary for query-style lookups.
6074 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6075 have been a typo for LFLAGS, so it has been changed.
6077 13. The install script calls Exim with "-C /dev/null" in order to find the
6078 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6079 to be output. However, since Exim outputs its version number before the
6080 error, it didn't break the script. It just looked ugly. I fixed this by
6081 always allowing "-C /dev/null" if the caller is root.
6083 14. Ignore overlarge ACL variable number when reading spool file - insurance
6084 against a later release with more variables having written the file.
6086 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6087 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6088 this, as well as the form without the "IPv6" on the front.
6090 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6091 OS/Makefile-Darwin file.
6093 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6094 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6096 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6097 variable. However, some OS use ints for inodes. I've added cast to long int
6098 to get rid of the compiler warning.
6100 19. I had forgotten to lock out "/../" in configuration file names when
6101 ALT_CONFIG_PREFIX was set.
6103 20. Routers used for verification do not need to specify transports. However,
6104 if such a router generated a host list, and callout was configured, Exim
6105 crashed, because it could not find a port number from the (non-existent)
6106 transport. It now assumes port 25 in this circumstance.
6108 21. Added the -t option to exigrep.
6110 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6111 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6112 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6113 them is set, LOOKUP_LSEARCH is forced.
6115 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6117 24. Added sender and host information to the "rejected by local_scan()" log
6118 line; previously there was no indication of these.
6120 25. Added .include_if_exists.
6122 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6123 Linux. It turns out that not all file systems support this. Apparently some
6124 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6125 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6126 not supported on the file descriptor, is now ignored when Exim is trying to
6127 sync a directory. This applies only to Linux.
6129 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6131 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6132 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6135 29. Added support for maildirsize files from supplied patch (modified a bit).
6137 30. The use of :fail: followed by an empty string could lead Exim to respond to
6138 sender verification failures with (e.g.):
6140 550 Verification failed for <xxx>
6141 550 Sender verify failed
6143 where the first response line was missing the '-' that indicates it is not
6144 the final line of the response.
6146 31. The loop for finding the name of the user that called Exim had a hardwired
6147 limit of 10; it now uses the value of finduser_retries, which is used for
6148 all other user lookups.
6150 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6153 33. Exim was neglecting to zero errno before one call of strtol() when
6154 expanding a string and expecting an integer value. On some systems this
6155 resulted in spurious "integer overflow" errors. Also, it was casting the
6156 result into an int without checking.
6158 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6159 did not work. The code looks as if it has *never* worked, though it appears
6160 to have been documented since at least release 1.62. I have made it work.
6162 35. The "timeout_DNS" error in retry rules, also documented since at least
6163 1.62, also never worked. As it isn't clear exactly what this means, and
6164 clearly it isn't a major issue, I have abolished the feature by treating it
6165 as "timeout", and writing a warning to the main and panic logs.
6167 36. The display of retry rules for -brt wasn't always showing the error code
6170 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6171 timeout_connect_A, timeout_connect_MX.
6173 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6174 to the empty sender.
6176 39. The daemon was not analysing the content of -oX till after it had closed
6177 stderr and disconnected from the controlling terminal. This meant that any
6178 syntax errors were only noted on the panic log, and the return code from
6179 the command was 0. By re-arranging the code a little, I've made the
6180 decoding happen first, so such errors now appear on stderr, and the return
6181 code is 1. However, the actual setting up of the sockets still happens in
6182 the disconnected process, so errors there are still only recorded on the
6185 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6186 connections (as happens on some IP stacks) was logged at start up time as
6187 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6188 it from "IPv6 and IPv4", which means that two separate sockets are being
6191 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6192 says whether AF_INET or AF_INET6 was passed as an argument.
6194 42. Exiwhat output was messed up when time zones were included in log
6197 43. Exiwhat now gives more information about the daemon's listening ports,
6198 and whether -tls-on-connect was used.
6200 44. The "port" option of the smtp transport is now expanded.
6202 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6203 silently ignored. Now an error message is written to the main and panic
6206 46. There's a new ACL modifier called "logwrite" which writes to a log file
6207 as soon as it is encountered.
6209 47. Added $local_user_uid and $local_user_gid at routing time.
6211 48. Exim crashed when trying to verify a sender address that was being
6214 49. Exim was recognizing only a space character after ".include". It now also
6215 recognizes a tab character.
