1 Change log file for Exim from version 4.21
2 ------------------------------------------
3 This document describes *changes* to previous versions, that might
4 affect Exim's operation, with an unchanged configuration file. For new
5 options, and new features, see the NewStuff file next to this ChangeLog.
8 Since Exim version 4.90
11 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
12 Previously only the last row was returned.
18 JH/01 Rework error string handling in TLS interface so that the caller in
19 more cases is responsible for logging. This permits library-sourced
20 string to be attached to addresses during delivery, and collapses
21 pairs of long lines into single ones.
23 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
24 during configuration. Wildcards are allowed and expanded.
26 JH/02 Rework error string handling in DKIM to pass more info back to callers.
27 This permits better logging.
29 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
30 do not close it down and have the child transport start it up again on
31 the passed-on TCP connection. Instead, proxy the child (and any
32 subsequent ones) for TLS via a unix-domain socket channel. Logging is
33 affected: the continued delivery log lines do not have any DNSSEC, TLS
34 Certificate or OCSP information. TLS cipher information is still logged.
36 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
37 identical IP addresses on different listening ports. Will also affect
40 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
41 add noisy ifdef guards to special-case this sillyness.
42 Patch from Bernd Kuhls.
44 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
45 than 255 are no longer allowed.
47 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
48 Disable the session-cache too, which might reduce our load. Since we
49 currrectly use a new context for every connection, both as server and
50 client, there is no benefit for these.
51 GnuTLS appears to not support tickets server-side by default (we don't
52 call gnutls_session_ticket_enable_server()) but client side is enabled
53 by default on recent versions (3.1.3 +) unless the PFS priority string
56 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
57 <https://reproducible-builds.org/specs/source-date-epoch/>.
59 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
60 the check for any unsuccessful recipients did not notice the limit, and
61 erroneously found still-pending ones.
63 JH/08 Pipeline CHUNKING command and data together, on kernels that support
64 MSG_MORE. Only in-clear (not on TLS connections).
66 JH/09 Avoid using a temporary file during transport using dkim. Unless a
67 transport-filter is involved we can buffer the headers in memory for
68 creating the signature, and read the spool data file once for the
69 signature and again for transmission.
71 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
72 4.77 as the kernel support then wasn't solid, having issues in 64bit
73 mode. Now, it's been long enough. Add support for FreeBSD also.
75 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
76 case where the routing stage had gathered several addresses to send to
77 a host before calling the transport for the first, we previously failed
78 to close down TLS in the old transport process before passing the TCP
79 connection to the new process. The new one sent a STARTTLS command
80 which naturally failed, giving a failed delivery and bloating the retry
81 database. Investigation and fix prototype from Wolfgang Breyha.
83 JH/12 Fix check on SMTP command input synchronisation. Previously there were
84 false-negatives in the check that the sender had not preempted a response
85 or prompt from Exim (running as a server), due to that code's lack of
86 awareness of the SMTP input buffering.
88 PP/04 Add commandline_checks_require_admin option.
89 Exim drops privileges sanely, various checks such as -be aren't a
90 security problem, as long as you trust local users with access to their
91 own account. When invoked by services which pass untrusted data to
92 Exim, this might be an issue. Set this option in main configuration
93 AND make fixes to the calling application, such as using `--` to stop
96 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
97 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
98 responses to those, into a single TLS record each way (this usually means
99 a single packet). As a side issue, smtp_enforce_sync now works on TLS
102 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
103 affects you only if you're dancing at the edge of the param size limits.
104 If you are, and this message makes sense to you, then: raise the
105 configured limit or use OpenSSL 1.1. Nothing we can do for older
108 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
109 cmdline element to get the documented default one. Previously it was
112 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
113 are unsuable for non-message-specific reasons - eg. connection timeout,
114 banner-time rejection.
116 JH/16 Fix logging of delivery remote port, when specified by router, under
119 PP/06 Repair manualroute's ability to take options in any order, even if one
120 is the name of a transport.
123 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
125 JH/17 Change the list-building routines interface to use the expanding-string
126 triplet model, for better allocation and copying behaviour.
128 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
129 Previously it was constructed the first time a possibly-matching string
130 was met in the configuration file input during startup; now it is done
133 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
134 compatible one, to avoid the (poorly documented) possibility of a config
135 file in the working directory redirecting the DB files, possibly correpting
136 some existing file. CVE-2017-10140 assigned for BDB.
138 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
139 cache-hot. Previously, although the result was properly cached, the
140 initial verify call returned a defer.
142 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
143 the main verify for receipient in uncached-mode.
145 JH/22 Retire historical build files to an "unsupported" subdir. These are
146 defined as "ones for which we have no current evidence of testing".
148 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
149 if present. Previously it was ignored.
151 JH/24 Start using specified-initialisers in C structure init coding. This is
152 a C99 feature (it's 2017, so now considered safe).
154 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
155 if was a fixed-sized field and bitmask ops via macros; it is now more
158 PP/07 GitHub PR 56: Apply MariaDB build fix.
159 Patch provided by Jaroslav Škarvada.
161 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
162 during Coverity cleanups [4.87 JH/47]
163 Diagnosis and fix provided by Michael Fischer v. Mollard.
165 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
166 the right size to place the terminating semicolon on its own folded
167 line, the header hash was calculated to an incorrect value thanks to
168 the (relaxed) space the fold became.
170 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
171 and confused the parent.
173 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
174 which could crash as a result. This could lead to undeliverable messages.
176 JH/28 Logging: "next input sent too soon" now shows where input was truncated
179 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
180 matters on fast-turnover and PID-randomising systems, which were getting
181 out-of-order delivery.
183 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
184 a possibly-overlapping copy. The symptom was that "Remote host closed
185 connection in response to HELO" was logged instead of the actual 4xx
188 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
189 Previously only that bufferd was discarded, resulting in SYMTP command
192 JH/32 DKIM: when a message has multiple signatures matching an identity given
193 in dkim_verify_signers, run the dkim acl once for each. Previously only
194 one run was done. Bug 2189.
196 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
197 "panic the current process" to "deliberately defer". The panic log is
198 still written with the problem list name; the mail and reject logs now
199 get a temp-reject line for the message that was being handled, saying
200 something like "domains check lookup or other defer". The SMTP 451
201 message is still "Temporary local problem".
203 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
204 A crafted sequence of BDAT commands could result in in-use memory beeing
205 freed. CVE-2017-16943.
207 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
208 from SMTP input. Previously it was always done; now only done for DATA
209 and not BDAT commands. CVE-2017-16944.
211 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
212 to the message (such as an overlong header line). Previously this was
213 not done and we did not exit BDAT mode. Followon from the previous item
214 though a different problem.
220 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
221 than -2003 did; needs libidn2 in addition to libidn.
223 JH/02 The path option on a pipe transport is now expanded before use.
225 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
226 Patch provided by "Björn", documentation fix added too.
228 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
229 missing a wire-to-host endian conversion.
231 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
232 close after a BDAT command line could be taken as a following command,
233 giving a synch failure. Fix by only checking for synch immediately
234 before acknowledging the chunk.
236 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
237 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
238 macro. Patches provided by Josh Soref.
240 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
241 Previously we did not; the RFC seems ambiguous and VRFY is not listed
242 by IANA as a service extension. However, John Klensin suggests that we
245 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
246 the dkim code may be unix-mode line endings rather than smtp wire-format
247 CRLF, so prepend a CR to any bare LF.
249 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
250 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
252 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
253 the first were themselves being wrongly included in the feed into dkim
254 processing; with most chunk sizes in use this resulted in an incorrect
255 body hash calculated value.
257 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
258 DKIM signature block, for verification. Although advised against by
259 standards it is specifically not ruled illegal.
261 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
263 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
264 missing a body hash (the bh= tag).
266 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
267 It seems that HAProxy sends the Proxy Protocol information in clear and
268 only then does a TLS startup, so do the same.
270 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
271 TCP connections (such as for Spamd) unless the daemon successfully set
272 Fast Open mode on its listening sockets. This fixes breakage seen on
273 too-old kernels or those not configured for Fast Open, at the cost of
274 requiring both directions being enabled for TFO, and TFO never being used
275 by non-daemon-related Exim processes.
277 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
278 endings, at least on the first header line. Try to canonify any that get
279 past that check, despite the cost.
281 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
282 now limited to an arbitrary five deep, while parsing addresses with the
283 strip_excess_angle_brackets option enabled.
285 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
286 instead leave the unprompted TLS handshake in socket buffer for the
287 TLS library to consume.
289 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
291 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
293 JH/16 Drop variables when they go out of scope. Memory management drops a whole
294 region in one operation, for speed, and this leaves assigned pointers
295 dangling. Add checks run only under the testsuite which checks all
296 variables at a store-reset and panics on a dangling pointer; add code
297 explicitly nulling out all the variables discovered. Fixes one known
298 bug: a transport crash, where a dangling pointer for $sending_ip_address
299 originally assigned in a verify callout, is re-used.
301 PP/06 Drop '.' from @INC in various Perl scripts.
303 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
305 PP/08 Reduce a number of compilation warnings under clang; building with
306 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
307 should be warning-free.
309 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
311 HS/01 Fix portability problems introduced by PP/08 for platforms where
312 realloc(NULL) is not equivalent to malloc() [SunOS et al].
314 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
315 chunk. This allows us to accept broken chunked messages. We need a more
316 general solution here.
318 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
319 already-broken messages in the queue.
321 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
323 JH/19 Fix reference counting bug in routing-generated-address tracking.
329 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
330 supports it and a size is available (ie. the sending peer gave us one).
332 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
333 years of being deprecated). The replacements are the ACLs
334 acl_smtp_mime and acl_not_smtp_mime.
336 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
337 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
338 or even an in-clear connection were permitted. Now, if the host lookup
339 was dnssec and dane was requested then the host is only used if the
340 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
341 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
342 if one fails this test.
343 This means that a poorly-configured remote DNS will make it incommunicado;
344 but it protects against a DNS-interception attack on it.
346 JH/04 Bug 1810: make continued-use of an open smtp transport connection
347 non-noisy when a race steals the message being considered.
349 JH/05 If main configuration option tls_certificate is unset, generate a
350 self-signed certificate for inbound TLS connections.
352 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
353 in rewrites and routers.
355 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
356 and logged a warning sing 4.83; now they are a configuration file error.
358 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
359 (lacking @domain). Apply the same qualification processing as RCPT.
361 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
363 JH/10 Support ${sha256:} applied to a string (as well as the previous
366 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
367 a cutthrough deliver is pending, as we always want to make a connection.
368 This also avoids re-routing the message when later placing the cutthrough
369 connection after a verify cache hit.
370 Do not update it with the verify result either.
372 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
373 when routing results in more than one destination address.
375 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
376 signing (which inhibits the cutthrough capability). Previously only
377 the presence of an option was tested; now an expansion evaluating as
378 empty is permissible (obviously it should depend only on data available
379 when the cutthrough connection is made).
381 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
382 the relevant preceding SMTP command did not note the pipelining mode.
384 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
385 Previously they were not counted.
387 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
388 as one having no matching records. Previously we deferred the message
389 that needed the lookup.
391 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
392 distinguished as "(=".
394 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
395 for missing MX records. Previously it only worked for missing A records.
397 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
399 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
400 after the data-go-ahead and data-ack. Patch from Jason Betts.
402 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
403 even for a "none" policy. Patch from Tony Meyer.
405 JH/22 Fix continued use of a connection for further deliveries. If a port was
406 specified by a router, it must also match for the delivery to be
409 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
410 When one of the recipients of a message was unsuitable for the connection
411 (has no matching addresses), we lost track of needing to mark it
412 deferred. As a result mail would be lost.
414 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
416 JH/25 Decoding ACL controls is now done using a binary search; the source code
417 takes up less space and should be simpler to maintain. Merge the ACL
418 condition decode tables also, with similar effect.
420 JH/26 Fix problem with one_time used on a redirect router which returned the
421 parent address unchanged. A retry would see the parent address marked as
422 delivered, so not attempt the (identical) child. As a result mail would
425 JH/27 Fix a possible security hole, wherein a process operating with the Exim
426 UID can gain a root shell. Credit to http://www.halfdog.net/ for
427 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
430 JH/28 Enable {spool,log} filesystem space and inode checks as default.
431 Main config options check_{log,spool}_{inodes,space} are now
432 100 inodes, 10MB unless set otherwise in the configuration.
434 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
435 Previously it only applied to the main-section connection policy
438 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
440 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
441 by me. Added RFC7919 DH primes as an alternative.
443 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
444 are not in the system include path.
446 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
447 GnuTLS, when a session startup failed (eg because the client disconnected)
448 Exim did stdio operations after fclose. This was exposed by a recent
449 change which nulled out the file handle after the fclose.
451 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
452 signed directly by the cert-signing cert, rather than an intermediate
453 OCSP-signing cert. This is the model used by LetsEncrypt.
455 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
457 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
458 an incoming connection.
460 HS/02 Bug 1802: Do not half-close the connection after sending a request
463 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
464 fallback to "prime256v1".
466 JH/34 SECURITY: Use proper copy of DATA command in error message.
467 Could leak key material. Remotely exploitable. CVE-2016-9963.
473 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
474 and 3.4.4 - once the server is enabled to respond to an OCSP request
475 it does even when not requested, resulting in a stapling non-aware
476 client dropping the TLS connection.
478 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
479 support variable-length bit vectors. No functional change.
481 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
482 The I= interface field on outgoing lines is now after the H= remote
483 host field, same as incoming lines. There is a separate
484 outgoing_interface log selector which allows you to disable the
487 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
488 If not running log_selector +smtp_connection the mainlog would be held
489 open indefinitely after a "too many connections" event, including to a
490 deleted file after a log rotate. Leave the per net connection logging
491 leaving it open for efficiency as that will be quickly detected by the
492 check on the next write.
494 HS/01 Bug 1671: Fix post transport crash.
495 Processing the wait-<transport> messages could crash the delivery
496 process if the message IDs didn't exist for some reason. When
497 using 'split_spool_directory=yes' the construction of the spool
498 file name failed already, exposing the same netto behaviour.
500 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
501 mime_regex ACL conditions.
503 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
504 to DSN fail messages (bounces): remote IP, remote greeting, remote response
505 to HELO, local diagnostic string.
507 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
508 log line to debug. Even when configured with a tls authenticator many
509 client connections are expected to not authenticate in this way, so
510 an authenticate fail is not an error.
512 HS/02 Add the Exim version string to the process info. This way exiwhat
513 gives some more detail about the running daemon.
515 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
516 matter for fast-change records such as DNSBLs.
518 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
519 retry record, even if constant. There may be multiple transports with
520 different interface settings and the retry behaviour needs to be kept
523 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
525 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
527 JH/10 Bug 840: fix log_defer_output option of pipe transport
529 JH/11 Bug 830: use same host for all RCPTS of a message, even under
530 hosts_randomize. This matters a lot when combined with mua_wrapper.
532 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
533 ${quote_pgsql:<string>} operator.
535 JH/13 Bug 1708: avoid misaligned access in cached lookup.
537 JH/14 Change header file name for freeradius-client. Relevant if compiling
538 with Radius support; from the Gentoo tree and checked under Fedora.
540 JH/15 Bug 1712: Introduce $prdr_requested flag variable
542 JH/16 Bug 1714: Permit an empty string as expansion result for transport
543 option transport_filter, meaning no filtering.
545 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
547 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
548 defaults to "*" (all hosts). The variable is now available when not built
549 with TLS, default unset, mainly to enable keeping the testsuite sane.
550 If a server certificate is not supplied (via tls_certificate) an error is
551 logged, and clients will find TLS connections fail on startup. Presumably
552 they will retry in-clear.
553 Packagers of Exim are strongly encouraged to create a server certificate
554 at installation time.
556 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
557 with the $config_file variable.
559 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
560 in transport context, after the attempt, and per-recipient. The latter type
561 is per host attempted. The event data is the error message, and the errno
562 information encodes the lookup type (A vs. MX) used for the (first) host,
563 and the trailing two digits of the smtp 4xx response.
565 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
566 to write to mainlog (or rejectlog, paniclog) in the window between file
567 creation and permissions/ownership being changed. Particularly affects
568 installations where exicyclog is run as root, rather than exim user;
569 result is that the running daemon panics and dies.
571 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
573 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
574 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
575 "pri" and "weight". Note that the previous implicit priority given by the
576 list order is no longer honoured.
578 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
581 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
582 by defining SUPPORT_SOCKS.
584 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
585 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
586 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
587 variables are renamed to proxy_{local,external}_{address,port}.
589 JH/27 Move Internationalisation support from Experimental to mainline, enabled
590 for a build by defining SUPPORT_I18N
592 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
593 of the query string, and make ${quote_redis:} do that quoting.
595 JH/29 Move Events support from Experimental to mainline, enabled by default
596 and removable for a build by defining DISABLE_EVENT.
598 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
600 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
601 cached by the daemon.
603 JH/32 Move Redis support from Experimental to mainline, enabled for a build
604 by defining LOOKUP_REDIS. The libhiredis library is required.
606 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
607 keys are given for lookup.
609 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
610 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
611 only supported when built with TLS support. The PolarSSL SHA routines
612 are still used when the TLS library is too old for convenient support.
614 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
615 openssl_options), for security. OpenSSL forces this from version 1.1.0
616 server-side so match that on older versions.
618 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
619 allocation for $value could be released as the expansion processing
620 concluded, but leaving the global pointer active for it.
622 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
623 and to use the domains and local_parts ACL conditions.
625 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
626 incorrectly not doubled on cutthrough transmission, hence seen as a
627 body-termination at the receiving system - resulting in truncated mails.
628 Commonly the sender saw a TCP-level error, and retransmitted the message
629 via the normal store-and-forward channel. This could result in duplicates
630 received - but deduplicating mailstores were liable to retain only the
631 initial truncated version.
633 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
635 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
637 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
638 we're in there, support oversigning also; bug 1309.
640 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
642 HS/04 Add support for keep_environment and add_environment options.
644 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
645 either intentional arithmetic overflow during PRNG, or testing config-
648 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
649 delivery resulted in actual delivery. Cancel cutthrough before DATA
652 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
653 rejects a recipient: pass the reject to the originator.