6217 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6218 extracting the relevant information from the specification. The man page no
6219 longer contains scrambled data for the -d option, and I've added a section
6220 at the front about calling Exim under different names.
6222 51. Added "extra_headers" argument to the "mail" command in filter files.
6224 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6227 53. Installed eximstats 1.29.
6229 54. Added transport_filter_timeout as a generic transport option.
6231 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6232 Cc: header lines. This was required by RFC 822, but it not required by RFC
6235 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6236 incoming messages that did not have them. Now it does so only if the
6237 message originates locally, that is, if there is no associated remote host
6238 address. When Resent- header lines are present, this applies to the Resent-
6239 lines rather than the non-Resent- lines.
6241 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6242 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6244 58. Messages for configuration errors now include the name of the main
6245 configuration file - useful now that there may be more than one file in a
6246 list (.included file names were always shown).
6248 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6249 for those rare installations that do not start the daemon as root or run it
6250 setuid root. I've cut out the call to initgroups() if the daemon is not
6253 60. The Exim user and group can now be bound into the binary as text strings
6254 that are looked up at the start of Exim's processing.
6256 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6258 62. Added $mailstore_basename variable.
6260 63. Installed patch to sieve.c from Michael Haardt.
6262 64. When Exim failed to open the panic log after failing to open the main log,
6263 the original message it was trying to log was written to stderr and debug
6264 output, but if they were not available (the usual case in production), it
6265 was lost. Now it is written to syslog before the two lines that record the
6266 failures to open the logs.
6268 65. Users' Exim filters run in subprocesses under the user's uid. It is
6269 possible for a "deliver" command or an alias in a "personal" command to
6270 provoke an address rewrite. If logging of address rewriting is configured,
6271 this fails because the process is not running as root or exim. There may be
6272 a better way of dealing with this, but for the moment (because 4.30 needs
6273 to be released), I have disabled address rewrite logging when running a
6274 filter in a non-root, non-exim process.
6280 1. The buildconfig auxiliary program wasn't quoting the value set for
6281 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6282 not defined. This bug was masked in 4.22 by the effect that was fixed in
6285 2. Some messages that were rejected after a message id was allocated were
6286 shown as "incomplete" by exigrep. It no longer does this for messages that
6287 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6289 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6290 have allow_domain_literals set, the ID did not get logged in the <= line.
6291 Domain literals are now always recognized in Message-ID: header lines.
6293 4. The first argument for a ${extract expansion item is the key name or field
6294 number. Leading and trailing spaces in this item were not being ignored,
6295 causing some misleading effects.
6297 5. When deliver_drop_privilege was set, single queue runner processes started
6298 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6299 same command in the process it spins off) were not dropping privilege.
6301 6. When the daemon running as "exim" started a queue runner, it always
6302 re-executed Exim in the spun-off process. This is a waste of effort when
6303 deliver_drop_privilege is set. The new process now just calls the
6304 queue-runner function directly.
6310 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6313 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6314 ignored. Though the use of -oP was forcing the writing of a pid file, it
6315 was always written to the default place.
6317 3. If the message "no IP address found for host xxxx" is generated during
6318 incoming verification, it is now followed by identification of the incoming
6319 connection (so you can more easily find what provoked it).
6321 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6323 5. Added some features to "harden" Exim a bit more against certain attacks:
6325 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6326 be put in Local/Makefile. This is like the never_users runtime option,
6327 but it cannot be overridden. The default setting is "root".
6329 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6330 prefix string with which any file named in a -C command line option
6333 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6334 is retained for -C and -D only if the caller of Exim is root. Without
6335 it, the exim user may also use -C and -D and retain privilege.
6337 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6338 command line option is disabled.
6340 6. Macro names set by the -D option must start with an upper case letter, just
6341 like macro names defined in the configuration file.
6343 7. Added "dereference=" facility to LDAP.
6345 8. Two instances of the typo "uknown" in the source files are fixed.
6347 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6348 the Configure-Makefile script screwed up while processing it.
6350 10. Incorporated PCRE 4.4.
6352 11. The SMTP synchronization check was not operating right at the start of an
6353 SMTP session. For example, it could not catch a HELO sent before the client
6354 waited for the greeting. There is now a check for outstanding input at the
6355 point when the greeting is written. Because of the duplex, asynchronous
6356 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6357 way, but not yet received, when the check is performed.
6359 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6360 on TCP/IP sockets, because this apparently causes some broken clients to
6363 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6364 unchanged) from the Cygwin maintainer.
6366 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6367 was supported instead of testing for mbx. Effectively a typo.
6369 15. The spa authenticator server code was not checking that the input it
6370 received was valid base64.
6372 16. The debug output line for the "set" modifier in ACLs was not showing the
6373 name of the variable that was being set.
6375 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6377 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6378 lookup on its own. Something else was needed to trigger the lookup. For
6379 example, a match in host_lookup or the need for a host name in a host list.
6380 Now, if $sender_host_name is referenced and the host name has not yet been
6381 looked up, a lookup is performed. If the lookup fails, the variable remains
6382 empty, and $host_lookup_failed is set to "1".
6384 19. Added "eqi" as a case-independent comparison operator.
6386 20. The saslauthd authentication condition could segfault if neither service
6387 nor realm was specified.
6389 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6390 error message that was logged was misleading, and incoming SMTP
6391 connections were dropped. The message is now more accurate, and temporary
6392 errors are given to SMTP connections.
6394 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6395 (except RSET) with a 421 error, until QUIT is received. However, it was
6396 failing to send a response to QUIT.
6398 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6399 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6400 ACL is now run after the helo_try_verify_hosts code.
6402 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6403 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6404 case-independent, so other case variants are also recognized). Apparently
6405 some systems use these upper case variants.
6407 25. If more than two messages were waiting for the same host, and a transport
6408 filter was specified for the transport, Exim sent two messages over the
6409 same TCP/IP connection, and then failed with "socket operation on non-
6410 socket" when it tried to send the third.
6412 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6414 27. The extern definition of crypt16() in expand.c was not being excluded when
6415 the OS had its own crypt16() function.
6417 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6418 as a preferred synonym for return_size_limit, both as an option and as an
6421 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6423 30. Changed the default configuration ACL to relax the local part checking rule
6424 for addresses that are not in any local domains. For these addresses,
6425 slashes and pipe symbols are allowed within local parts, but the sequence
6426 /../ is explicitly forbidden.
6428 31. SPA server authentication was not clearing the challenge buffer before
6431 32. log_message in a "warn" ACL statement was writing to the reject log as
6432 well as to the main log, which contradicts the documentation and doesn't
6433 seem right (because no rejection is happening). So I have stopped it.
6435 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6436 However, I am unable to do any testing of this.
6438 34. Fixed an infelicity in the appendfile transport. When checking directories
6439 for a mailbox, to see if any needed to be created, it was accidentally
6440 using path names with one or more superfluous leading slashes; tracing
6441 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6443 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6444 contents are added to the log line that is written for every discarded
6445 recipient. (Previously a log_message setting was ignored.)
6447 36. The ${quote: operator now quotes the string if it is empty.
6449 37. The install script runs exim in order to find its version number. If for
6450 some reason other than non-existence or emptiness, which it checks, it
6451 could not run './exim', it was installing it with an empty version number,
6452 i.e. as "exim-". This error state is now caught, and the installation is
6455 38. An argument was missing from the function that creates an error message
6456 when Exim fails to connect to the socket for saslauthd authentication.
6457 This could cause Exim to crash, or give a corrupted message.
6459 39. Added isip, isip4, and isip6 to ${if conditions.
6461 40. The ACL variables $acl_xx are now saved with the message, and can be
6462 accessed later in routers, transports, and filters.
6464 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6465 strings in the file are not string-expanded.
6467 42. If a MAIL command specified a SIZE value that was too large to fit into an
6468 int variable, the check against message_size_limit failed. Such values are
6469 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6470 day this will have to be increased, but I don't think I want to be around
6471 when emails are that large.
6478 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6479 iconv() is not standard in FreeBSD.
6481 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6482 IPv6 enabled. The observed symptom was a segmentation fault on return from
6483 the function os_common_find_running_interfaces() in src/os.c.
6485 3. In the check_special_case() function in daemon.c I had used "errno" as an
6486 argument name, which causes warnings on some systems. This was basically a
6487 typo, since it was named "eno" in the comments!