655 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
656 Many were false-positives and ignorable, but it's worth fixing the
659 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
660 for the new environment-manipulation done at startup. Move the routines
661 from being local to tls.c to being global via the os.c file.
663 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
664 an extract embedded as result-arg for a map, the first arg for extract
665 is unavailable so we cannot tell if this is a numbered or keyed
666 extraction. Accept either.
672 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
675 JH/02 The smtp transport option "multi_domain" is now expanded.
677 JH/03 The smtp transport now requests PRDR by default, if the server offers
680 JH/04 Certificate name checking on server certificates, when exim is a client,
681 is now done by default. The transport option tls_verify_cert_hostnames
682 can be used to disable this per-host. The build option
683 EXPERIMENTAL_CERTNAMES is withdrawn.
685 JH/05 The value of the tls_verify_certificates smtp transport and main options
686 default to the word "system" to access the system default CA bundle.
687 For GnuTLS, only version 3.0.20 or later.
689 JH/06 Verification of the server certificate for a TLS connection is now tried
690 (but not required) by default. The verification status is now logged by
691 default, for both outbound TLS and client-certificate supplying inbound
694 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
697 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
698 Status Notification (bounce) messages are now MIME format per RFC 3464.
699 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
700 under the control of the dsn_advertise_hosts option, and routers may
701 have a dsn_lasthop option.
703 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
704 default, modifiable by a malware= option. The list separator for
705 the options can now be changed in the usual way. Bug 68.
707 JH/10 The smtp_receive_timeout main option is now expanded before use.
709 JH/11 The incoming_interface log option now also enables logging of the
710 local interface on delivery outgoing connections.
712 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
713 if the interface and destination host and port all match.
715 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
718 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
719 Patch from Andrew Lewis.
721 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
722 now supports optional time-restrictions, weighting, and priority
723 modifiers per server. Patch originally by <rommer@active.by>.
725 JH/16 The spamd_address main option now supports a mixed list of local
726 and remote servers. Remote servers can be IPv6 addresses, and
727 specify a port-range.
729 JH/17 Bug 68: The spamd_address main option now supports an optional
730 timeout value per server.
732 JH/18 Bug 1581: Router and transport options headers_add/remove can
733 now have the list separator specified.
735 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
738 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
741 JH/21 Support for the A6 type of dns record is withdrawn.
743 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
744 rather than the verbs used.
746 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
747 from 255 to 1024 chars.
749 JH/24 Verification callouts now attempt to use TLS by default.
751 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
752 are generic router options now. The defaults didn't change.
754 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
755 Original patch from Alexander Shikoff, worked over by JH.
757 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
758 files if ZCAT_COMMAND is not executable.
760 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
762 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
764 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
765 Normally benign, it bites when the pair was led to by a CNAME;
766 modern usage is to not canonicalize the domain to a CNAME target
767 (and we were inconsistent anyway for A-only vs AAAA+A).
769 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
771 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
772 when evaluating $sender_host_dnssec.
774 JH/31 Check the HELO verification lookup for DNSSEC, adding new
775 $sender_helo_dnssec variable.
777 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
779 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
781 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
783 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
784 documented as working, but never had. Support all but $spam_report.
786 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
787 added for tls authenticator.
789 HS/03 Add perl_taintmode main config option
795 TL/01 When running the test suite, the README says that variables such as
796 no_msglog_check are global and can be placed anywhere in a specific
797 test's script, however it was observed that placement needed to be near
798 the beginning for it to behave that way. Changed the runtest perl
799 script to read through the entire script once to detect and set these
800 variables, reset to the beginning of the script, and then run through
801 the script parsing/test process like normal.
803 TL/02 The BSD's have an arc4random API. One of the functions to induce
804 adding randomness was arc4random_stir(), but it has been removed in
805 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
806 function when detected.
808 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
809 cause callback expansion.
811 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
812 syntax errors in an expansion can be treated as a string instead of
813 logging or causing an error, due to the internal use of bool_lax
814 instead of bool when processing it.
816 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
817 server certificates when making smtp deliveries.
819 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
821 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
823 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
825 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
826 Merged patch from Sebastian Wiedenroth.
828 JH/05 Fix results-pipe from transport process. Several recipients, combined
829 with certificate use, exposed issues where response data items split
830 over buffer boundaries were not parsed properly. This eventually
831 resulted in duplicates being sent. This issue only became common enough
832 to notice due to the introduction of connection certificate information,
833 the item size being so much larger. Found and fixed by Wolfgang Breyha.
835 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
836 size buffer was used, resulting in syntax errors when an expansion
839 JH/07 Add support for directories of certificates when compiled with a GnuTLS
840 version 3.3.6 or later.
842 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
843 is EXPERIMENTAL_EVENT, the main-configuration and transport options
844 both become "event_action", the variables become $event_name, $event_data
845 and $event_defer_errno. There is a new variable $verify_mode, usable in
846 routers, transports and related events. The tls:cert event is now also
847 raised for inbound connections, if the main configuration event_action
850 TL/06 In test suite, disable OCSP for old versions of openssl which contained
851 early OCSP support, but no stapling (appears to be less than 1.0.0).
853 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
854 server certificate names available under the smtp transport option
855 "tls_verify_cert_hostname" now do not permit multi-component wildcard
858 JH/10 Time-related extraction expansions from certificates now use the main
859 option "timezone" setting for output formatting, and are consistent
860 between OpenSSL and GnuTLS compilations. Bug 1541.
862 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
863 encoded parameter in the incoming message. Bug 1558.
865 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
866 include certificate info, eximon was claiming there were spoolfile
869 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
871 JH/14 Log delivery-related information more consistently, using the sequence
872 "H=<name> [<ip>]" wherever possible.
874 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
875 are problematic for Debian distribution, omit them from the release
878 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
880 JH/16 Fix string representation of time values on 64bit time_t architectures.
883 JH/17 Fix a null-indirection in certextract expansions when a nondefault
884 output list separator was used.
889 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
890 checkers that were complaining about end of non-void function with no
893 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
894 This was a regression introduced in 4.83 by another bugfix.
896 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
898 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
899 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
905 TF/01 Correctly close the server side of TLS when forking for delivery.
907 When a message was received over SMTP with TLS, Exim failed to clear up
908 the incoming connection properly after forking off the child process to
909 deliver the message. In some situations the subsequent outgoing
910 delivery connection happened to have the same fd number as the incoming
911 connection previously had. Exim would try to use TLS and fail, logging
912 a "Bad file descriptor" error.
914 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
915 utilities have not been installed.
917 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
918 temporary space as the ACL may create new global variables.
920 TL/01 LDAP support uses per connection or global context settings, depending
921 upon the detected version of the libraries at build time.
923 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
924 to extract and use the src ip:port in logging and expansions as if it
925 were a direct connection from the outside internet. PPv2 support was
926 updated based on HAProxy spec change in May 2014.
928 JH/02 Add ${listextract {number}{list}{success}{fail}}.
930 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
931 Properly escape header and check for NULL return.
933 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
936 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
938 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
939 characters in header names, implemented as a verify condition.
940 Contributed by Michael Fischer v. Mollard.
942 TL/05 Rename SPF condition results err_perm and err_temp to standardized
943 results permerror and temperror. Previous values are deprecated but
944 still accepted. In a future release, err_perm and err_temp will be
945 completely removed, which will be a backward incompatibility if the
946 ACL tests for either of these two old results. Patch contributed by
947 user bes-internal on the mailing list.
949 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
951 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
952 selectors, in both main and reject logs.
954 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
957 JH/07 Add malware type "sock" for talking to simple daemon.
959 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
961 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
962 routers/transports under cutthrough routing.
964 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
965 numbers. Touch up "bool" conditional to keep the same definition.
967 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
969 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
971 JH/12 Expand items in router/transport headers_add or headers_remove lists
972 individually rather than the list as a whole. Bug 1452.
974 Required for reasonable handling of multiple headers_ options when
975 they may be empty; requires that headers_remove items with embedded
976 colons must have them doubled (or the list-separator changed).
978 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
979 view the policy declared in the DMARC record. Currently, $dmarc_status
980 is a combined value of both the record presence and the result of the
983 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
985 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
986 dnslookup router and the smtp transport (applying to the forward
989 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
990 of ldap servers used for a specific lookup. Patch provided by Heiko
993 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
994 New variable $lookup_dnssec_authenticated for observability.
996 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
997 Patch submitted by Lars Timman.
999 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1001 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1002 Requires trusted mode and valid format message id, aborts otherwise.
1003 Patch contributed by Heiko Schlichting.
1005 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1006 certextract with support for various fields. Bug 1358.
1008 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1009 is requested by default, modifiable by smtp transport option
1012 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1013 operate on certificate variables to give certificate fingerprints
1014 Also new ${sha256:cert_variable}.
1016 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1018 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1019 Christian Aistleitner.
1021 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1023 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1024 file. Patch from Wolfgang Breyha.
1026 JH/25 Expand the coverage of the delivery $host and $host_address to
1027 client authenticators run in verify callout. Bug 1476.
1029 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1030 align with daemon_smtp_ports. Bug 72.
1032 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1033 support and error reporting did not work properly.
1035 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1036 and is readable. Patch from Andrew Colin Kissa.
1038 TL/14 Enhance documentation of ${run expansion and how it parses the
1039 commandline after expansion, particularly in the case when an
1040 unquoted variable expansion results in an empty value.
1042 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1044 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1045 by renaming away from it. Addresses GH issue 15, reported by
1048 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1050 TL/15 SECURITY: prevent double expansion in math comparison functions
1051 (can expand unsanitized data). Not remotely exploitable.
1058 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1060 PP/02 Make -n do something, by making it not do something.
1061 When combined with -bP, the name of an option is not output.
1063 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1066 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1067 $sender_host_name and config options to manage this, and basic check
1070 PP/05 DSCP support for outbound connections and control modifier for inbound.
1072 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1073 (Only plugin which currently uses this is kerberos4, which nobody should
1074 be using, but we should make it available and other future plugins might
1075 conceivably use it, even though it would break NAT; stuff *should* be
1076 using channel bindings instead).
1078 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1079 name; added for Sendmail compatibility; requires admin caller.
1080 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1081 just ignore it); requires trusted caller.
1082 Also parse but ignore: -Ac -Am -X<logfile>
1085 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1087 TL/02 Add +smtp_confirmation as a default logging option.
1089 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1090 Patch by Magnus Holmgren from 2007-02-20.
1092 TL/04 Bugzilla 1281 - Spec typo.
1093 Bugzilla 1283 - Spec typo.
1094 Bugzilla 1290 - Spec grammar fixes.
1096 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1098 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1100 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1101 mailing list by by Dmitry Isaikin.
1103 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1105 JH/02 Support "G" suffix to numbers in ${if comparisons.
1107 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1109 NM/01 Bugzilla 1197 - Spec typo
1110 Bugzilla 1196 - Spec examples corrections
1112 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1114 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1115 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1118 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1119 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1121 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1122 "acl {{name}{arg}...}", and optional args on acl condition
1125 JH/05 Permit multiple router/transport headers_add/remove lines.
1127 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1129 JH/07 Avoid using a waiting database for a single-message-only transport.
1130 Performance patch from Paul Fisher. Bugzilla 1262.
1132 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1135 JH/09 Add $headers_added variable, with content from use of ACL modifier
1136 add_header (but not yet added to the message). Bugzilla 199.
1138 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1139 Pulled from Bugzilla 817 by Wolfgang Breyha.
1141 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1143 (nb: this is the same fix as in Exim 4.80.1)
1145 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1148 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1151 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1153 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1154 advertise SMTP AUTH mechanism to us, instead of a generic
1155 protocol violation error. Also, make Exim more robust to bad
1156 data from the Dovecot auth socket.
1158 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1160 When a queue runner is handling a message, Exim first routes the
1161 recipient addresses, during which it prunes them based on the retry
1162 hints database. After that it attempts to deliver the message to
1163 any remaining recipients. It then updates the hints database using
1166 So if a recipient address works intermittently, it can get repeatedly
1167 deferred at routing time. The retry hints record remains fresh so the
1168 address never reaches the final cutoff time.
1170 This is a fairly common occurrence when a user is bumping up against
1171 their storage quota. Exim had some logic in its local delivery code
1172 to deal with this. However it did not apply to per-recipient defers
1173 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1175 This change adds a proper retry rule check during routing so that the
1176 final cutoff time is checked against the message's age. We only do
1177 this check if there is an address retry record and there is not a
1178 domain retry record; this implies that previous attempts to handle
1179 the address had the retry_use_local_parts option turned on. We use
1180 this as an approximation for the destination being like a local
1181 delivery, as in LMTP.
1183 I suspect this new check makes the old local delivery cutoff check
1184 redundant, but I have not verified this so I left the code in place.
1186 TF/02 Correct gecos expansion when From: is a prefix of the username.
1188 Test 0254 submits a message to Exim with the header
1192 When I ran the test suite under the user fanf2, Exim expanded
1193 the header to contain my full name, whereas it should have added
1194 a Resent-Sender: header. It erroneously treats any prefix of the
1195 username as equal to the username.
1197 This change corrects that bug.
1199 GF/01 DCC debug and logging tidyup
1200 Error conditions log to paniclog rather than rejectlog.
1201 Debug lines prefixed by "DCC: " to remove any ambiguity.
1203 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1205 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1206 Bug spotted by Jeremy Harris; was flawed since initial commit.
1207 Would have resulted in OCSP responses post-SNI triggering an Exim
1208 NULL dereference and crash.
1210 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1212 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1213 Bug detection, analysis and fix by Samuel Thibault.
1214 Bugzilla 1331, Debian bug #698092.
1216 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1218 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1219 Server implementation by Todd Lyons, client by JH.
1220 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1221 config variable "prdr_enable" controls whether the server
1222 advertises the facility. If the client requests PRDR a new
1223 acl_data_smtp_prdr ACL is called once for each recipient, after
1224 the body content is received and before the acl_smtp_data ACL.
1225 The client is controlled by both of: a hosts_try_prdr option
1226 on the smtp transport, and the server advertisement.
1227 Default client logging of deliveries and rejections involving
1228 PRDR are flagged with the string "PRDR".
1230 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1231 fclose(). Diagnosis by Todd Lyons.
1233 PP/17 Update configure.default to handle IPv6 localhost better.
1234 Patch by Alain Williams (plus minor tweaks).
1237 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1238 This is now consistent with GnuTLS, and is now documented: the
1239 previous undocumented portable approach to treating the option as
1240 unset was to force an expansion failure. That still works, and
1241 an empty string is now equivalent.
1243 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1244 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1245 not performing validation itself.
1247 PP/20 Added force_command boolean option to pipe transport.
1248 Patch from Nick Koston, of cPanel Inc.
1250 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1253 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1255 PP/21 Fix eximon continuous updating with timestamped log-files.
1256 Broken in a format-string cleanup in 4.80, missed when I repaired the
1257 other false fix of the same issue.
1258 Report and fix from Heiko Schlichting.
1261 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1262 Report from Prashanth Katuri.
1264 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1265 It's SecureTransport, so affects any MacOS clients which use the
1266 system-integrated TLS libraries, including email clients.
1268 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1269 using a MIME ACL for non-SMTP local injection.
1270 Report and assistance in diagnosis by Warren Baker.
1272 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1274 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1276 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1277 last id that failed so it may be referenced in subsequent ACL's.
1279 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1282 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1283 ldap library initialization, allowing self-signed CA's to be
1284 used. Also properly sets require_cert option later in code by
1285 using NULL (global ldap config) instead of ldap handle (per
1286 session). Bug diagnosis and testing by alxgomz.
1288 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1289 the src/util/ subdirectory.
1291 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1292 renamed to Transport Post Delivery Action by Jeremy Harris, as
1295 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1296 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1297 redis_servers = needs to be configured which will be used by the redis
1298 lookup. Patch from Warren Baker, of The Packet Hub.
1300 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1302 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1303 hostname or reverse DNS when processing a host list. Used suggestions
1304 from multiple comments on this bug.
1306 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1308 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1309 Missed a few lines, added it to make the runtest require no keyboard
1312 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1313 contains upper case chars. Make router use caseful_local_part.
1315 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1316 support when GnuTLS has been built with p11-kit.
1322 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1324 This, or similar/improved, will also be change PP/11 of 4.82.
1330 PP/01 Handle short writes when writing local log-files.
1331 In practice, only affects FreeBSD (8 onwards).
1332 Bugzilla 1053, with thanks to Dmitry Isaikin.
1334 NM/01 Bugzilla 949 - Documentation tweak
1336 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1339 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1341 PP/02 Implemented gsasl authenticator.
1343 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1345 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1346 `pkg-config foo` for cflags/libs.
1348 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1349 with rest of GSASL and with heimdal_gssapi.
1351 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1352 `pkg-config foo` for cflags/libs for the TLS implementation.
1354 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1355 properties get this fed in as external SSF. A number of robustness
1356 and debugging improvements to the cyrus_sasl authenticator.
1358 PP/08 cyrus_sasl server now expands the server_realm option.
1360 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1361 Patch by Jeremy Harris.
1363 PP/10 Added dbmjz lookup type.
1365 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1367 PP/12 MAIL args handles TAB as well as SP, for better interop with
1368 non-compliant senders.
1369 Analysis and variant patch by Todd Lyons.
1371 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1372 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1373 Patch from Dirk Mueller <dmueller@suse.com>
1375 PP/13 tls_peerdn now print-escaped for spool files.
1376 Observed some $tls_peerdn in wild which contained \n, which resulted
1377 in spool file corruption.
1379 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1380 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1381 or write after TLS renegotiation, which otherwise led to messages
1384 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1385 as a tracking header (ie: a signed header comes before the signature).
1386 Patch from Wolfgang Breyha.
1388 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1389 comma-sep list; embedded commas doubled.
1391 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1393 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1395 Report and patch from Dmitry Banschikov.
1397 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1398 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1399 failures. We appear to now support TLS1.1+ with Exim.
1401 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1402 lets Exim select keys and certificates based upon TLS SNI from client.
1403 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1404 before an outbound SMTP session. New log_selector, +tls_sni.