6489 4. The code that waits for the clock to tick (at a resolution of some fraction
6490 of a second) so as to ensure message-id uniqueness was always waiting for
6491 at least one whole tick, when it could have waited for less. [This is
6492 almost certainly not relevant at current processor speeds, where it is
6493 unlikely to ever wait at all. But we try to future-proof.]
6495 5. The function that sleeps for a time interval that includes fractions of a
6496 second contained a race. It did not block SIGALRM between setting the
6497 timer, and suspending (a couple of lines later). If the interval was short
6498 and the sigsuspend() was delayed until after it had expired, the suspension
6499 never ended. On busy systems this could lead to processes getting stuck for
6502 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6503 process, before it forks any delivery processes. The open lookup caching
6504 mechanism meant that the open file or database connection was passed into
6505 the delivery process. The problem was that delivery processes always tidy
6506 up cached lookup data. This could cause a problem for the next delivery
6507 process started by the queue runner, because the external queue runner
6508 process does not know about the closure. So the next delivery process
6509 still has data in the lookup cache. In the case of a file lookup, there was
6510 no problem because closing a file descriptor in a subprocess doesn't affect
6511 the parent. However, if the lookup was caching a connection to a database,
6512 the connection was closed, and the second delivery process was likely to
6513 see errors such as "PGSQL: query failed: server closed the connection
6514 unexpectedly". The problem has been fixed by closing all cached lookups
6515 in a queue runner before running a delivery process.
6517 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6518 seem to have the "const" qualifier which it has on other OS. I've
6521 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6522 *of the same type* (client or server) with the same public name that an
6523 error should be diagnosed.
6525 9. When Exim looked up a host name for an IP address, but failed to find the
6526 original IP address when looking up the host name (a safety check), it
6527 output the message "<ip address> does not match any IP for NULL", which was
6528 confusing, to say the least. The bug was that the host name should have
6529 appeared instead of "NULL".
6531 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6532 user that is built into the binary, and the -C or -D options is used, root
6533 privilege is dropped before the configuration file is read. In addition,
6534 logging is switched to stderr instead of the normal log files. If the
6535 configuration then re-defines the Exim user, the unprivileged environment
6536 is probably not what is expected, so Exim logs a panic warning message (but
6539 However, if deliver_drop_privilege is set, the unprivileged state may well
6540 be exactly what is intended, so the warning has been cut out in that case,
6541 and Exim is allowed to try to write to its normal log files.
6547 1. smtp_return_error_details was not giving details for temporary sender
6548 or receiver verification errors.
6550 2. Diagnose a configuration error if two authenticators have the same public
6553 3. Exim used not to create the message log file for a message until the first
6554 delivery attempt. This could be confusing when incoming messages were held
6555 for policy or load reasons. The message log file is now created at the time
6556 the message is received, and an initial "Received" line is written to it.
6558 4. The automatically generated man page for command line options had a minor
6559 bug that caused no ill effects; however, a more serious problem was that
6560 the procedure for building the man page automatically didn't always
6561 operate. Consequently, release 4.20 contains an out-of-date version. This
6562 shouldn't happen again.
6564 5. When building Exim with embedded Perl support, the script that builds the
6565 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6566 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6568 6. The freeze_tell option was not being used for messages that were frozen on
6569 arrival, either by an ACL or by local_scan().
6571 7. Added the smtp_incomplete_transaction log selector.
6573 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6574 was accepting AUTH without a new EHLO.
6576 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6577 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6580 10. Logging of TCP/IP connections (when configured) now happens in the main
6581 daemon process instead of the child process, so that the TCP/IP connection
6582 count is more accurate (but it can never be perfect).
6584 11. The use of "drop" in a nested ACL was not being handled correctly in the
6585 outer ACL. Now, if condition failure induced by the nested "drop" causes
6586 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6587 or "require"), the connection is dropped.
6589 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6590 that yield "discard" can now be used with an "accept" or a "discard" verb,
6591 but an error is generated for any others (because I can't see a useful way
6592 to define what should happen).
6594 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6595 are now processed in the same way as lines in the Exim configuration file.
6596 In particular, continuation lines are supported.
6598 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6600 15. Added -ti meaning -t -i.
6602 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6603 domains, and warn by logging if others are found.
6605 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6606 returned from GIFCONF when Exim is trying to find the list of interfaces on
6607 a host. The code in os.c has been modified to copy each ifreq to an aligned
6608 structure in all cases.
6610 Also, in some cases, the returned ifreq's were being copied to a 'struct
6611 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6612 means the last couple of bytes of an IPv6 address could be chopped if the
6613 ifreq contained only a normal sockaddr (14 bytes storage).
6615 18. Named domain lists were not supported in the hosts_treat_as_local option.
6616 An entry such as +xxxx was not recognized, and was treated as a literal
6619 19. Ensure that header lines added by a DATA ACL are included in the reject log
6620 if the ACL subsequently rejects the message.
6622 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6623 MD5 (which is deprecated).
6625 21. When testing a filter file using -bf, Exim was writing a message when it
6626 took the sender from a "From " line in the message, but it was not doing so
6627 when it took $return_path from a Return-Path: header line. It now does.
6629 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6630 with a valid header line field name (a series of printing characters
6631 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6633 23. Changed "disc" in the source to "disk" to conform to the documentation and
6634 the book and for uniformity.
6636 24. Ignore Sendmail's -Ooption=value command line item.
6638 25. When execve() failed while trying to run a command in a pipe transport,
6639 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6640 could be confused with a return value of 69 from the command itself. This
6641 has been changed to 127, the value the shell returns if it is asked to run
6642 a non-existent command. The wording for the related log line suggests a
6643 non-existent command as the problem.
6645 26. If received_header_text expands to an empty string, do not add a Received:
6646 header line to the message. (Well, it adds a token one on the spool, but
6647 marks it "old" so that it doesn't get used or transmitted.)
6649 27. Installed eximstats 1.28 (addition of -nt option).
6651 28. There was no check for failure on the call to getsockname() in the daemon
6652 code. This can fail if there is a shortage of resources on the system, with
6653 ENOMEM, for example. A temporary error is now given on failure.
6655 29. Contrary to the C standard, it seems that in some environments, the
6656 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6657 program. Exim now does this explicitly; it affects the formatting of
6658 timestamps using strftime().
6660 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6661 complaints. I've now initialized all the variables, to avoid this.
6663 32. Header lines added by a system filter were not being "seen" during
6664 transport-time rewrites.
6666 33. The info_callback() function passed to OpenSSL is set up with type void
6667 (*)(SSL *, int, int), as described somewhere. However, when calling the
6668 function (actually a macro) that sets it up, the type void(*)() is
6669 expected. I've put in a cast to prevent warnings from picky compilers.
6671 34. If a DNS black list lookup found a CNAME record, but there were no A
6672 records associated with the domain it pointed at, Exim crashed.
6674 35. If a DNS black list lookup returned more than one A record, Exim ignored
6675 all but the first. It now scans all returned addresses if a particular IP
6676 value is being sought. In this situation, the contents of the
6677 $dnslist_value variable are a list of all the addresses, separated by a
6680 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6681 to accept a host name and all its aliases if the forward lookup for any of
6682 them yielded the IP address of the incoming connection. Now it accepts only
6683 those names whose forward lookup yields the correct IP address. Any other
6684 names are discarded. This closes a loophole whereby a rogue DNS
6685 administrator could create reverse DNS records to break through a
6686 wildcarded host restriction in an ACL.
6688 37. If a user filter or a system filter that ran in a subprocess used any of
6689 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6690 the wrong values were passed to the pipe command ($thisaddress had the
6691 value of $0, $0 had the value of $1, etc). This bug was introduced by
6692 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6694 38. Improved the line breaking for long SMTP error messages from ACLs.
6695 Previously, if there was no break point between 40 and 75 characters, Exim
6696 left the rest of the message alone. Two changes have been made: (a) I've
6697 reduced the minimum length to 35 characters; (b) if it can't find a break
6698 point between 35 and 75 characters, it looks ahead and uses the first one
6699 that it finds. This may give the occasional overlong line, but at least the
6700 remaining text gets split now.
6702 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6703 file descriptors might be low, and that setting 1000 would always raise it.
6704 It turns out that in some environments, the limit is already over 1000 and
6705 that lowering it causes trouble. So now Exim takes care not to decrease it.
6707 40. When delivering a message, the value of $return_path is set to $sender_
6708 address at the start of routing (routers may change the value). By an
6709 oversight, this default was not being set up when an address was tested by
6710 -bt or -bv, which affected the outcome if any router or filter referred to
6713 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6714 log only when "message" or "log_message" are set. However, if one of the
6715 conditions was an address verification, or a call to a nested ACL, the
6716 messages generated by the underlying test were being passed through. This
6717 no longer happens. The underlying message is available in $acl_verify_
6718 message for both "message" and "log_message" expansions, so it can be
6719 passed through if needed.