1406 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1407 NULL dereference. Report and patch from Alun Jones.
1409 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1410 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1411 Not seeing resolver debug output on NetBSD, but suspect this is a
1412 resolver implementation change.
1414 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1415 Left warnings. Added "eximon gdb" invocation mode.
1417 PP/21 Defaulting "accept_8bitmime" to true, not false.
1419 PP/22 Added -bw for inetd wait mode support.
1421 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1422 locate the relevant includes and libraries. Made this the default.
1424 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1425 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1427 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1428 This may cause build issues on older platforms.
1430 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1431 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1432 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1433 Added SNI support via GnuTLS too.
1434 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1436 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1438 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1439 Applied second patch from Janne, implementing suggestion to default
1440 multiple-strings-in-record handling to match SPF spec.
1442 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1444 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1445 read-only, out of scope).
1446 Patch from Wolfgang Breyha, report from Stuart Northfield.
1448 PP/29 Fix three issues highlighted by clang analyser static analysis.
1449 Only crash-plausible issue would require the Cambridge-specific
1450 iplookup router and a misconfiguration.
1451 Report from Marcin Mirosław.
1453 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1455 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1456 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1457 As part of this, removing so much warning spew let me fix some minor
1458 real issues in debug logging.
1460 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1461 assignment on my part. Fixed.
1463 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1464 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1465 Janne Snabb (who went above and beyond: thank you).
1467 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1468 string otherwise requires a connection and a bunch more work and it's
1469 relatively easy to get wrong. Should also expose TLS library linkage
1472 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1473 64-bit ${eval} (JH/03).
1475 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1476 GNU libc to support some of the 64-bit stuff, should not lead to
1477 conflicts. Defined before os.h is pulled in, so if a given platform
1478 needs to override this, it can.
1480 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1481 protection layer was required, which is not implemented.
1482 Bugzilla 1254, patch from Wolfgang Breyha.
1484 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1485 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1486 tls_dhparam take prime identifiers. Also unbreak combination of
1487 OpenSSL+DH_params+TLSSNI.
1489 PP/39 Disable SSLv2 by default in OpenSSL support.
1495 PP/01 Solaris build fix for Oracle's LDAP libraries.
1496 Bugzilla 1109, patch from Stephen Usher.
1498 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1500 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1503 TF/02 Fix a couple more cases where we did not log the error message
1504 when unlink() failed. See also change 4.74-TF/03.
1506 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1507 lock up or crash if it happened to be inside a call to libc when it
1508 got a SIGUSR1 from exiwhat.
1510 The SIGUSR1 handler appends the current process status to the process
1511 log which is later printed by exiwhat. It used to use the general
1512 purpose logging code to do this, but several functions it calls are
1513 not safe for signals.
1515 The new output code in the SIGUSR1 handler is specific to the process
1516 log, and simple enough that it's easy to inspect for signal safety.
1517 Removing some special cases also simplifies the general logging code.
1518 Removing the spurious timestamps from the process log simplifies
1521 TF/04 Improved ratelimit ACL condition.
1523 The /noupdate option has been deprecated in favour of /readonly which
1524 has clearer semantics. The /leaky, /strict, and /readonly update modes
1525 are mutually exclusive. The update mode is no longer included in the
1526 database key; it just determines when the database is updated. (This
1527 means that when you upgrade Exim will forget old rate measurements.)
1529 Exim now checks that the per_* options are used with an update mode that
1530 makes sense for the current ACL. For example, when Exim is processing a
1531 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1532 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1533 must specify per_mail/readonly. If you omit the update mode it defaults to
1534 /leaky where that makes sense (as before) or /readonly where required.
1536 The /noupdate option is now undocumented but still supported for
1537 backwards compatibility. It is equivalent to /readonly except that in
1538 ACLs where /readonly is required you may specify /leaky/noupdate or
1539 /strict/noupdate which are treated the same as /readonly.
1541 A useful new feature is the /count= option. This is a generalization
1542 of the per_byte option, so that you can measure the throughput of other
1543 aggregate values. For example, the per_byte option is now equivalent
1544 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1546 The per_rcpt option has been generalized using the /count= mechanism
1547 (though it's more complicated than the per_byte equivalence). When it is
1548 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1549 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1550 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1551 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1552 Note that using per_rcpt with a non-readonly update mode in more than
1553 one ACL will cause the recipients to be double-counted. (The per_mail
1554 and per_byte options don't have this problem.)
1556 The handling of very low rates has changed slightly. If the computed rate
1557 is less than the event's count (usually one) then this event is the first
1558 after a long gap. In this case the rate is set to the same as this event's
1559 count, so that the first message of a spam run is counted properly.
1561 The major new feature is a mechanism for counting the rate of unique
1562 events. The new per_addr option counts the number of different
1563 recipients that someone has sent messages to in the last time period. It
1564 behaves like per_rcpt if all the recipient addresses are different, but
1565 duplicate recipient addresses do not increase the measured rate. Like
1566 the /count= option this is a general mechanism, so the per_addr option
1567 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1568 example, measure the rate that a client uses different sender addresses
1569 with the options per_mail/unique=$sender_address. There are further
1570 details in the main documentation.
1572 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1574 TF/06 Removed a few PCRE remnants.
1576 TF/07 Automatically extract Exim's version number from tags in the git
1577 repository when doing development or release builds.
1579 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1580 Bugzilla 879. Patch from Paul Fisher.
1582 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1583 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1586 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1588 PP/05 Variable $av_failed, true if the AV scanner deferred.
1589 Bugzilla 1078. Patch from John Horne.
1591 PP/06 Stop make process more reliably on build failure.
1592 Bugzilla 1087. Patch from Heiko Schlittermann.
1594 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1595 Bugzilla 1089. Patch from Heiko Schlittermann.
1597 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1598 Bugzilla 1131. Patch from Holger Weiß.
1600 PP/09 Handle IPv6 addresses with SPF.
1601 Bugzilla 860. Patch from Wolfgang Breyha.
1603 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1605 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1608 PP/11 match_* no longer expand right-hand-side by default.
1609 New compile-time build option, EXPAND_LISTMATCH_RHS.
1610 New expansion conditions, "inlist", "inlisti".
1612 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1614 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1616 PP/14 fix log_write() format string regression from TF/03.
1617 Bugzilla 1152. Patch from Dmitry Isaikin.
1623 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1625 PP/02 Harmonised TLS library version reporting; only show if debugging.
1626 Layout now matches that introduced for other libraries in 4.74 PP/03.
1628 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1630 PP/04 New "dns_use_edns0" global option.
1632 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1635 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1636 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1638 TK/01 Updated PolarSSL code to 0.14.2.
1639 Bugzilla 1097. Patch from Andreas Metzler.
1641 PP/07 Catch divide-by-zero in ${eval:...}.
1642 Fixes bugzilla 1102.
1644 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1647 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1648 format-string attack -- SECURITY: remote arbitrary code execution.
1650 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1651 time unintentionally subject to list matching rules, letting the header
1652 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1653 arbitrary string expansion). This allowed for information disclosure.
1655 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1656 INT_MIN/-1 -- value coerced to INT_MAX.
1662 NM/01 Workaround for PCRE version dependency in version reporting
1665 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1666 This fixes portability to compilers other than gcc, notably
1667 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1669 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1670 makefiles for portability to HP-UX and POSIX correctness.
1672 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1673 Also via indented variable definition in the Makefile.
1674 (Debugging by Oliver Heesakkers).
1676 PP/02 Restore caching of spamd results with expanded spamd_address.
1677 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1679 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1680 Improves build reliability. Fix from: Frank Elsner
1682 NM/02 Fix wide character breakage in the rfc2047 coding
1683 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1685 NM/03 Allow underscore in dnslist lookups
1686 Fixes bug 1026. Patch from Graeme Fowler
1688 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1689 Code patches from Adam Ciarcinski of NetBSD.
1691 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1694 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1695 is logged, to avoid truncation. Patch from John Horne.
1697 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1698 Patch from Jakob Hirsch.
1700 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1701 SQL string expansion failure details.
1702 Patch from Andrey Oktyabrski.
1704 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1705 Patch from Simon Arlott.
1707 PP/09 New lookups functionality failed to compile on old gcc which rejects
1708 extern declarations in function scope.
1709 Patch from Oliver Fleischmann
1711 PP/10 Use sig_atomic_t for flags set from signal handlers.
1712 Check getgroups() return and improve debugging.
1713 Fixed developed for diagnosis in bug 927 (which turned out to be
1716 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1717 Patch from Mark Zealey.
1719 PP/12 Bugzilla 1056: Improved spamd server selection.
1720 Patch from Mark Zealey.
1722 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1723 Based on patch from Heiko Schlittermann.
1725 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1726 Patch from Uwe Doering, sign-off by Michael Haardt.
1728 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1729 variable declaration deep within a block. Bug and patch from
1732 PP/15 lookups-Makefile IRIX compatibility coercion.
1734 PP/16 Make DISABLE_DKIM build knob functional.
1736 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1737 Patch by Simon Arlott
1739 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1740 variable argument macros. Our copy now differs from upstream.
1746 TF/01 Failure to get a lock on a hints database can have serious
1747 consequences so log it to the panic log.
1749 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1750 controlled using the smtp_confirmation log selector.
1752 TF/03 Include the error message when we fail to unlink a spool file.
1754 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1755 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1756 for maintaining out-of-tree patches for some time.
1758 PP/01 Bugzilla 139: Documentation and portability issues.
1759 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1760 Handle per-OS dynamic-module compilation flags.
1762 PP/02 Let /dev/null have normal permissions.
1763 The 4.73 fixes were a little too stringent and complained about the
1764 permissions on /dev/null. Exempt it from some checks.
1765 Reported by Andreas M. Kirchwitz.
1767 PP/03 Report version information for many libraries, including
1768 Exim version information for dynamically loaded libraries. Created
1769 version.h, now support a version extension string for distributors
1770 who patch heavily. Dynamic module ABI change.
1772 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1773 privilege escalation vulnerability whereby the Exim run-time user
1774 can cause root to append content of the attacker's choosing to
1777 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1780 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1781 If dropping privileges for untrusted macros, we disabled normal logging
1782 on the basis that it would fail; for the Exim run-time user, this is not
1783 the case, and it resulted in successful deliveries going unlogged.
1784 Fixed. Reported by Andreas Metzler.
1790 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1791 only prepend for the Resent-* case. Fixes regression introduced in
1792 Exim 4.70 by NM/22 for Bugzilla 607.
1794 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1795 increasing numbers of administrators be bitten by this.
1797 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1799 PP/03 Bugzilla 994: added openssl_options main configuration option.
1801 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1803 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1805 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1807 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1808 without a peer certificate, leading to a segfault because of an
1809 assumption that peers always have certificates. Be a little more
1810 paranoid. Problem reported by Martin Tscholak.
1812 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1813 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1814 NB: ClamAV planning to remove STREAM in "middle of 2010".
1815 CL also introduces -bmalware, various -d+acl logging additions and
1816 more caution in buffer sizes.
1818 PP/09 Implemented reverse_ip expansion operator.
1820 PP/10 Bugzilla 937: provide a "debug" ACL control.
1822 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1824 PP/12 Bugzilla 973: Implement --version.
1826 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1828 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1830 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1832 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1833 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1834 ignore trailing whitespace.
1836 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1838 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1841 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1842 Notification from Dr Andrew Aitchison.
1844 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1845 ExtendedDetectionInfo response format.
1846 Notification from John Horne.
1848 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1851 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1852 XSL and documented dependency on system catalogs, with examples of how
1855 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1858 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1859 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1860 configuration file which is writeable by the Exim user or group.
1862 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1863 of configuration files to cover files specified with the -C option if
1864 they are going to be used with root privileges, not just the default
1867 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1868 option (effectively making it always true).
1870 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1871 files to be used while preserving root privileges.
1873 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1874 that rogue child processes cannot use them.
1876 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1877 run-time user, instead of root.
1879 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1880 Exim run-time user without dropping privileges.
1882 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1883 result string, instead of calling string_vformat() twice with the same
1886 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1887 for other users. Others should always drop root privileges if they use
1888 -C on the command line, even for a whitelisted configure file.
1890 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1892 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1898 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1899 $data_path, and $header_path variables; fixed documentation bugs and
1902 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1903 exipick to access non-standard spools, including the "frozen" queue
1906 NM/01 Bugzilla 965: Support mysql stored procedures.
1907 Patch from Alain Williams
1909 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1911 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1912 Patch from Andreas Metzler
1914 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1915 Patch from Kirill Miazine
1917 NM/05 Bugzilla 671: Added umask to procmail example.
1919 JJ/03 installed exipick 20100323.0, fixing doc bug
1921 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1922 directory. Notification and patch from Dan Rosenberg.
1924 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1926 TK/02 Improve log output when DKIM signing operation fails.
1928 MH/01 Treat the transport option dkim_domain as a colon separated
1929 list, not as a single string, and sign the message with each element,
1930 omitting multiple occurences of the same signer.
1932 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1933 Bugzilla 985, 986. Patch by Simon Arlott
1935 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1936 Patch by Simon Arlott
1938 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1939 MBX locking. Notification from Dan Rosenberg.
1945 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1947 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1949 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1951 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1953 NM/04 Bugzilla 915: Flag broken perl installation during build.
1959 TK/01 Added patch by Johannes Berg that expands the main option
1960 "spamd_address" if it starts with a dollar sign.
1962 TK/02 Write list of recipients to X-Envelope-Sender header when building
1963 the mbox-format spool file for content scanning (suggested by Jakob
1966 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1967 (http://www.dcc-servers.net/) support via dccifd. Activated by
1968 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1970 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1971 by Mark Daniel Reidel <mr@df.eu>.
1973 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1974 When building exim an external PCRE library is now needed -
1975 PCRE is a system library on the majority of modern systems.
1976 See entry on PCRE_LIBS in EDITME file.
1978 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1979 conversation. Added nologin parameter to request.
1980 Patch contributed by Kirill Miazine.
1982 TF/01 Do not log submission mode rewrites if they do not change the address.
1984 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1986 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1987 log files in place. Contributed by Roberto Lima.
1989 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1991 TF/03 Bugzilla 615: When checking the local_parts router precondition
1992 after a local_part_suffix or local_part_prefix option, Exim now
1993 does not use the address's named list lookup cache, since this
1994 contains cached lookups for the whole local part.
1996 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1997 Robert Millan. Documentation is in experimental-spec.txt.
1999 TF/04 Bugzilla 668: Fix parallel build (make -j).
2001 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2003 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2004 Patch provided by Jan Srzednicki.
2006 TF/05 Leading white space used to be stripped from $spam_report which
2007 wrecked the formatting. Now it is preserved.
2009 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2010 that they are available at delivery time.
2012 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2014 TF/08 TLS error reporting now respects the incoming_interface and
2015 incoming_port log selectors.
2017 TF/09 Produce a more useful error message if an SMTP transport's hosts
2018 setting expands to an empty string.
2020 NM/06 Bugzilla 744: EXPN did not work under TLS.
2021 Patch provided by Phil Pennock.
2023 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2024 Patch provided by Richard Godbee.
2026 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2027 acl_smtp_notquit, added index entry.
2029 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2030 Patch provided by Eugene Bujak.
2032 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2033 accept(). Patch provided by Maxim Dounin.
2035 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2036 Patch provided by Phil Pennock.
2038 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2040 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2041 Patch provided by Brad "anomie" Jorsch.
2043 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2044 Patch provided by Dean Brooks.
2046 TK/05 Add native DKIM support (does not depend on external libraries).
2048 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2049 Patch provided by Graeme Fowler.
2051 NM/16 Bugzilla 851: Documentation example syntax fix.
2053 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2055 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2058 NM/19 Bugzilla 745: TLS version reporting.
2059 Patch provided by Phil Pennock.
2061 NM/20 Bugzilla 167: bool: condition support.
2062 Patch provided by Phil Pennock.
2064 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2065 clients. Patch provided by Phil Pennock.
2067 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2068 Patch provided by Brad "anomie" Jorsch.
2070 NM/23 Bugzilla 687: Fix misparses in eximstats.
2071 Patch provided by Heiko Schlittermann.
2073 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2074 Patch provided by Heiko Schlittermann.
2076 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2077 plus update to original patch.
2079 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2081 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2082 Patch provided by David Brownlee.
2084 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2086 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2088 NM/30 Bugzilla 888: TLS documentation bugfixes.
2090 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2092 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2093 Unlike the original bugzilla I have changed all shell scripts in src tree.
2095 NM/33 Bugzilla 898: Transport filter timeout fix.
2096 Patch by Todd Rinaldo.
2098 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2099 Patch by Serge Demonchaux.
2101 NM/35 Bugzilla 39: Base64 decode bug fixes.
2102 Patch by Jakob Hirsch.
2104 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2106 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2108 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2110 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2116 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2117 ALT-N's libdkim that I have put here:
2118 http://duncanthrax.net/exim-experimental/
2120 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2121 were called 'true' and it seems that C99 defines that as a reserved
2122 keyword to be used with 'bool' variable types. That means you could
2123 not include C99-style headers which use bools without triggering
2124 build errors in sieve.c.
2126 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2127 as mailq or other aliases. Changed the --help handling significantly
2128 to do whats expected. exim_usage() emits usage/help information.
2130 SC/01 Added the -bylocaldomain option to eximstats.
2132 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2134 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2136 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2142 PH/01 Another patch from the Sieve maintainer.
2144 PH/02 When an IPv6 address is converted to a string for single-key lookup
2145 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2146 dots are used instead of colons so that keys in lsearch files need not
2147 contain colons. This was done some time before quoting was made available
2148 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2149 (notated using the quote facility) so as to distinguish them from IPv4
2150 keys. This meant that lookups for IP addresses in host lists did not work
2151 for iplsearch lookups.
2153 This has been fixed by arranging for IPv6 addresses to be expressed with
2154 colons if the lookup type is iplsearch. This is not incompatible, because
2155 previously such lookups could never work.
2157 The situation is now rather anomalous, since one *can* have colons in
2158 ordinary lsearch keys. However, making the change in all cases is
2159 incompatible and would probably break a number of configurations.
2161 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2164 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2165 conversion specification without a maximum field width, thereby enabling
2166 a rogue spamd server to cause a buffer overflow. While nobody in their
2167 right mind would setup Exim to query an untrusted spamd server, an
2168 attacker that gains access to a server running spamd could potentially
2169 exploit this vulnerability to run arbitrary code as the Exim user.
2171 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2172 $primary_hostname instead of what libspf2 thinks the hosts name is.
2174 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2175 a directory entry by the name of the lookup key. Previously, if a
2176 symlink pointed to a non-existing file or a file in a directory that
2177 Exim lacked permissions to read, a lookup for a key matching that
2178 symlink would fail. Now it is enough that a matching directory entry
2179 exists, symlink or not. (Bugzilla 503.)
2181 PH/03 The body_linecount and body_zerocount variables are now exported in the
2184 PH/04 Added the $dnslist_matched variable.
2186 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2187 This means they are set thereafter only if the connection becomes
2190 PH/06 Added the client_condition to authenticators so that some can be skipped
2191 by clients under certain conditions.
2193 PH/07 The error message for a badly-placed control=no_multiline_responses left
2194 "_responses" off the end of the name.
2196 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2198 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2199 (without spaces) instead of just copying the configuration text.
2201 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2203 PH/11 Added $max_received_linelength.
2205 PH/12 Added +ignore_defer and +include_defer to host lists.
2207 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2208 way in which PCRE > 7.0 is built.
2210 PH/14 Implemented queue_only_load_latch.
2212 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2213 MAIL command. The effect was to mangle the value on 64-bit systems.
2215 PH/16 Another patch from the Sieve maintainer.
2217 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2219 PH/18 If a system quota error occurred while trying to create the file for
2220 a maildir delivery, the message "Mailbox is full" was not appended to the
2221 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2222 only to a quota excession during the actual writing of the file.
2224 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2225 characters?) which causes problems in log lines. The DN values are now
2226 passed through string_printing() before being added to log lines.
2228 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2229 and InterBase are left for another time.)
2231 PH/21 Added message_body_newlines option.
2233 PH/22 Guard against possible overflow in moan_check_errorcopy().
2235 PH/23 POSIX allows open() to be a macro; guard against that.
2237 PH/24 If the recipient of an error message contained an @ in the local part
2238 (suitably quoted, of course), incorrect values were put in $domain and
2239 $local_part during the evaluation of errors_copy.
2245 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2246 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2249 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2250 issue a MAIL command.
2252 PH/02 In an ACL statement such as
2254 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2256 if a client was not listed at all, or was listed with a value other than
2257 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2258 the condition was not true (as it should be), so access was not denied.
2259 The bug was that the ! inversion was incorrectly passed on to the second
2260 item. This has been fixed.
2262 PH/03 Added additional dnslists conditions == and =& which are different from
2263 = and & when the dns lookup returns more than one IP address.
2265 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2266 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2268 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2269 FSYNC, which compiles an option called disable_fsync that allows for
2270 bypassing fsync(). The documentation is heavily laced with warnings.
2272 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2274 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2275 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2276 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2277 including adding "make clean"; (3) Added -fPIC when compiling the test
2278 dynamically loaded module, to get rid of a warning.
2280 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2281 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2282 = 0s. The bug is otherwise harmless.
2284 PH/07 There was a bug in the dovecot authenticator such that the value of
2285 $auth1 could be overwritten, and so not correctly preserved, after a
2286 successful authentication. This usually meant that the value preserved by
2287 the server_setid option was incorrect.
2289 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2291 PH/09 Installed PCRE release 7.0.
2293 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2294 run for batched SMTP input. It is now run at the start of every message
2295 in the batch. While fixing this I discovered that the process information
2296 (output by running exiwhat) was not always getting set for -bs and -bS
2297 input. This is fixed, and it now also says "batched" for BSMTP.
2299 PH/11 Added control=no_pipelining.
2301 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2302 patch, slightly modified), and move the expansion of helo_data till after
2303 the connection is made in the smtp transport (so it can use these
2306 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2308 PH/14 Added log_selector = +pid.
2310 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2312 PH/16 Add ${if forany and ${if forall.
2314 PH/17 Added dsn_from option to vary the From: line in DSNs.
2316 PH/18 Flush SMTP output before performing a callout, unless control =
2317 no_callout_flush is set.
2319 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2320 was true (the default) a successful delivery failed to delete the retry
2321 item, thus causing premature timeout of the address. The bug is now
2324 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2326 PH/21 Long custom messages for fakedefer and fakereject are now split up
2327 into multiline responses in the same way that messages for "deny" and
2328 other ACL rejections are.
2330 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2331 with slight modification.
2333 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2334 draft, changing the syntax and factoring some duplicate code".
2336 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2337 for deliveries of the second and subsequent messages over the same SMTP
2340 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2341 ${reduce, with only minor "tidies".
2343 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2345 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2346 expansion side effects.
2348 PH/27 When a message times out after an over-quota error from an Exim-imposed
2349 quota, the bounce message says "mailbox is full". This message was not
2350 being given when it was a system quota that was exceeded. It now should
2353 MH/03 Made $recipients available in local_scan(). local_scan() already has
2354 better access to the recipient list through recipients_list[], but
2355 $recipients can be useful in postmaster-provided expansion strings.
2357 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2358 in the case of a MAIL command with additional options following the
2359 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2360 were accidentally chopped off.
2362 PH/29 SMTP synchronization checks are implemented when a command is read -
2363 there is a check that no more input is waiting when there shouldn't be
2364 any. However, for some commands, a delay in an ACL can mean that it is
2365 some time before the response is written. In this time, more input might
2366 arrive, invalidly. So now there are extra checks after an ACL has run for
2367 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2368 pipelining has not been advertised.
2370 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2372 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2373 returned in the SMTP response when smtp_return_error_details was false.
2374 This has been fixed.
2376 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2377 instead of the C I/O that was originally supplied, because problems were
2378 reported on Solaris.
2380 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2381 Exim which did not show up earlier: it was assuming that a call to
2382 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2383 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2384 was a macro that became an assignment, so it seemed to work. This has
2385 changed to a proper function call with a void return, hence the compile
2386 error. Exim's code has been fixed.
2388 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2391 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2393 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2395 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2396 because it turns out to be too big or there is a timeout) there is no
2397 "Completed" line in the log. When some messages of this type were
2398 selected by exigrep, they were listed as "not completed". Others were
2399 picked up by some special patterns. I have improved the selection
2400 criteria to be more general.
2402 PH/38 The host_find_failed option in the manualroute router can now be set
2403 to "ignore", to completely ignore a host whose IP address cannot be
2404 found. If all hosts are ignored, the behaviour is controlled by the new
2405 host_all_ignored option.
2407 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2408 homing or because of multiple MX records with /mx) generated more than
2409 one IP address, and the following item turned out to be the local host,
2410 all the secondary addresses of the first item were incorrectly removed
2411 from the list, along with the local host and any following hosts (which
2412 is what is supposed to happen).
2414 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2415 whoever called Exim into the -H file. In the case of the daemon it was
2416 behaving confusingly. When first started, it used values for whoever
2417 started the daemon, but after a SIGHUP it used the Exim user (because it
2418 calls itself on a restart). I have changed the code so that it now always
2421 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2422 message are rejected with the same error (e.g. no authentication or bad
2423 sender address), and a DATA command is nevertheless sent (as can happen
2424 with PIPELINING or a stupid MUA), the error message that was given to the
2425 RCPT commands is included in the rejection of the DATA command. This is
2426 intended to be helpful for MUAs that show only the final error to their
2429 PH/42 Another patch from the Sieve maintainer.
2431 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2432 Eximstats - Fixed some broken HTML links and added missing column headers
2434 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2435 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2437 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2439 PH/43 Yet another patch from the Sieve maintainer.
2441 PH/44 I found a way to check for a TCP/IP connection going away before sending
2442 the response to the final '.' that terminates a message, but only in the
2443 case where the client has not sent further data following the '.'
2444 (unfortunately, this is allowed). However, in many cases there won't be
2445 any further data because there won't be any more messages to send. A call
2446 to select() can be used: if it shows that the input is "ready", there is
2447 either input waiting, or the socket has been closed. An attempt to read
2448 the next input character can distinguish the two cases. Previously, Exim
2449 would have sent an OK response which the client would never have see.
2450 This could lead to message repetition. This fix should cure that, at
2451 least in a lot of common cases.
2453 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2454 advertised in response to EHLO.
2460 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2461 fixed by 4.65/MH/01 (is this a record?) are fixed:
2463 (i) An empty string was always treated as zero by the numeric comparison
2464 operators. This behaviour has been restored.
2466 (ii) It is documented that the numeric comparison operators always treat
2467 their arguments as decimal numbers. This was broken in that numbers
2468 starting with 0 were being interpreted as octal.
2470 While fixing these problems I realized that there was another issue that
2471 hadn't been noticed. Values of message_size_limit (both the global option
2472 and the transport option) were treated as octal if they started with 0.
2473 The documentation was vague. These values are now always treated as
2474 decimal, and I will make that clear in the documentation.
2480 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2481 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2484 MH/01 Don't check that the operands of numeric comparison operators are
2485 integers when their expansion is in "skipping" mode (fixes bug
2486 introduced by 4.64-PH/07).
2488 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2489 child addresses, Exim now panics and dies. Previously, because the count
2490 is held in a short int, deliveries were likely to be lost. As such a
2491 large number of recipients for a single message is ridiculous
2492 (performance will be very, very poor), I have chosen to impose a limit
2493 rather than extend the field.
2499 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2500 leftover -K file (the existence of which was triggered by #402).
2501 While we were at it, introduced process PID as part of the -K
2502 filename. This should rule out race conditions when creating
2505 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2506 processing considerably. Previous code took too long for large mails,
2507 triggering a timeout which in turn triggers #401.
2509 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2510 in the DK code in transports.c. sendfile() is not really portable,
2511 hence the _LINUX specificness.
2513 TF/01 In the add_headers option to the mail command in an Exim filter,
2514 there was a bug that Exim would claim a syntax error in any
2515 header after the first one which had an odd number of characters
2518 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2519 callout verification, Exim cached a "reject" for the entire domain. This
2520 is correct for most verifications, but it is not correct for a recipient
2521 verification with use_sender or use_postmaster set, because in that case
2522 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2523 case of MAIL FROM:<> rejection from other early rejections (e.g.
2524 rejection of HELO). When verifying a recipient using a non-null MAIL
2525 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2526 Whatever the result of the callout, the value of the domain cache is
2527 left unchanged (for any other kind of callout, getting as far as trying
2528 RCPT means that the domain itself is ok).
2530 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2533 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2534 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2535 session, the daemon ignores ECONNECT errors and logs others; it now
2536 ignores EPIPE as well.
2538 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2539 (quoted-printable decoding).
2541 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2542 later the small subsequent patch to fix an introduced bug.
2544 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2546 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2548 PH/08 An error is now given if message_size_limit is specified negative.
2550 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2551 to be given (somewhat) arbitrary names.
2553 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2556 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2557 miscellaneous code fixes
2559 PH/10 Added the log_reject_target ACL modifier to specify where to log
2562 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2563 hostname. This is wrong, because it relates to the incoming message (and
2564 probably the interface on which it is arriving) and not to the outgoing
2565 callout (which could be using a different interface). This has been
2566 changed to use the value of the helo_data option from the smtp transport
2567 instead - this is what is used when a message is actually being sent. If
2568 there is no remote transport (possible with a router that sets up host
2569 addresses), $smtp_active_hostname is used.
2571 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2572 tweaks were necessary in order to get it to work (see also 21 below):
2573 (a) The code assumed that strncpy() returns a negative number on buffer
2574 overflow, which isn't the case. Replaced with Exim's string_format()
2576 (b) There were several signed/unsigned issues. I just did the minimum
2577 hacking in of casts. There is scope for a larger refactoring.
2578 (c) The code used strcasecmp() which is not a standard C function.
2579 Replaced with Exim's strcmpic() function.
2580 (d) The code set only $1; it now sets $auth1 as well.
2581 (e) A simple test gave the error "authentication client didn't specify
2582 service in request". It would seem that Dovecot has changed its
2583 interface. Fortunately there's a specification; I followed it and
2584 changed what the client sends and it appears to be working now.
2586 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2589 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2590 address A is aliased to B and C, where B exists and C does not. Without
2591 -v the output is "A verified" because verification stops after a
2592 successful redirection if more than one address is generated. However,
2593 with -v the child addresses are also verified. Exim was outputting "A
2594 failed to verify" and then showing the successful verification for C,
2595 with its parentage. It now outputs "B failed to verify", showing B's
2596 parentage before showing the successful verification of C.
2598 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2599 look up a TXT record in a specific list after matching in a combined
2602 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2603 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2604 they consult the DNS. I had assumed they would set it the way they
2605 wanted; and indeed my experiments on Linux seem to show that in some
2606 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2607 To be on the safe side, however, I have now made the interface to
2608 host_find_byname() similar to host_find_bydns(), with an argument
2609 containing the DNS resolver options. The host_find_byname() function now
2610 sets these options at its start, just as host_find_bydns() does. The smtp
2611 transport options dns_qualify_single and dns_search_parents are passed to
2612 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2613 of host_find_byname() use the default settings of RES_DEFNAMES
2614 (qualify_single) but not RES_DNSRCH (search_parents).
2616 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2617 spool_read_header() do less string testing, by means of a preliminary
2618 switch on the second character of optional "-foo" lines. (This is
2619 overdue, caused by the large number of possibilities that now exist.
2620 Originally there were few.) While I was there, I also converted the
2621 str(n)cmp tests so they don't re-test the leading "-" and the first
2622 character, in the hope this might squeeze out yet more improvement.
2624 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2625 flag allowing group syntax was set by the header_syntax check but not
2626 turned off, possible causing trouble later; (2) The flag was not being
2627 set at all for the header_verify test, causing "group"-style headers to
2628 be rejected. I have now set it in this case, and also caused header_
2629 verify to ignore an empty address taken from a group. While doing this, I
2630 came across some other cases where the code for allowing group syntax
2631 while scanning a header line wasn't quite right (mostly, not resetting
2632 the flag correctly in the right place). These bugs could have caused
2633 trouble for malformed header lines. I hope it is now all correct.
2635 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2636 with the "reply" argument non-NULL. The code, however (which originally
2637 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2638 but it didn't always do it. This confused somebody who was copying the
2639 code for some other use. I have removed all the tests.
2641 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2642 feature that was used to support insecure browsers during the U.S. crypto
2643 embargo. It requires special client support, and Exim is probably the
2644 only MTA that supported it -- and would never use it because real RSA is
2645 always available. This code has been removed, because it had the bad
2646 effect of slowing Exim down by computing (never used) parameters for the
2647 RSA_EXPORT functionality.
2649 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2650 authenticator to fail if there's a tab character in the incoming data
2651 (there should never be unless someone is messing about, as it's supposed
2652 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2653 if the connection is using TLS or if the remote IP is the same as the
2654 local IP, and the "valid-client-cert option" if a client certificate has
2657 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2658 authenticators. This can be used for authorization after authentication
2659 succeeds. (In the case of plaintext, it servers for both authentication
2662 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2663 if any retry times were supplied.
2665 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2666 connection, where there is no client IP address to check. In this
2667 situation, the verify now always succeeds.
2669 PH/25 Applied John Jetmore's -Mset patch.
2671 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2673 PH/27 In a string expansion for a processed (not raw) header when multiple
2674 headers of the same name were present, leading whitespace was being
2675 removed from all of them, but trailing whitespace was being removed only
2676 from the last one. Now trailing whitespace is removed from each header
2677 before concatenation. Completely empty headers in a concatenation (as
2678 before) are ignored.
2680 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2681 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2683 PH/29 [Removed. This was a change that I later backed out, and forgot to
2684 correct the ChangeLog entry (that I had efficiently created) before
2685 committing the later change.]
2687 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2688 address errors (4xx response to RCPT) over the same connection as other
2689 messages routed to the same hosts. Such deliveries are always "forced",
2690 so retry times are not inspected. This resulted in far too many retries
2691 for the affected addresses. The effect occurred only when there were more
2692 hosts than the hosts_max_try setting in the smtp transport when it had
2693 the 4xx errors. Those hosts that it had tried were not added to the list
2694 of hosts for which the message was waiting, so if all were tried, there
2695 was no problem. Two fixes have been applied:
2697 (i) If there are any address or message errors in an SMTP delivery, none
2698 of the hosts (tried or untried) are now added to the list of hosts
2699 for which the message is waiting, so the message should not be a
2700 candidate for sending over the same connection that was used for a
2701 successful delivery of some other message. This seems entirely
2702 reasonable: after all the message is NOT "waiting for some host".
2703 This is so "obvious" that I'm not sure why it wasn't done
2704 previously. Hope I haven't missed anything, but it can't do any
2705 harm, as the worst effect is to miss an optimization.
2707 (ii) If, despite (i), such a delivery is accidentally attempted, the
2708 routing retry time is respected, so at least it doesn't keep
2709 hammering the server.
2711 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2712 in ${readsocket because some servers need this prod.
2714 PH/32 Added some extra debug output when updating a wait-xxx database.
2716 PH/33 The hint "could be header name not terminated by colon", which has been
2717 given for certain expansion errors for a long time, was not being given
2718 for the ${if def:h_colon_omitted{... case.
2720 PH/34 The spec says: "With one important exception, whenever a domain list is
2721 being scanned, $domain contains the subject domain." There was at least
2722 one case where this was not true.
2724 PH/35 The error "getsockname() failed: connection reset by peer" was being
2725 written to the panic log as well as the main log, but it isn't really
2726 panic-worthy as it just means the connection died rather early on. I have
2727 removed the panic log writing for the ECONNRESET error when getsockname()
2730 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2731 runs only) independently of the message's sender address. This meant
2732 that, if the 4xx error was in fact related to the sender, a different
2733 message to the same recipient with a different sender could confuse
2734 things. In particular, this can happen when sending to a greylisting
2735 server, but other circumstances could also provoke similar problems.
2736 I have changed the default so that the retry time for these errors is now
2737 based a combination of the sender and recipient addresses. This change
2738 can be overridden by setting address_retry_include_sender=false in the
2741 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2742 remote server are returned as part of bounce messages. This was not
2743 happening for LMTP over a pipe (the lmtp transport), but now it is the
2744 same for both kinds of LMTP.
2746 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2747 in header lines that were in fact CNAMEs. This is no longer the case.
2749 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2750 and queue runs started by the daemon processed all messages. This has
2751 been fixed so that -R and -S can now usefully be given with -q<time>.
2753 PH/40 Import PCRE release 6.7 (fixes some bugs).
2755 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2757 PH/42 Give an error if -q is specified more than once.
2759 PH/43 Renamed the variables $interface_address and $interface_port as
2760 $received_ip_address and $received_port, to make it clear that these
2761 values apply to message reception, and not to the outgoing interface when
2762 a message is delivered. (The old names remain recognized, of course.)
2764 PH/44 There was no timeout on the connect() call when using a Unix domain
2765 socket in the ${readsocket expansion. There now is.
2767 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2768 be meaningful with "accept".
2770 SC/01 Eximstats V1.43
2771 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2773 SC/02 Eximstats V1.44
2774 Use a glob alias rather than an array ref in the generated
2775 parser. This improves both readability and performance.
2777 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2778 Collect SpamAssassin and rejection statistics.
2779 Don't display local sender or destination tables unless
2780 there is data to show.
2781 Added average volumes into the top table text output.
2783 SC/04 Eximstats V1.46
2784 Collect data on the number of addresses (recipients)
2785 as well as the number of messages.
2787 SC/05 Eximstats V1.47
2788 Added 'Message too big' to the list of mail rejection
2789 reasons (thanks to Marco Gaiarin).
2791 SC/06 Eximstats V1.48
2792 Mainlog lines which have GMT offsets and are too short to
2793 have a flag are now skipped.
2795 SC/07 Eximstats V1.49 (Alain Williams)
2796 Added the -emptyok flag.
2798 SC/08 Eximstats V1.50
2799 Fixes for obtaining the IP address from reject messages.
2801 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2802 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2803 whitespace changes from 4.64-PH/27
2805 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2808 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2809 are found, allow negative numbers in numeric criteria)
2811 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2813 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2814 to match changes made in 4.64-PH/43
2816 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2818 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2819 "domain" is actually an IP address, and force a failure. This locks out
2820 those revolvers/nameservers that support "A-for-A" lookups, in
2821 contravention of the specifications.
2823 PH/48 When a host name was looked up from an IP address, and the subsequent
2824 forward lookup of the name timed out, the host name was left in
2825 $sender_host_name, contrary to the specification.
2827 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2828 restricted to single-key lookups, Exim was not diagnosing an error if
2829 * or *@ was used with a query-style lookup.
2831 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2833 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2834 long ago, but noone interested enough thought of it. Let's just say that
2835 the "1.1" means that there are some new functions that weren't there at
2836 some point in the past.
2838 PH/51 Error processing for expansion failure of helo_data from an smtp
2839 transport during callout processing was broken.
2841 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2842 tested/used via the -bh/-bhc/-bs options.
2844 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2845 bug, fixed in subsequent PCRE releases).
2847 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2848 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2850 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2856 SC/01 Use a glob alias rather than an array ref in eximstats generated
2857 parser. This improves both readability and performance.
2859 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2860 Don't display local sender or destination tables in eximstats unless
2861 there is data to show.
2862 Added average volumes into the eximstats top table text output.
2864 SC/03 Collect data on the number of addresses (recipients) as well
2865 as the number of messages in eximstats.
2867 TF/01 Correct an error in the documentation for the redirect router. Exim
2868 does (usually) call initgroups() when daemonizing.
2870 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2871 with consistent privilege compared to when running as a daemon.
2873 TF/03 Note in the spec that $authenticated_id is not set for local
2874 submissions from trusted users.
2876 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2877 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2879 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2880 by adding some example configuration directives to the default
2881 configuration file. A little bit of work is required to uncomment the
2882 directives and define how usernames and passwords are checked, but
2883 there is now a framework to start from.
2885 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2886 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2887 without this. I don't know how relevant this is to other LDAP libraries.
2889 PH/02 Add the verb name to the "unknown ACL verb" error.
2891 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2893 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2895 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2896 directory not expanded when it should be if an expanded home directory
2897 was set for the address (which is overridden by the transport).
2899 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2902 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2903 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2904 because it is too late at that time, and has no effect.
2906 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2907 security issue with \' (bugzilla #107). I could not use the
2908 PQescapeStringConn() function, because it needs a PGconn value as one of
2911 PH/08 When testing addresses using -bt, indicate those final addresses that
2912 are duplicates that would not cause an additional delivery. At least one
2913 person was confused, thinking that -bt output corresponded to deliveries.
2914 (Suppressing duplicates isn't a good idea as you lose the information
2915 about possibly different redirections that led to the duplicates.)
2917 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2918 systems where poll() doesn't work, in particular OS X.
2920 PH/10 Added more information to debugging output for retry time not reached.
2922 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2923 operations in malware.c.
2925 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2928 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2929 syslog_duplication set false, log lines that would normally be written
2930 both the the main log and to the reject log were not written to syslog at
2933 PH/14 In the default configuration, change the use of "message" in ACL warn
2934 statements to "add_header".
2936 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2937 not followed by a command (e.g. "seen endif").
2939 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2940 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2943 PH/17 Added extra conditions to the default value of delay_warning_condition
2947 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2948 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2949 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2952 The Auto-Submitted: and various List- headers are standardised, whereas I
2953 don't think Precedence: ever was.
2955 PH/18 Refactored debugging code in route_finduser() to show more information,
2956 in particular, the error code if getpwnam() issues one.
2958 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2959 This is apparently needed in addition to the PH/07 change above to avoid
2960 any possible encoding problems.
2962 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2963 but not after initializing Perl.
2965 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2966 output them only if debugging. By default they are written stderr,
2967 apparently, which is not desirable.
2969 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2972 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2975 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2977 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2978 authenticated or an ident call has been made. Suppress the default
2979 values for $authenticated_id and $authenticated_sender (but permit -oMai
2980 and -oMas) when testing with -bh.
2982 PH/24 Re-jigged the order of the tests in the default configuration so that the
2983 tests for valid domains and recipients precede the DNS black list and CSA
2984 tests, on the grounds that those ones are more expensive.
2986 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2987 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2988 This bug exists in every version of Exim that I still have, right back to
2991 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2992 However, an attempt to turn on case-sensitivity in a regex key by
2993 including (?-i) didn't work because the subject string was already
2994 lowercased, and the effects were non-intuitive. It turns out that a
2995 one-line patch can be used to allow (?-i) to work as expected.
3001 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3002 other effects) broke the use of negated acl sub-conditions.
3004 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3007 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3008 "Deny" causes Exim to reject the incoming connection with a 554 error.
3009 Unfortunately, if there is a major crisis, such as a disk failure,
3010 tcp-wrappers gives "deny", whereas what one would like would be some
3011 kind of temporary error. A kludge has been added to help with this.
3012 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3013 554 error is used if errno is still zero or contains ENOENT (which occurs
3014 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3017 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3019 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3020 errors. Otherwise a message that provokes a temporary error (when other
3021 messages do not) can cause a whole host to time out.
3023 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3024 addresses were routed directly to files or pipes from a redirect router.
3025 File deliveries just didn't batch; pipe deliveries might have suffered
3028 PH/06 A failure to get a lock for a hints database would erroneously always say
3029 "Failed to get write lock", even when it was really a read lock.
3031 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3032 of 0600. This has been changed to use the value of the lockfile_mode
3033 option (which defaults to 0600).
3035 PH/08 Applied small patch from the Sieve maintainer.
3037 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3038 folder from quota calculations, a direct delivery into this folder messed
3039 up the contents of the maildirsize file. This was because the regex was
3040 used only to exclude .Trash (or whatever) when the size of the mailbox
3041 was calculated. There was no check that a delivery was happening into an
3042 excluded directory. This bug has been fixed by ignoring all quota
3043 processing for deliveries into excluded directories.
3045 PH/10 Added the maildirfolder_create_regex option to appendfile.
3051 PH/01 The code for finding all the local interface addresses on a FreeBSD
3052 system running IPv6 was broken. This may well have applied to all BSD
3053 systems, as well as to others that have similar system calls. The broken
3054 code found IPv4 interfaces correctly, but gave incorrect values for the
3055 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3056 that it would not match correctly against @[] and not recognize the IPv6
3059 PH/02 The ipliteral router was not recognizing addresses of the form user@
3060 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3062 PH/03 Added disable_ipv6.
3064 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3065 decoded form, because it is most often used to construct To: headers
3066 lines in autoreplies, and the decoded form may well be syntactically
3067 invalid. However, $reply_address has leading white space removed, and all
3068 newlines turned into spaces so that the autoreply transport does not
3071 PH/05 If group was specified without a user on a router, and no group or user
3072 was specified on a transport, the group from the router was ignored.
3074 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3075 for visible compile-time settings that can be used to change these
3076 numbers, for those that want even more. Backwards compatibility with old
3077 spool files has been maintained. However, going back to a previous Exim
3078 release will lost any variables that are in spool files.
3080 PH/07 Two small changes when running in the test harness: increase delay when
3081 passing a TCP/IP connection to a new process, in case the original
3082 process has to generate a bounce, and remove special handling of
3083 127.0.0.2 (sic), which is no longer necessary.
3085 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3086 be the same on different OS.
3088 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3091 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3092 whether --show-vars was specified or not
3094 JJ/02 exipick: Added support for new ACL variable spool format introduced
3097 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3098 syntactically invalid From: or Reply-to: line, and a filter used this to
3099 generate an autoreply, and therefore failed to obtain an address for the
3100 autoreply, Exim could try to deliver to a non-existent relative file
3101 name, causing unrelated and misleading errors. What now happens is that
3102 it logs this as a hard delivery error, but does not attempt to create a
3105 PH/11 The exinext utility has a -C option for testing purposes, but although
3106 the given file was scanned by exinext itself; it wasn't being passed on
3107 when Exim was called.
3109 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3110 an end-of-file indication when reading a command response.
3112 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3113 compiled. In many other places in Exim, IPv6 addresses are always
3114 recognized, so I have changed this. It also means that IPv4 domain
3115 literals of the form [IPV4:n.n.n.n] are now always recognized.
3117 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3118 used if the router is not running as root, for example, when verifying at
3119 ACL time, or when using -bh. The debugging output from this situation was
3120 non-existent - all you got was a failure to exec. I have made two
3123 (a) Failures to set uid/gid, the current directory, or a process leader
3124 in a subprocess such as that created by queryprogram now generate
3125 suitable debugging output when -d is set.
3127 (b) The queryprogram router detects when it is not running as root,
3128 outputs suitable debugging information if -d is set, and then runs
3129 the subprocess without attempting to change uid/gid.
3131 PH/15 Minor change to Makefile for building test_host (undocumented testing
3134 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3135 additional section of a DNS packet that returns MX or SRV records.
3136 Instead, it always explicitly searches for A/AAAA records. This avoids
3137 major problems that occur when a DNS server includes only records of one
3138 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3139 fixed another bug: if SRV records were looked up and the corresponding
3140 address records were *not* found in the additional section, the port
3141 values from the SRV records were lost.
3143 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3144 using the correct key (the original address) when searching the retry
3145 rules in order to find which one to use for generating the retry hint.
3147 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3148 adding the default one. Similarly, if it contains a Reply-To: header, the
3149 errors_reply_to option, if set, is not used.
3151 PH/19 When calculating a retry time, Exim used to measure the "time since
3152 failure" by looking at the "first failed" field in the retry record. Now
3153 it does not use this if it is later than than the arrival time of the
3154 message. Instead it uses the arrival time. This makes for better
3155 behaviour in cases where some deliveries succeed, thus re-setting the
3156 "first failed" field. An example is a quota failure for a huge message
3157 when small messages continue to be delivered. Without this change, the
3158 "time since failure" will always be short, possible causing more frequent
3159 delivery attempts for the huge message than are intended.
3160 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3162 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3163 $1, $2, $3) because the numerical variables can be reset during some
3164 expansion items (e.g. "match"), thereby losing the authentication data.
3166 PH/21 Make -bV show the size of off_t variables so that the test suite can
3167 decide whether to run tests for quotas > 2G.
3169 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3170 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3171 filecount value is greater than 2G or if a quota value is greater than 2G
3172 on a system where the size of off_t is not greater than 4, a panic error
3175 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3176 never match. The debug and -bh output now contains an explicit error
3177 message indicating a malformed IPv4 address or mask.
3179 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3180 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3181 PH/23 above applies.
3183 PH/25 Do not write to syslog when running in the test harness. The only
3184 occasion when this arises is a failure to open the main or panic logs
3185 (for which there is an explicit test).
3187 PH/26 Added the /no_tell option to "control=freeze".
3189 PH/27 If a host name lookup failed very early in a connection, for example, if
3190 the IP address matched host_lookup and the reverse lookup yielded a name
3191 that did not have a forward lookup, an error message of the form "no IP
3192 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3193 could be logged. Now it outputs the IP address instead of "NULL".
3195 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3196 allows the sender and the authenticated sender to be set when
3197 submitting a message from within Exim. Since child_open_exim() is
3198 documented for local_scan(), the new function should be too.
3200 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3201 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3202 results in an empty string is now treated as unset.
3204 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3206 PH/31 Added sender_verify_fail logging option.
3208 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3209 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3210 tidied the source and removed it altogether.
3212 PH/33 When a queue run was abandoned because the load average was too high, a
3213 log line was always written; now it is written only if the queue_run log
3214 selector is set. In addition, the log line for abandonment now contains
3215 information about the queue run such as the pid. This is always present
3216 in "start" and "stop" lines but was omitted from the "abandon" line.
3218 PH/34 Omit spaces between a header name and the colon in the error message that
3219 is given when verify = headers_syntax fails (if there are lots of them,
3220 the message gets confusing).
3222 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3223 names, as there are now some PTR records that contain slashes. This check
3224 is only to protect against broken name servers that fall over on strange
3225 characters, so the fact that it applies to all lookups doesn't matter.
3227 PH/36 Now that the new test suite is complete, we can remove some of the
3228 special code in Exim that was needed for the old test suite. For example,
3229 sorting DNS records because real resolvers return them in an arbitrary
3230 order. The new test suite's fake resolver always returns records in the
3233 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3234 bounces) except when queue_only is set, to avoid logging races between
3235 the different processes.
3237 PH/38 Panic-die if .include specifies a non-absolute path.
3239 PH/39 A tweak to the "H" retry rule from its user.
3241 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3242 a label. They prevented compilation on older perls.
3244 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3245 a warning to be raised on newish perls.
3247 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3248 on queue. Changes to match documented behaviour of showing count of
3249 messages matching specified criteria.
3251 PH/40 Changed the default ident timeout from 30s to 5s.
3253 PH/41 Added support for the use of login_cap features, on those BSD systems
3254 that have them, for controlling the resources used by pipe deliveries.
3256 PH/42 The content-scanning code uses fopen() to create files in which to put
3257 message data. Previously it was not paying any attention to the mode of
3258 the files. Exim runs with umask(0) because the rest of the code creates
3259 files with open(), and sets the required mode explicitly. Thus, these
3260 files were ending up world-writeable. This was not a big issue, because,
3261 being within the spool directory, they were not world-accessible. I have
3262 created a function called modefopen, which takes an additional mode
3263 argument. It sets umask(777), creates the file, chmods it to the required
3264 mode, then resets the umask. All the relevant calls to fopen() in the
3265 content scanning code have been changed to use this function.
3267 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3268 to 24 hours. This avoids potential overflow problems when processing G
3269 and H retry rules. I suspect nobody ever tinkers with this value.
3271 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3273 PH/45 When the plaintext authenticator is running as a client, the server's
3274 challenges are checked to ensure they are valid base64 strings. By
3275 default, the authentication attempt is cancelled if an invalid string is
3276 received. Setting client_ignore_invalid_base64 true ignores these errors.
3277 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3278 they are received. Thus, the responses can be made to depend on the
3279 challenges. If an invalid string is ignored, an empty string is placed in
3282 PH/46 Messages that are created by the autoreply transport now contains a
3283 References: header, in accordance with RFCs 2822 and 3834.
3285 PH/47 Added authenticated_sender_force to the smtp transport.
3287 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3289 PH/49 Installed latest patch from the Sieve maintainer.
3291 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3292 was also set, the appendfile transport was unnecessarily scanning a
3293 directory of message files (e.g. for maildir delivery) to find the count
3294 of files (along with the size), even though it did not need this
3295 information. It now does the scan only if it needs to find either the
3296 size of the count of files.
3298 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3300 PH/52 Two bugs concerned with error handling when the smtp transport is
3303 (i) Exim was not creating retry information for temporary errors given
3304 for individual recipients after the DATA command when the smtp transport
3305 was used in LMTP mode. This meant that they could be retried too
3306 frequently, and not timed out correctly.
3308 (ii) Exim was setting the flag that allows error details to be returned
3309 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3310 recipients that were returned after the DATA command.
3312 PH/53 This is related to PH/52, but is more general: for any failing address,
3313 when detailed error information was permitted to be returned to the
3314 sender, but the error was temporary, then after the final timeout, only
3315 "retry timeout exceeded" was returned. Now it returns the full error as
3316 well as "retry timeout exceeded".
3318 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3319 do this, and (what is worse) MTAs that accept it.
3321 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3322 will now be deprecated.
3324 PH/56 New os.c-cygwin from the Cygwin maintainer.
3326 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3327 formats (previously only available in exim formats via -bpr, -bpru,
3328 and -bpra. Now also available in native and exiqgrep formats)
3330 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3331 with very large, slow to parse queues
3333 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3335 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3337 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3338 responses to authentication challenges, though it was showing the
3339 challenges; (ii) I've removed the CR characters from the debug output for
3342 PH/58 Allow for the insertion of a newline as well as a space when a string
3343 is turned into more than one encoded-word during RFC 2047 encoding. The
3344 Sieve code now uses this.
3346 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3347 data_4xx, lost_connection, tls_required.
3349 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3350 message was being sent as an SMTP response.
3352 PH/61 Add -l and -k options to exicyclog.
3354 PH/62 When verifying, if an address was redirected to one new address, so that
3355 verification continued, and the new address failed or deferred after
3356 having set something in $address_data, the value of $address_data was not
3357 passed back to the ACL. This was different to the case when no
3358 redirection occurred. The value is now passed back in both cases.
3360 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3361 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3362 use login_cap.h, so on its own it isn't the distinguishing feature. The
3363 new name refers directly to the setclassresources() function.
3365 PH/65 Added configuration files for NetBSD3.
3367 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3369 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3370 is preferred over IPv4.
3372 PH/68 The bounce_return_message and bounce_return_body options were not being
3373 honoured for bounces generated during the reception of non-SMTP messages.
3374 In particular, this applied to messages rejected by the ACL. This bug has
3375 been fixed. However, if bounce_return_message is true and bounce_return_
3376 body is false, the headers that are returned for a non-SMTP message
3377 include only those that have been read before the error was detected.
3378 (In the case of an ACL rejection, they have all been read.)
3380 PH/69 The HTML version of the specification is now built in a directory called
3381 spec_html instead of spec.html, because the latter looks like a path with
3382 a MIME-type, and this confuses some software.
3384 PH/70 Catch two compiler warnings in sieve.c.
3386 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3387 function verify_get_ident() calls ip_connect() to connect a socket, but
3388 if the "connect()" function timed out, ip_connect() used to close the
3389 socket. However, verify_get_ident() also closes the socket later, and in
3390 between Exim writes to the log, which may get opened at this point. When
3391 the socket was closed in ip_connect(), the log could get the same file
3392 descriptor number as the socket. This naturally causes chaos. The fix is
3393 not to close the socket in ip_connect(); the socket should be closed by
3394 the function that creates it. There was only one place in the code where
3395 this was missing, in the iplookup router, which I don't think anybody now
3396 uses, but I've fixed it anyway.
3398 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3399 well as to direct DNS lookups. Otherwise the handling of names in host
3400 lists is inconsistent and therefore confusing.
3406 PH/01 Two changes to the default runtime configuration:
3408 (1) Move the checks for relay_from_hosts and authenticated clients from
3409 after to before the (commented out) DNS black list checks.
3411 (2) Add control=submission to the relay_from_hosts and authenticated
3412 clients checks, on the grounds that messages accepted by these
3413 statements are most likely to be submissions.
3415 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3417 (1) Generate an error if the third argument for the ${prvs expansion is
3420 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3423 (3) Reset the variables that are obtained from the first argument of
3424 ${prvscheck and used in the second argument before leaving the code,
3425 because their memory is reclaimed, so using them afterwards may do
3428 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3429 one (it's much easier than Tom thought :-).
3431 (5) Because of (4), we can now allow for the use of $prvscheck_result
3432 inside the third argument.
3434 PH/03 For some reason, the default setting of PATH when running a command from
3435 a pipe transport was just "/usr/bin". I have changed it to
3438 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3439 anything to be listed in the output from -bV.
3441 PH/05 When a filter generated an autoreply, the entire To: header line was
3442 quoted in the delivery log line, like this:
3444 => >A.N.Other <ano@some.domain> <original@ddress> ...
3446 This has been changed so that it extracts the operative address. There
3447 may be more than one such address. If so, they are comma-separated, like
3450 => >ano@some.domain,ona@other.domain <original@ddress> ...
3452 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3453 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3454 not being looked up in the rDNS to get a host name, Exim was showing the
3455 IP address twice in Received: lines, even though the IP addresses were
3456 identical. For example:
3458 Received: from [1.2.3.4] (helo=[1.2.3.4])
3460 However, if the real host name was known, it was omitting the HELO data
3461 if it matched the actual IP address. This has been tidied up so that it
3462 doesn't show the same IP address twice.
3464 PH/07 When both +timestamp and +memory debugging was on, the value given by
3465 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3466 called by the debug printing, thereby overwriting the timestamp buffer.
3467 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3469 PH/08 When the original message was included in an autoreply transport, it
3470 always said "this is a copy of the message, including all the headers",
3471 even if body_only or headers_only was set. It now gives an appropriate
3474 PH/09 Applied a patch from the Sieve maintainer which:
3476 o fixes some comments
3477 o adds the (disabled) notify extension core
3478 o adds some debug output for the result of if/elsif tests
3479 o points to the current vacation draft in the documentation
3480 and documents the missing references header update
3484 o fixes a bug in processing the envelope test (when testing
3485 multiple envelope elements, the last element determined the
3488 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3489 Electronic Mail") by including:
3491 Auto-submitted: auto-generated
3493 in the messages that it generates (bounce messages and others, such as
3494 warnings). In the case of bounce messages for non-SMTP messages, there was
3495 also a typo: it was using "Auto_submitted" (underscore instead of
3496 hyphen). Since every message generated by Exim is necessarily in response
3497 to another message, thes have all been changed to:
3499 Auto-Submitted: auto-replied
3501 in accordance with these statements in the RFC:
3503 The auto-replied keyword:
3505 - SHOULD be used on messages sent in direct response to another
3506 message by an automatic process,
3508 - MUST NOT be used on manually-generated messages,
3510 - MAY be used on Delivery Status Notifications (DSNs) and Message
3511 Disposition Notifications (MDNs),
3513 - MUST NOT be used on messages generated by automatic or periodic
3514 processes, except for messages which are automatic responses to
3517 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3518 to the default Received: header definition.
3520 PH/12 Added log selector acl_warn_skipped (default on).
3522 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3523 variables because (a) they are in the wrong storage pool and (b) even if
3524 they were copied, it wouldn't work properly because of the caching.
3526 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3527 checking when decoding. Apparently there are clients that generate
3528 overlong encoded strings. Why am I not surprised?
3530 PH/15 If the first argument of "${if match_address" was not empty, but did not
3531 contain an "@" character, Exim crashed. Now it writes a panic log message
3532 and treats the condition as false.
3534 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3536 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3537 extension "envelope-auth". The code is finished and in agreement with
3538 other implementations, but there is no documentation so far and in fact,
3539 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3540 not changing the active code.
3542 Print executed "if" and "elsif" statements when debugging is used. This
3543 helps a great deal to understand what a filter does.
3545 Document more things not specified clearly in RFC3028. I had all this
3546 sorted out, when out of a sudden new issues came to my mind. Oops."
3548 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3551 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3552 canonical form (as documented). However, after a host name lookup from
3553 the IP address, check_host() was doing a simple string comparison with
3554 addresses acquired from the DNS when checking that the found name did
3555 have the original IP as one of its addresses. Since any found IPv6
3556 addresses are likely to be in abbreviated form, the comparison could
3557 fail. Luckily, there already exists a function for doing the comparison
3558 by converting both addresses to binary, so now that is used instead of
3559 the text comparison.
3561 PH/20 There was another similar case to PH/19, when a complete host name was
3562 given in a host list; looking up its IP address could give an abbreviated
3563 form, whereas the current host's name might or might not be abbreviated.
3564 The same fix has been applied.
3570 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3571 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3574 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3575 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3577 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3579 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3580 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3581 Also, for "Auto-Submitted", treat anything other than "no" as
3582 non-personal, in accordance with RFC 3834. (Previously it treated
3583 anything starting "auto-" as non-personal.)
3585 TF/01 The control=submission/name=... option had a problem with syntax
3586 errors if the name included a slash character. The /name= option
3587 now slurps the rest of the string, so it can include any characters
3588 but it must come last in the list of options (after /sender_retain
3591 PH/05 Some modifications to the interface to the fake nameserver for the new
3599 TK/01 Added the "success_on_redirect" address verification option. See
3600 NewStuff for rationale and an example.
3602 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3604 PH/02 Patch to exigrep to allow it to work on syslog lines.
3606 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3607 fread() to skip over the body file's header line, because in Cygwin the
3608 header line is locked and is inaccessible.
3610 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3611 co-exist for some time) to make it clear that it is the Exim ID that is
3612 referenced, not the Message-ID: header line.
3614 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3615 string_format() function, because snprintf() does not exist on all
3618 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3619 ${stat: expansion item.
3621 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3622 protocol synchronization error", to keep the pedants happy.
3624 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3625 well as for IRIX systems, when gcc is being used. See the host.c source
3628 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3630 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3633 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3634 transport and to the smtp transport in LMTP mode.
3636 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3638 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3639 run a filter in a subprocess. This could lead to confusion in subsequent
3640 lookups in the parent process. There should also be a search_tidyup() at
3641 the end of the subprocess.
3643 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3644 only if the host matched helo_try_verify_hosts, which caused the
3645 verification to occur when the EHLO/HELO command was issued. The ACL just
3646 tested the remembered result. Now, if a previous verification attempt has
3647 not happened, "verify = helo" does it there and then.
3649 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3651 TK/03 Fix log output including CR from clamd.
3653 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3654 exist provoked a memory error which could cause a segfault.
3656 PH/15 Installed PCRE 6.2
3658 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3660 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3661 of the problem. Specifically, suggested +O2 rather than +O1 for the
3664 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3666 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3667 record, the port was not correctly logged when the outgoing_port log
3668 selector was set (it logged the transort's default port).
3670 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3671 fallback_hosts, and "hosts" in the smtp transport.
3673 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3674 host errors such as "Connection refused".
3676 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3677 authentication with radiusclient 0.4.9:
3679 - Error returned from rc_read_config was caught wrongly
3680 - Username/password not passed on to radius server due to wrong length.
3682 The presumption is that some radiusclient API changes for 4.51/PH/17
3683 were not taken care of correctly. The code is still untested by me (my
3684 Linux distribution still has 0.3.2 of radiusclient), but it was
3685 contributed by a Radius user.
3687 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3688 expanding the "port" option of the smtp transport.
3690 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3691 while reading a MIME header. Thanks to Tom Hughes for a patch.
3693 PH/24 Include config.h inside local_scan.h so that configuration settings are
3696 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3697 that in an ACL for RCPT (for example), you can examine exactly what was
3700 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3701 commands, but it was not correctly comparing the address with the actual
3702 client host address. Thus, it would show the EHLO address in Received:
3703 header lines when this was not necessary.
3705 PH/27 Added the % operator to ${eval:}.
3707 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3708 it should be ignoring failures (because with -C, for example, it has lost
3709 privilege). It wasn't ignoring creation failures other than "already
3712 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3715 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3716 because an input line was too long, either on its own, or by virtue of
3717 too many continuations, the temporary file was not being removed, and the
3718 return code was incorrect.
3720 PH/31 Missing "BOOL" in function definition in filtertest.c.
3722 PH/32 Applied Sieve patches from the maintainer.
3724 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3726 PH/33 Added "verify = not_blind".
3728 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3729 Local/Makefile (with some defaults set). These are used in built scripts
3730 such as exicyclog, but they have never been used in the exim_install
3731 script (though there are many overriding facilities there). I have
3732 arranged that the exim_install script now takes note of these two
3735 PH/35 Installed configuration files for Dragonfly.
3737 PH/36 When a locally submitted message by a trusted user did not contain a
3738 From: header, and the sender address was obtained from -f or from an SMTP
3739 MAIL command, and the trusted user did not use -F to supply a sender
3740 name, $originator_name was incorrectly used when constructing a From:
3741 header. Furthermore, $originator_name was used for submission mode
3742 messages from external hosts without From: headers in a similar way,
3743 which is clearly wrong.
3745 PH/37 Added control=suppress_local_fixups.
3747 PH/38 When log_selector = +received_sender was set, and the addition of the
3748 sender made the log line's construction buffer exactly full, or one byte
3749 less than full, an overflow happened when the terminating "\n" was
3752 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3753 when the result of a list match is failure because a DNS lookup failed.
3755 PH/40 RM_COMMAND is now used in the building process.
3757 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3758 the "build-* directories that it finds.
3760 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3761 address in a domain literal was a prefix of an interface address.
3763 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3764 when verifying a sender address, unless rewrite_headers is false.
3766 PH/44 Wrote a long comment about why errors_to addresses are verified as
3767 recipients, not senders.
3769 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3770 the ratelimit ACL was added.
3772 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3774 PH/46 Added extra information about PostgreSQL errors to the error string.
3776 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3777 avoiding the need to install special zones in a real server. This is
3778 backwards compatible; if it can't find the fake resolver, it drops back.
3779 Thus, both old and new test suites can be run.
3781 TF/02 Added util/ratelimit.pl
3783 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3784 clock is set back in time.
3786 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3787 Candler <B.Candler@pobox.com>.
3789 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3790 OK for addresses that are the result of redirections.
3792 PH/48 A number of further additions for the benefit of the new test suite,
3793 including a fake gethostbyname() that interfaces to the fake DNS resolver
3796 TF/06 The fix for widen_domains has also been applied to qualify_single and
3797 search_parents which are the other dnslookup options that can cause
3800 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3803 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3805 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3806 service" if it encountered an MX record with an empty target hostname.
3807 The message is now "an MX or SRV record indicated no SMTP service".
3809 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3810 if the DNS of the sending site is misconfigured. This is quite a
3811 common situation. This change restores the behaviour of treating a
3812 helo verification defer as a failure.
3814 PH/51 If self=fail was set on a router, the bounce message did not include the
3815 actual error message.
3821 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3823 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3824 command itself ended in error, the underlying message about the transport
3825 filter timeout was being overwritten with the pipe command error. Now the
3826 underlying error message should be appended to the second error message.
3828 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3830 PH/02 Reduce the amount of output that "make" produces by default. Full output
3831 can still be requested.
3833 PH/03 The warning log line about a condition test deferring for a "warn" verb
3834 was being output only once per connection, rather than after each
3835 occurrence (because it was using the same function as for successful
3836 "warn" verbs). This seems wrong, so I have changed it.
3838 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3839 it should not have, which might have caused a crash in the right
3840 circumstances, but probably never did.
3842 PH/04 Installed a modified version of Tony Finch's patch to make submission
3843 mode fix the return path as well as the Sender: header line, and to
3844 add a /name= option so that you can make the user's friendly name appear
3847 TF/03 Added the control = fakedefer ACL modifier.
3849 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3850 Mark Lowes for thorough testing.
3852 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3854 TK/03 Merged latest SRS patch from Miles Wilton.
3856 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3857 with the definition in sysexits.h (which is #included earlier).
3858 Fortunately, Exim does not actually use EX_OK. The code used to try to
3859 preserve the sysexits.h value, by assuming that macro definitions were
3860 scanned for macro replacements. I have been disabused of this notion,
3861 so now the code just undefines EX_OK before #including unistd.h.
3863 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3864 in the smtp transport. When a block could not be written in a single
3865 write() function, the timeout was being re-applied to each part-write.
3866 This seems wrong - if the receiver was accepting one byte at a time it
3867 would take for ever. The timeout is now adjusted when this happens. It
3868 doesn't have to be particularly precise.
3870 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3871 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3873 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3874 without a domain if the check to <postmaster@domain> fails.
3876 SC/01 Eximstats: added -xls and the ability to specify output files
3877 (patch written by Frank Heydlauf).
3879 SC/02 Eximstats: use FileHandles for outputting results.
3881 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3883 SC/04 Eximstats: fixed display of large numbers with -nvr option
3885 SC/05 Eximstats: fixed merging of reports with empty tables.
3887 SC/06 Eximstats: added the -include_original_destination flag
3889 SC/07 Eximstats: removed tabs and trailing whitespace.
3891 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3893 TK/06 MBOX spool code: Add real "From " MBOX separator line
3894 so the .eml file is really in mbox format (even though
3895 most programs do not really care). Patch from Alex Miller.
3897 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3898 The latter is generated from $received_to and is only set if the
3899 message has one envelope recipient. SA can use these headers,
3900 obviously out-of-the-box. Patch from Alex Miller.
3902 PH/08 The ${def test on a variable was returning false if the variable's
3903 value was "0", contrary to what the specification has always said!
3904 The result should be true unless the variable is empty.
3906 PH/09 The syntax error of a character other than { following "${if
3907 def:variable_name" (after optional whitespace) was not being diagnosed.
3908 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3909 accidental colon was present, for example, could give incorrect results.
3911 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3912 result is used (not including appendfile, where other changes are about
3915 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3916 This involved changing a lot of size variables from int to off_t. It
3917 should work with maildirs and everything.
3919 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3920 spamd dying while we are connected to it.
3922 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3925 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3926 function for BDB 4.3.
3928 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3930 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3931 the sources for PCRE have been re-arranged and more files are now
3934 PH/15 The code I had for printing potentially long long variables in PH/11
3935 above was not the best (it lost precision). The length of off_t variables
3936 is now inspected at build time, and an appropriate printing format (%ld
3937 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3938 to be "long long int" or "long int". This is needed for the internal
3939 formatting function string_vformat().
3941 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3942 the configuration file to be ":syslog", then the script "guesses" where
3943 the logs files are, rather than using the compiled in default. In our
3944 case the guess is not the same as the compiled default, so the script
3945 suddenly stopped working when I started to use syslog. The patch checks
3946 to see if log_file_path is "". If so, it attempts to read it from exim
3947 with no configuration file to get the compiled in version, before it
3948 falls back to the previous guessing code."
3950 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3951 implementing BATV in an Exim configuration. See NewStuff for the gory
3954 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3955 Makefile that are specific to HP-UX.
3957 PH/18 If the "use_postmaster" option was set for a recipient callout together
3958 with the "random" option, the postmaster address was used as the MAIL
3959 FROM address for the random test, but not for the subsequent recipient
3960 test. It is now used for both.
3962 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3963 patch removes a few documentation additions to RFC 3028, because the
3964 latest draft now contains them. It adds the new en;ascii-case comparator
3965 and a new error check for 8bit text in MIME parts. Comparator and
3966 require names are now matched exactly. I enabled the subaddress
3967 extension, but it is not well tested yet (read: it works for me)."
3969 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3970 rework some of the code of TK/09 above to avoid the hardwired use of
3971 "%lld" and "long long". Replaced the call to snprintf() with a call to
3974 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3975 records point to non-existent hosts", "retry timeout exceeded", and
3976 "retry time not reached for any host after a long failure period".
3978 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3979 experimental DomainKeys support:
3981 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3982 (2) On an error such as an illegally used "control", the wrong name for
3983 the control was given.
3985 These problems did NOT occur unless DomainKeys support was compiled.
3987 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3989 PH/24 Added ${if match_ip condition.
3991 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3992 databases so that it will be absolutely obvious if a crash occurs in the
3993 DB library. This is a regular occurrence (often caused by mis-matched
3996 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3997 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3998 picked up on a user's system that detects such things. There doesn't seem
3999 to be a gcc warning option for this - only an attribute that has to be
4000 put on the function's prototype. It seems that in Fedora Core 4 they have
4001 set this on a number of new functions. No doubt there will be more in due
4004 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4005 specify a transport. However, if an address that was verified by such a
4006 router was the subject of a callout, Exim crashed because it tried to
4007 read the rcpt_include_affixes from the non-existent transport. Now it
4008 just assumes that the setting of that option is false. This bug was
4009 introduced by 4.51/PH/31.
4011 PH/28 Changed -d+all to exclude +memory, because that information is very
4012 rarely of interest, but it makes the output a lot bigger. People tend to
4013 do -d+all out of habit.
4015 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4016 code in os-type was giving problems when libc.so lives in lib64, like on
4019 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4020 aren't the modern standard, and it seems that some systems' include files
4021 don't always have them. Exim was already checking for some of the newer
4022 ones like T_AAAA, and defining it itself. I've added checks for all the
4023 record types that Exim uses.
4025 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4026 not automatically generating a new one, as it is supposed to. This
4027 prevented TLS from working. If the file did exist, but contained invalid
4028 data, a new version was generated, as expected. It was only the case of a
4029 non-existent file that was broken.
4031 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4032 with a change in libdomainkeys > 0.64.
4034 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4035 from DNS. If the selector record carries the flag, it now has
4036 precedence over the domain-wide flag.
4038 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4040 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4041 the use of a port name that isn't defined in /etc/services) occurred, the
4042 message was deferred as in a normal delivery, and thus remained on the
4043 spool, instead of being failed because of the mua_wrapper setting. This
4044 is now fixed, and I tidied up some of the mua_wrapper messages at the
4047 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4048 the messages in a hash of arrays rather than using individual hashes.
4049 This is a bit cleaner and results in dramatic memory savings, albeit
4050 at a slight CPU cost.
4052 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4053 as requested by Marc Sherman.
4055 SC/10 Eximstats: added histograms for user specified patterns as requested
4058 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4060 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4061 fopen() in the content-scanning modules that did not already have it.
4067 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4068 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4070 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4072 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4074 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4075 to test Sieve filters that use "vacation".
4077 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4078 that changes the way the GnuTLS parameters are stored in the cache file.
4079 The new format can be generated externally. For backward compatibility,
4080 if the data in the cache doesn't make sense, Exim assumes it has read an
4081 old-format file, and it generates new data and writes a new file. This
4082 means that you can't go back to an older release without removing the
4085 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4086 work if there are any delivery delays because "one_time" forces the
4087 parent to be marked "delivered", so its unseen clone is never tried
4088 again. For this reason, Exim now forbids the simultaneous setting of
4091 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4092 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4093 entry if you want to know the details. The fix, however, neglected to
4094 consider the case when local delivery batching is involved. The test for
4095 "previously delivered" was not happening when checking to see if an
4096 address could be batched with a previous (undelivered) one; under
4097 certain circumstances this could lead to multiple deliveries to the same
4100 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4101 in its include files, and this causes problems building Exim.
4103 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4104 header_syntax) but Exim was just ignoring anything given after a slash.
4105 In particular, this caused confusion with an attempt to use "verify =
4106 reverse_host_lookup/defer_ok". An error is now given when options are
4107 supplied for verify items that do not have them. (Maybe reverse_host_
4108 lookup should have a defer_ok option, but that's a different point.)
4110 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4111 defined by RFC 821) to 2048, because there were problems with some AUTH
4112 commands, and RFC 1869 says the size should be increased for extended
4113 SMTP commands that take arguments.
4115 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4118 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4119 "unknown" error; now it says that the functionality isn't in the binary.
4121 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4122 an address' error message when a string expansion fails (syntax or
4123 whatever). Otherwise the password may appear in the log. Following change
4124 PH/42 below, there is no longer a chance of it appearing in a bounce
4127 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4129 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4130 of its addresses was ever tried. (Bugzilla bug #2.)
4132 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4133 the result incorrectly in the debug output. (It correctly added a newline
4134 to what was transported.)
4136 TF/01 Added $received_time.
4138 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4139 commented out examples of how to interface to a virus scanner and to
4140 SpamAssassin. Also added commented examples of av_scanner and
4141 spamd_address settings.
4143 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4144 and controls are allowed in which ACLs. There were a couple of minor
4145 errors. Some of the entries in the conditions table (which is a table of
4146 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4147 negation of where the condition IS allowed.
4149 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4151 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4152 header file does not have a version number, so I've had to invent a new
4153 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4154 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4155 radiusclient), but it was contributed by a Radius user.
4157 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4158 files or directories, for OpenSSL.
4160 PH/19 When an Exim process that is running as root has to create an Exim log
4161 file, it does so in a subprocess that runs as exim:exim so as to get the
4162 ownership right at creation (otherwise, other Exim processes might see
4163 the file with the wrong ownership). There was no test for failure of this
4164 fork() call, which would lead to the process getting stuck as it waited
4165 for a non-existent subprocess. Forks do occasionally fail when resources
4166 run out. I reviewed all the other calls to fork(); they all seem to check
4169 PH/20 When checking for unexpected SMTP input at connect time (before writing
4170 the banner), Exim was not dealing correctly with a non-positive return
4171 from the read() function. If the client had disconnected by this time,
4172 the result was a log entry for a synchronization error with an empty
4173 string after "input=" when read() returned zero. If read() returned -1
4174 (an event I could not check), uninitialized data bytes were printed.
4175 There were reports of junk text (parts of files, etc) appearing after
4178 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4180 PH/22 Added support for macro redefinition, and (re)definition in between
4181 driver and ACL definitions.
4183 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4184 forgetting to use the resulting value; it was using the unexpanded value.
4186 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4187 hadn't been configured. The fix is from Juergen Kreileder, who
4188 understands it better than I do:
4190 "Here's what I see happening with three configured cyrus_sasl
4191 authenticators configured (plain, login, cram-md5):
4193 On startup auth_cyrus_sasl_init() gets called for each of these.
4194 This means three calls to sasl_listmech() without a specified mech_list.
4195 => SASL tests which mechs of all available mechs actually work
4196 => three warnings about OTP not working
4197 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4199 With the patch, sasl_listmech() also gets called three times. But now
4200 SASL's mech_list option is set to the server_mech specified in the the
4201 authenticator. Or in other words, the answer from sasl_listmech()
4202 gets limited to just the mech you're testing for (which is different
4204 => the return list contains just 'plain' or 'login', 'cram-md5' or
4205 nothing depending on the value of ob->server_mech.
4207 I've just tested the patch: Authentication still works fine,
4208 unavailable mechs specified in the exim configuration are still
4209 caught, and the auth.log warnings about OTP are gone."
4211 PH/25 When debugging is enabled, the contents of the command line are added
4212 to the debugging output, even when log_selector=+arguments is not
4215 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4216 answer is "GNU", and only if the return is "GNU/something" is the answer
4219 PH/27 $acl_verify_message is now set immediately after the failure of a
4220 verification in an ACL, and so is available in subsequent modifiers. In
4221 particular, the message can be preserved by coding like this:
4223 warn !verify = sender
4224 set acl_m0 = $acl_verify_message
4226 Previously, $acl_verify_message was set only while expanding "message"
4227 and "log_message" when a very denied access.
4229 PH/28 Modified OS/os.c-Linux with
4231 -#ifndef OS_LOAD_AVERAGE
4232 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4234 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4235 nomenclature these days.)
4237 PH/29 Installed patch from the Sieve maintainer that adds the options
4238 sieve_useraddress and sieve_subaddress to the redirect router.
4240 PH/30 In these circumstances:
4241 . Two addresses routed to the same list of hosts;
4242 . First host does not offer TLS;
4243 . First host accepts first address;
4244 . First host gives temporary error to second address;
4245 . Second host offers TLS and a TLS session is established;
4246 . Second host accepts second address.
4247 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4248 and peerdn, if requested) that were in fact used only for the second
4251 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4252 attention to any local part prefix or suffix that was matched by the
4253 router that accepted the address. It now behaves in the same way as it
4254 does for delivery: the affixes are removed from the local part unless
4255 rcpt_include_affixes is set on the transport.
4257 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4258 timeout during the DATA phase of an incoming message.
4260 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4261 applied a patch sanctioned by the Sieve maintainer.
4263 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4264 the uid or gid is negative. A case of a negative gid caused this to be
4265 noticed. The fix allows for either to be negative.
4267 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4268 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4271 PH/36 The change PH/12 above was broken. Fixed it.
4273 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4274 the grounds that routing the same address twice would always produce the
4275 same answer. This might have been true once, but it is certainly no
4276 longer true now. Routing a child address may depend on the previous
4277 routing that produced that child. Some complicated redirection strategies
4278 went wrong when messages had multiple recipients, and made Exim's
4279 behaviour dependent on the order in which the addresses were given.
4281 I have moved the duplicate checking until after the routing is complete.
4282 Exim scans the addresses that are assigned to local and remote
4283 transports, and removes any duplicates. This means that more work will be
4284 done, as duplicates will always all be routed, but duplicates are
4285 presumably rare, so I don't expect this is of any significance.
4287 For deliveries to pipes, files, and autoreplies, the duplicate checking
4288 still happens during the routing process, since they are not going to be
4291 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4292 It corrects a timeout issue with spamd. This is Ian's comment: "The
4293 background is that sometimes spamd either never reads data from a
4294 connection it has accepted, or it never writes response data. The exiscan
4295 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4296 blindly assumes that writes won't block so it may never time out."
4298 PH/39 Allow G after quota size as well as K and M.
4300 PH/40 The value set for $authenticated_id in an authenticator may not contain
4301 binary zeroes or newlines because the value is written to log lines and
4302 to spool files. There was no check on this. Now the value is run through
4303 the string_printing() function so that such characters are converted to
4304 printable escape sequences.
4306 PH/41 $message_linecount is a new variable that contains the total number of
4307 lines in the message. Compare $body_linecount, which is the count for the
4310 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4311 bounce and delay warning messages, except in certain special cases, which
4314 (a) An SMTP error message from a remote host;
4315 (b) A message specified in a :fail: redirection;
4316 (c) A message specified in a "fail" command in a system filter;
4317 (d) A message specified in a FAIL return from the queryprogram router;
4318 (e) A message specified by the cannot_route_message router option.
4320 In these cases only, Exim does include the error details in bounce and
4321 warning messages. There are also a few cases where bland messages such
4322 as "unrouteable address" or "local delivery error" are given.
4324 PH/43 $value is now also set for the "else" part of a ${run expansion.
4326 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4327 being worked on, but at least Exim now implements the latest version to
4330 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4331 process to complete was treated as a delivery failure, a timeout while
4332 writing the message to the pipe was logged, but erroneously treated as a
4333 successful delivery. Such timeouts include transport filter timeouts. For
4334 consistency with the overall process timeout, these timeouts are now
4335 treated as errors, giving rise to delivery failures by default. However,
4336 there is now a new Boolean option for the pipe transport called
4337 timeout_defer, which, if set TRUE, converts the failures into defers for
4338 both kinds of timeout. A transport filter timeout is now identified in
4341 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4342 systems where "make" and "gmake" are different, calling "gmake" at top
4343 level broke things. I've arranged for the value of $(MAKE) to be passed
4344 from the Makefile to this script so that it can call the same version of
4348 A note about Exim versions 4.44 and 4.50
4349 ----------------------------------------
4351 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4352 changes of various kinds. As a consequence, a big documentation update was
4353 needed. This delayed the release for rather longer than seemed good, especially
4354 in the light of a couple of (minor) security issues. Therefore, the changes
4355 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4356 release. So 4.44 and 4.50 are in effect two different branches that both start
4359 I have left the 4.50 change log unchanged; it contains all the changes since
4360 4.43. The change log for 4.44 is below; many of its items are identical to
4361 those for 4.50. This seems to be the most sensible way to preserve the
4362 historical information.
4368 1. Minor wording change to the doc/README.SIEVE file.
4370 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4371 computation of the current number of files was incorrect.
4373 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4374 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4377 4. Give more explanation in the error message when the command for a transport
4378 filter fails to execute.
4380 5. There are several places where Exim runs a non-Exim command in a
4381 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4382 was being done only for the command run by the queryprogram router. It is
4383 now done for all such subprocesses. The other cases are: ${run, transport
4384 filters, and the commands run by the lmtp and pipe transports.
4386 6. Added CONFIGURE_GROUP build-time option.
4388 7. Some older OS have a limit of 256 on the maximum number of file
4389 descriptors. Exim was using setrlimit() to set 1000 as a large value
4390 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4391 systems. I've change it so that if it can't get 1000, it tries for 256.
4393 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4394 was an oversight, and furthermore, ever since the addition of extra
4395 controls (e.g. 4.43/32), the checks on when to allow different forms of
4396 "control" were broken. There should now be diagnostics for all cases when a
4397 control that does not make sense is encountered.
4399 9. Added the /retain_sender option to "control=submission".
4401 10. $recipients is now available in the predata ACL (oversight).
4403 11. Tidy the search cache before the fork to do a delivery from a message
4404 received from the command line. Otherwise the child will trigger a lookup
4405 failure and thereby defer the delivery if it tries to use (for example) a
4406 cached ldap connection that the parent has called unbind on.
4408 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4409 of $address_data from the recipient verification was clobbered by the
4410 sender verification.
4412 13. The value of address_data from a sender verification is now available in
4413 $sender_address_data in subsequent conditions in the ACL statement.
4415 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4417 15. Added a new option "connect=<time>" to callout options, to set a different
4420 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4421 was its contents. (It was OK if the option was not defined at all.)
4423 17. A "Completed" log line is now written for messages that are removed from
4424 the spool by the -Mrm option.
4426 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4427 information about exactly what failed.
4429 19. Added -dd to debug only the daemon process.
4431 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4432 handles timeouts, both on the server side and network timeouts. Renamed the
4433 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4435 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4436 It is now set to "smtps".
4438 22. $host_address is now set to the target address during the checking of
4439 ignore_target_hosts.
4441 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4442 being passed; this would have caused $sender_host_name to have been used if
4443 matching the list had actually called for a host name (not very likely,
4444 since this list is usually IP addresses). A host name is now passed as
4447 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4448 code that specifies a non-restarting handler (typically sigaction() in
4449 modern systems) in an attempt to fix a rare and obscure crash bug.
4451 25. Narrowed the window for a race in the daemon that could cause it to ignore
4452 SIGCHLD signals. This is not a major problem, because they are used only to
4453 wake it up if nothing else does.
4455 26. A malformed maildirsize file could cause Exim to calculate negative values
4456 for the mailbox size or file count. Odd effects could occur as a result.
4457 The maildirsize information is now recalculated if the size or filecount
4460 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4461 support for a long time. Removed HAVE_SYS_VFS_H.
4463 28. Installed the latest version of exipick from John Jetmore.
4465 29. In an address list, if the pattern was not a regular expression, an empty
4466 subject address (from a bounce message) matched only if the pattern was an
4467 empty string. Non-empty patterns were not even tested. This was the wrong
4468 because it is perfectly reasonable to use an empty address as part of a
4469 database query. An empty address is now tested by patterns that are
4470 lookups. However, all the other forms of pattern expect the subject to
4471 contain a local part and a domain, and therefore, for them, an empty
4472 address still always fails if the pattern is not itself empty.
4474 30. Exim went into a mad DNS loop when attempting to do a callout where the
4475 host was specified on an smtp transport, and looking it up yielded more
4476 than one IP address.
4478 31. Re-factored the code for checking spool and log partition space into a
4479 function that finds that data and another that does the check. The former
4480 is then used to implement four new variables: $spool_space, $log_space,
4481 $spool_inodes, and $log_inodes.
4483 32. The RFC2047 encoding function was originally intended for short strings
4484 such as real names; it was not keeping to the 75-character limit for
4485 encoded words that the RFC imposes. It now respects the limit, and
4486 generates multiple encoded words if necessary. To be on the safe side, I
4487 have increased the buffer size for the ${rfc2047: expansion operator from
4490 33. It is now permitted to omit both strings after an "if" condition; if the
4491 condition is true, the result is "true". As before, when the second string
4492 is omitted, a false condition yields an empty string. This makes it less
4493 cumbersome to write custom ACL and router conditions.
4495 34. Failure to deliver a bounce message always caused it to be frozen, even if
4496 there was an errors_to setting on the router. The errors_to setting is now
4499 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4500 canonical form (fully expanded) before being placed in
4501 $sender_host_address.
4503 36. The table in the code that translates DNS record types into text (T_A to
4504 "A" for instance) was missing entries for NS and CNAME. It is just possible
4505 that this could have caused confusion if both these types were looked up
4506 for the same domain, because the text type is used as part of Exim's
4507 per-process caching. But the chance of anyone hitting this buglet seems
4510 37. The dnsdb lookup has been extended in a number of ways.
4512 (1) There is a new type, "zns", which walks up the domain tree until it
4513 finds some nameserver records. It should be used with care.
4515 (2) There is a new type, "mxh", which is like "mx" except that it returns
4516 just the host names, not the priorities.
4518 (3) It is now possible to give a list of domains (or IP addresses) to be
4519 looked up. The behaviour when one of the lookups defers can be
4520 controlled by a keyword.
4522 (4) It is now possible to specify the separator character for use when
4523 multiple records are returned.
4525 38. The dnslists ACL condition has been extended: it is now possible to supply
4526 a list of IP addresses and/or domains to be looked up in a particular DNS
4529 39. Added log_selector=+queue_time_overall.
4531 40. When running the queue in the test harness, wait just a tad after forking a
4532 delivery process, to get repeatability of debugging output.
4534 41. Include certificate and key file names in error message when GnuTLS fails
4535 to set them up, because the GnuTLS error message doesn't include the name
4536 of the failing file when there is a problem reading it.
4538 42. Allow both -bf and -bF in the same test run.
4540 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4542 44. The "Exiscan patch" is now merged into the mainline Exim source.
4544 45. Sometimes the final signoff response after QUIT could fail to get
4545 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4546 < 0 before doing a fflush(). This bug looks as though it goes back to the
4547 introduction of TLS in release 3.20, but "sometimes" must have been rare
4548 because the tests only now provoked it.
4550 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4551 (this can affect the format of dates).
4553 47. exim_tidydb, when checking for the continued existence of a message for
4554 which it has found a message-specific retry record, was not finding
4555 messages that were in split spool directories. Consequently, it was
4556 deleting retry records that should have stayed in existence.
4558 48. Steve fixed some bugs in eximstats.
4560 49. The SPA authentication driver was not abandoning authentication and moving
4561 on to the next authenticator when an expansion was forced to fail,
4562 contradicting the general specification for all authenticators. Instead it
4563 was generating a temporary error. It now behaves as specified.
4565 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4566 (the order specifies the preference for clients). The order is now AES256,
4567 AES128, 3DES, ARCFOUR128.
4569 51. Small patch to Sieve code - explicitly set From: when generating an
4572 52. Exim crashed if a remote delivery caused a very long error message to be
4573 recorded - for instance if somebody sent an entire SpamAssassin report back
4574 as a large number of 550 error lines. This bug was coincidentally fixed by
4575 increasing the size of one of Exim's internal buffers (big_buffer) that
4576 happened as part of the Exiscan merge. However, to be on the safe side, I
4577 have made the code more robust (and fixed the comments that describe what
4580 53. Now that there can be additional text after "Completed" in log lines (if
4581 the queue_time_overall log selector is set), a one-byte patch to exigrep
4582 was needed to allow it to recognize "Completed" as not the last thing in
4585 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4586 patch that reportedly fixes this has been added. I am not expert enough to
4587 create a test for it. This is what the patch creator wrote:
4589 "I found a little strange behaviour of ldap code when working with
4590 Windows 2003 AD Domain, where users was placed in more than one
4591 Organization Units. When I tried to give exim partial DN, the exit code
4592 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4593 But simultaneously result of request was absolutely normal ldap result,
4594 so I produce this patch..."
4596 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4597 REFERENCE, so I have modified the code to exclude the patch when that macro
4600 55. Some experimental protocols are using DNS PTR records for new purposes. The
4601 keys for these records are domain names, not reversed IP addresses. The
4602 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4603 leaves it alone. Component reversal etc. now happens only for IP addresses.
4606 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4608 57. Double the size of the debug message buffer (to 2048) so that more of very
4609 long debug lines gets shown.
4611 58. The exicyclog utility now does better if the number of log files to keep
4612 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4614 59. Two changes related to the smtp_active_hostname option:
4616 (1) $smtp_active_hostname is now available as a variable.
4617 (2) The default for smtp_banner uses $smtp_active_hostname instead
4618 of $primary_hostname.
4620 60. The host_aton() function is supposed to be passed a string that is known
4621 to be a valid IP address. However, in the case of IPv6 addresses, it was
4622 not checking this. This is a hostage to fortune. Exim now panics and dies
4623 if the condition is not met. A case was found where this could be provoked
4624 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4625 components; fortuitously, this particular loophole had already been fixed
4626 by change 4.50/55 above.
4628 If there are any other similar loopholes, the new check in host_aton()
4629 itself should stop them being exploited. The report I received stated that
4630 data on the command line could provoke the exploit when Exim was running as
4631 exim, but did not say which command line option was involved. All I could
4632 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4633 running as the user.
4636 61. There was a buffer overflow vulnerability in the SPA authentication code
4637 (which came originally from the Samba project). I have added a test to the
4638 spa_base64_to_bits() function which I hope fixes it.
4641 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4642 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4644 63. The daemon start-up calls getloadavg() while still root for those OS that
4645 need the first call to be done as root, but it missed one case: when
4646 deliver_queue_load_max is set with deliver_drop_privilege. This is
4647 necessary for the benefit of the queue runner, because there is no re-exec
4648 when deliver_drop_privilege is set.
4650 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4651 This has been fixed.
4653 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4654 was in use, was not putting the data itself into the right store pool;
4655 consequently, it could be overwritten for a subsequent message in the same
4656 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4659 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4661 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4662 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4663 the function were treating the return as a boolean value, which happened to
4664 work because 0=false and not-0=true, but is not correct code.
4666 68. The host_aton() function was not handling scoped IPv6 addresses (those
4667 with, for example, "%eth0" on the end) correctly.
4669 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4670 negated items (that is, ~something) in unsigned ints. Some compilers
4671 apparently mutter when there is no cast.
4673 70. If an address verification called from an ACL failed, and did not produce a
4674 user-specific message (i.e. there was only a "system" message), nothing was
4675 put in $acl_verify_message. In this situation, it now puts the system
4678 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4679 session; change 4.31/43 added the unwanted input to the log line - except
4680 that it did not do this in the start of session case. It now does.
4682 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4683 This is wrong and can cause the other end to generate a synchronization
4684 error if it is another Exim or anything else that does the synchronization
4685 check. A QUIT command is no longer sent after a timeout.
4687 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4688 during host lookups.
4690 74. The defer_ok option of callout verification was not working if it was used
4691 when verifying addresses in header lines, that is, for this case:
4693 verify = header_sender/callout=defer_ok
4695 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4696 those file descriptors could be used for SMTP connections. If anything
4697 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4698 could be sent to the SMTP client, causing chaos. The daemon now opens
4699 stdin, stdout, and stderr to /dev/null when it puts itself into the
4702 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4703 log by default. The user can override this with suitable Perl magic.
4705 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4706 the log message when discard triggers, was not working for the DATA ACL or
4707 for the non-SMTP ACL.
4709 78. Error message wording change in sieve.c.
4711 79. If smtp_accept_max_per_host was set, the number of connections could be
4712 restricted to fewer than expected, because the daemon was trying to set up
4713 a new connection before checking whether the processes handling previous
4714 connections had finished. The check for completed processes is now done
4715 earlier. On busy systems, this bug wouldn't be noticed because something
4716 else would have woken the daemon, and it would have reaped the completed
4719 80. If a message was submitted locally by a user whose login name contained one
4720 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4721 It caused a spool format error. I have fixed the spool reading code. A
4722 related problem was that the "from" clause in the Received: line became
4723 illegal because of the space(s). It is now covered by ${quote_local_part.
4725 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4728 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4730 83. Patch from Sieve maintainer to fix unterminated string problem in
4731 "vacation" handling.
4733 84. Some minor changes to the Linux configuration files to help with other
4734 OS variants using glibc.
4736 85. One more patch for Sieve to update vacation handling to latest spec.
4739 ----------------------------------------------------
4740 See the note above about the 4.44 and 4.50 releases.
4741 ----------------------------------------------------
4747 1. Change 4.43/35 introduced a bug that caused file counts to be
4748 incorrectly computed when quota_filecount was set in an appendfile
4751 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4752 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4755 3. Give more explanation in the error message when the command for a transport
4756 filter fails to execute.
4758 4. There are several places where Exim runs a non-Exim command in a
4759 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4760 was being done only for the command run by the queryprogram router. It is
4761 now done for all such subprocesses. The other cases are: ${run, transport
4762 filters, and the commands run by the lmtp and pipe transports.
4764 5. Some older OS have a limit of 256 on the maximum number of file
4765 descriptors. Exim was using setrlimit() to set 1000 as a large value
4766 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4767 systems. I've change it so that if it can't get 1000, it tries for 256.
4769 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4770 was an oversight, and furthermore, ever since the addition of extra
4771 controls (e.g. 4.43/32), the checks on when to allow different forms of
4772 "control" were broken. There should now be diagnostics for all cases when a
4773 control that does not make sense is encountered.
4775 7. $recipients is now available in the predata ACL (oversight).
4777 8. Tidy the search cache before the fork to do a delivery from a message
4778 received from the command line. Otherwise the child will trigger a lookup
4779 failure and thereby defer the delivery if it tries to use (for example) a
4780 cached ldap connection that the parent has called unbind on.
4782 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4783 of $address_data from the recipient verification was clobbered by the
4784 sender verification.
4786 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4787 was its contents. (It was OK if the option was not defined at all.)
4789 11. A "Completed" log line is now written for messages that are removed from
4790 the spool by the -Mrm option.
4792 12. $host_address is now set to the target address during the checking of
4793 ignore_target_hosts.
4795 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4796 being passed; this would have caused $sender_host_name to have been used if
4797 matching the list had actually called for a host name (not very likely,
4798 since this list is usually IP addresses). A host name is now passed as
4801 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4802 code that specifies a non-restarting handler (typically sigaction() in
4803 modern systems) in an attempt to fix a rare and obscure crash bug.
4805 15. Narrowed the window for a race in the daemon that could cause it to ignore
4806 SIGCHLD signals. This is not a major problem, because they are used only to
4807 wake it up if nothing else does.
4809 16. A malformed maildirsize file could cause Exim to calculate negative values
4810 for the mailbox size or file count. Odd effects could occur as a result.
4811 The maildirsize information is now recalculated if the size or filecount
4814 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4815 support for a long time. Removed HAVE_SYS_VFS_H.
4817 18. Updated exipick to current release from John Jetmore.
4819 19. Allow an empty sender to be matched against a lookup in an address list.
4820 Previously the only cases considered were a regular expression, or an
4823 20. Exim went into a mad DNS lookup loop when doing a callout where the
4824 host was specified on the transport, if the DNS lookup yielded more than
4827 21. The RFC2047 encoding function was originally intended for short strings
4828 such as real names; it was not keeping to the 75-character limit for
4829 encoded words that the RFC imposes. It now respects the limit, and
4830 generates multiple encoded words if necessary. To be on the safe side, I
4831 have increased the buffer size for the ${rfc2047: expansion operator from
4834 22. Failure to deliver a bounce message always caused it to be frozen, even if
4835 there was an errors_to setting on the router. The errors_to setting is now
4838 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4839 canonical form (fully expanded) before being placed in
4840 $sender_host_address.
4842 24. Updated eximstats to version 1.33
4844 25. Include certificate and key file names in error message when GnuTLS fails
4845 to set them up, because the GnuTLS error message doesn't include the name
4846 of the failing file when there is a problem reading it.
4848 26. Expand error message when OpenSSL has problems setting up cert/key files.
4851 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4852 (this can affect the format of dates).
4854 28. exim_tidydb, when checking for the continued existence of a message for
4855 which it has found a message-specific retry record, was not finding
4856 messages that were in split spool directories. Consequently, it was
4857 deleting retry records that should have stayed in existence.
4859 29. eximstats updated to version 1.35
4860 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4861 1.35 - bugfix such that pie charts by volume are generated correctly
4863 30. The SPA authentication driver was not abandoning authentication and moving
4864 on to the next authenticator when an expansion was forced to fail,
4865 contradicting the general specification for all authenticators. Instead it
4866 was generating a temporary error. It now behaves as specified.
4868 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4869 (the order specifies the preference for clients). The order is now AES256,
4870 AES128, 3DES, ARCFOUR128.
4872 31. Small patch to Sieve code - explicitly set From: when generating an
4875 32. Exim crashed if a remote delivery caused a very long error message to be
4876 recorded - for instance if somebody sent an entire SpamAssassin report back
4877 as a large number of 550 error lines. This bug was coincidentally fixed by
4878 increasing the size of one of Exim's internal buffers (big_buffer) that
4879 happened as part of the Exiscan merge. However, to be on the safe side, I
4880 have made the code more robust (and fixed the comments that describe what
4883 33. Some experimental protocols are using DNS PTR records for new purposes. The
4884 keys for these records are domain names, not reversed IP addresses. The
4885 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4886 leaves it alone. Component reversal etc. now happens only for IP addresses.
4889 34. The host_aton() function is supposed to be passed a string that is known
4890 to be a valid IP address. However, in the case of IPv6 addresses, it was
4891 not checking this. This is a hostage to fortune. Exim now panics and dies
4892 if the condition is not met. A case was found where this could be provoked
4893 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4894 components; fortuitously, this particular loophole had already been fixed
4895 by change 4.50/55 or 4.44/33 above.
4897 If there are any other similar loopholes, the new check in host_aton()
4898 itself should stop them being exploited. The report I received stated that
4899 data on the command line could provoke the exploit when Exim was running as
4900 exim, but did not say which command line option was involved. All I could
4901 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4902 running as the user.
4905 35. There was a buffer overflow vulnerability in the SPA authentication code
4906 (which came originally from the Samba project). I have added a test to the
4907 spa_base64_to_bits() function which I hope fixes it.
4910 36. The daemon start-up calls getloadavg() while still root for those OS that
4911 need the first call to be done as root, but it missed one case: when
4912 deliver_queue_load_max is set with deliver_drop_privilege. This is
4913 necessary for the benefit of the queue runner, because there is no re-exec
4914 when deliver_drop_privilege is set.
4916 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4917 was in use, was not putting the data itself into the right store pool;
4918 consequently, it could be overwritten for a subsequent message in the same
4919 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4922 38. Sometimes the final signoff response after QUIT could fail to get
4923 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4924 < 0 before doing a fflush(). This bug looks as though it goes back to the
4925 introduction of TLS in release 3.20, but "sometimes" must have been rare
4926 because the tests only now provoked it.
4932 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4933 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4934 Now it returns an integer. A number of calls were still expecting a T/F
4935 return. Fortuitously, in all cases, the tests worked in OK situations,
4936 which is the norm. However, things would have gone wrong on any write
4937 failures on the smtp file descriptor. This function is used when sending
4938 messages over SMTP and also when doing verify callouts.
4940 2. When Exim is called to do synchronous delivery of a locally submitted
4941 message (the -odf or -odi options), it no longer closes stderr before doing
4944 3. Implemented the mua_wrapper option.
4946 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4948 5. Implemented the functions header_remove(), header_testname(),
4949 header_add_at_position(), and receive_remove_recipient(), and exported them
4952 6. If an ACL "warn" statement specified the addition of headers, Exim already
4953 inserted X-ACL-Warn: at the start if there was no header name. However, it
4954 was not making this test for the second and subsequent header lines if
4955 there were newlines in the string. This meant that an invalid header could
4956 be inserted if Exim was badly configured.
4958 7. Allow an ACL "warn" statement to add header lines at the start or after all
4959 the Received: headers, as well as at the end.
4961 8. Added the rcpt_4xx retry error code.
4963 9. Added postmaster_mailfrom=xxx to callout verification option.
4965 10. Added mailfrom=xxxx to the callout verification option, for verify=
4968 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4969 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4971 12. Inserted some casts to stop certain compilers warning when using pointer
4972 differences as field lengths or precisions in printf-type calls (mostly
4973 affecting debugging statements).
4975 13. Added optional readline() support for -be (dynamically loaded).
4977 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4978 same clock tick as a message's arrival, so that its received time was the
4979 same as the "first fail" time on the retry record, and that message
4980 remained on the queue past the ultimate address timeout, every queue runner
4981 would try a delivery (because it was past the ultimate address timeout) but
4982 after another failure, the ultimate address timeout, which should have then
4983 bounced the address, did not kick in. This was a "< instead of <=" error;
4984 in most cases the first failure would have been in the next clock tick
4985 after the received time, and all would be well.
4987 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4988 being recognized when the domain list was tested by the match_domain
4989 condition in an expansion string.
4991 16. Added the ${str2b64: operator.
4993 17. Exim was always calling setrlimit() to set a large limit for the number of
4994 processes, without checking whether the existing limit was already
4995 adequate. (It did check for the limit on file descriptors.) Furthermore,
4996 errors from getrlimit() and setrlimit() were being ignored. Now they are
4997 logged to the main and panic logs, but Exim does carry on, to try to do its
4998 job under whatever limits there are.
5000 18. Imported PCRE 5.0.
5002 19. Trivial typo in log message " temporarily refused connection" (the leading
5005 20. If the log selector return_path_on_delivery was set and an address was
5006 redirected to /dev/null, the delivery process crashed because it assumed
5007 that a return path would always be set for a "successful" delivery. In this
5008 case, the whole delivery is bypassed as an optimization, and therefore no
5011 21. Internal re-arrangement: the function for sending a challenge and reading
5012 a response while authentication was assuming a zero-terminated challenge
5013 string. It's now changed to take a pointer and a length, to allow for
5014 binary data in such strings.
5016 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5018 23. Exim was not respecting finduser_retries when seeking the login of the
5019 uid under which it was called; it was always trying 10 times. (The default
5020 setting of finduser_retries is zero.) Also, it was sleeping after the final
5021 failure, which is pointless.
5023 24. Implemented tls_on_connect_ports.
5025 25. Implemented acl_smtp_predata.
5027 26. If the domain in control=submission is set empty, Exim assumes that the
5028 authenticated id is a complete email address when it generates From: or
5029 Sender: header lines.
5031 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5032 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5033 chown and chgrp in /bin and hostname in /usr/bin.
5035 28. Exim was keeping the "process log" file open after each use, just as it
5036 does for the main log. This opens the possibility of it remaining open for
5037 long periods when the USR1 signal hits a daemon. Occasional processlog
5038 errors were reported, that could have been caused by this. Anyway, it seems
5039 much more sensible not to leave this file open at all, so that is what now
5042 29. The long-running daemon process does not normally write to the log once it
5043 has entered its main loop, and it closes the log before doing so. This is
5044 so that log files can straightforwardly be renamed and moved. However,
5045 there are a couple of unusual error situations where the daemon does write
5046 log entries, and I had neglected to close the log afterwards.
5048 30. The text of an SMTP error response that was received during a remote
5049 delivery was being truncated at 512 bytes. This is too short for some of
5050 the long messages that one sometimes sees. I've increased the limit to
5053 31. It is now possible to make retry rules that apply only when a message has a
5054 specific sender, in particular, an empty sender.
5056 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5057 it possible to be selective about when SMTP synchronization is enforced.