6721 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6722 new expansion $bh_ to give the encoded byte string without charset
6723 translation. Translation happens only if iconv() is available; HAVE_ICONV
6724 indicates this at build time. HEADERS_CHARSET gives the charset to
6725 translate to; headers_charset can change it in the configuration, and
6726 "headers charset" can change it in an individual filter file.
6728 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6729 that creates RFC 2047 encoded "words" labels them as that charset instead
6730 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6731 expansion operator; (ii) when Exim creates a From: line for a local
6732 message; (iii) when a header line is rewritten to include a "phrase" part.
6734 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6735 buggy, causing it to skip the first lines of messages whose message ID
6736 ended in 'D'. This would not have bitten before Exim release 4.14, because
6737 message IDs were unlikely to end in 'D' before then. The effect was to have
6738 incorrect size information for certain domains.
6740 45. #include "config.h" was missing at the start of the crypt16.c module. This
6741 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6744 46. If there was a timeout during a "random" callout check, Exim treated it as
6745 a failure of the random address, and carried on sending RSET and the real
6746 address. If the delay was just some slowness somewhere, the response to the
6747 original RCPT would be taken as a response to RSET and so on, causing
6748 mayhem of various kinds.
6750 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6751 when I implemented it. It didn't allow for the fact that some option values
6752 may legitimately be negative (e.g. size_addition), and it didn't even do
6753 the right test for positive values.
6755 48. Domain names in DNS records are case-independent. Exim always looks them up
6756 in lower case. Some resolvers return domain names in exactly the case they
6757 appear in the zone file, that is, they may contain uppercase letters. Not
6758 all resolvers do this - some return always lower case. Exim was treating a
6759 change of case by a resolver as a change of domain, similar to a widening
6760 of a domain abbreviation. This triggered its re-routing code and so it was
6761 trying to route what was effectively the same domain again. This normally
6762 caused routing to fail (because the router wouldn't handle the domain
6763 twice). Now Exim checks for this case specially, and just changes the
6764 casing of the domain that it ultimately uses when it transmits the message
6767 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6770 50. If a filter generated a file delivery with a non-absolute name (possible if
6771 no home directory exists for the router), the forbid_file option was not
6774 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6775 the existing equality matching.
6777 52. Exim was using ints instead of ino_t variables in some places where it was
6778 dealing with inode numbers.
6780 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6781 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6782 called TMPDIR, and if it finds it is different, it changes its value.
6784 54. The smtp_printf() function is now made available to local_scan() so
6785 additional output lines can be written before returning. There is also an
6786 smtp_fflush() function to enable the detection of a dropped connection.
6787 The variables smtp_input and smtp_batched_input are exported to
6790 55. Changed the default runtime configuration: the message "Unknown user"
6791 has been removed from the ACL, and instead placed on the localuser router,
6792 using the cannot_route_message feature. This means that any verification
6793 failures that generate their own messages won't get overridden. Similarly,
6794 the "Unrouteable address" message that was in the ACL for unverifiable
6795 relay addresses has also been removed.
6797 56. Added hosts_avoid_esmtp to the smtp transport.
6799 57. The exicyclog script was not checking for the esoteric option
6800 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6801 will work only if exicyclog is run under the appropriate euid.
6803 58. Following a discussion on the list, the rules by which Exim recognises line
6804 endings on incoming messages have been changed. The -dropcr and drop_cr
6805 options are now no-ops, retained only for backwards compatibility. The
6806 following line terminators are recognized: LF CRLF CR. However, special
6807 processing applies to CR:
6809 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6810 nor a local message in the state where . is a terminator.
6812 (ii) If a bare CR is encountered in a header line, an extra space is added
6813 after the line terminator so as not to end the header. The reasoning
6814 behind this is that bare CRs in header lines are most likely either
6815 to be mistakes, or people trying to play silly games.
6817 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6818 was being incorrectly given as 18 bytes larger than it should have been.
6819 This is a VOB (very old bug).
6821 60. This may never have affected anything current, but just in case it has:
6822 When the local host is found other than at the start of a list of hosts,
6823 the local host, those with the same MX, and any that follow, are discarded.
6824 When the list in question was part of a longer list of hosts, the following
6825 hosts (not currently being processed) were also being discarded. This no
6826 longer happens. I'm not sure if this situation could ever has previously
6829 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6832 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6833 header. This is something that is recommended in a new Internet Draft, and
6834 is something that is documented as being done by Sendmail. There are two
6835 possible values. For messages generated by the autoreply transport, Exim
6838 Auto-Submitted: auto-replied
6840 whereas for all other generated messages (e.g. bounces) it adds
6842 Auto-Submitted: auto-generated
6844 63. The "personal" condition in filters now includes a test for the
6845 Auto-Submitted: header. If it contains the string "auto-" the message it
6846 not considered personal.
6848 64. Added rcpt_include_affixes as a generic transport option.
6850 65. Added queue_only_override (default true).
6852 66. Added the syslog_duplication option.
6854 67. If what should have been the first header line of a message consisted of
6855 a space followed by a colon, Exim was mis-interpreting it as a header line.
6856 It isn't of course - it is syntactically invalid and should therefore be
6857 treated as the start of the message body. The misbehaviour could have
6858 caused a number of strange effects, including loss of data in subsequent
6859 header lines, and spool format errors.
6861 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6862 client host had authenticated. This control can now be exercised by an ACL
6863 for more flexibility.
6865 69. By default, callouts do not happen when testing with -bh. There is now a
6866 variant, -bhc, which does actually run the callout code, including
6867 consulting and updating the callout cache.
6869 70. Added support for saslauthd authentication, courtesy of Alexander
6872 71. If statvfs() failed on the spool or log directories while checking their
6873 size for availability, Exim confusingly gave the error "space shortage".
6874 Furthermore, in debugging mode it crashed with a floating point exception.
6875 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6876 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6877 serious problem, Exim now writes to the main and panic logs when this
6878 happens, with details of the failure. It then refuses to accept the
6879 incoming message, giving the message "spool directory problem" or "log
6880 directory problem" with a 421 code for SMTP messages.
6882 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6883 0, 1, and 2 exist, because some OS complain for execs without them (see
6884 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6885 descriptors. However, the code omitted to check that the open succeeded,
6886 causing mysterious errors if for some reason the permissions on /dev/null
6887 got screwed. Now Exim writes a message to the main and panic logs, and
6888 bombs out if it can't open /dev/null.
6890 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6891 interact so that it is all more flexible. It is supposed to remain
6892 backwards compatible. Also added extra_local_interfaces.
6894 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6895 to bomb out with an assertion failure - to the client this appears as a
6896 connection drop. This problem occurs in the part of the code that was taken
6897 from the Samba project. Fortunately, the assertion is in a very simple
6898 function, so I have fixed this by reproducing the function inline in the
6899 one place where it is called, and arranging for authentication to fail
6900 instead of killing the process with assert().
6902 75. The SPA client code was not working when the server requested OEM rather
6903 than Unicode encoding.
6905 76. Added code to make require_files with a specific uid setting more usable in
6906 the case where statting the file as root fails - usually a non-root-mounted
6907 NFS file system. When this happens and the failure is EACCES, Exim now
6908 forks a subprocess and does the per-uid checking as the relevant uid.
6910 77. Added process_log_path.
6912 78. If log_file_path was not explicitly set, a setting of check_log_space or
6913 check_log_inodes was ignored.
6915 79. If a space check for the spool or log partitions fails, the incident is now
6916 logged. Of course, in the latter case the data may get lost...
6918 80. Added the %p formatting code to string_format() so that it can be used to
6919 print addresses in debug_print(). Adjusted all the address printing in the
6920 debugging in store.c to use %p rather than %d.
6922 81. There was a concern that a line of code in smtp_in.c could overflow a
6923 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6924 initially expressed, the concern was not well-founded, because trailing
6925 spaces are removed early. However, if the trailing spaces were followed by
6926 a NULL, they did not get removed, so the overflow was possible. Two fixes
6929 (a) I re-wrote the offending code in a cleaner fashion.
6930 (b) If an incoming SMTP command contains a NULL character, it is rejected
6933 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6934 runs initgroups(), so that if the Exim user is in any additional groups,
6935 they will be used during message reception.
6941 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob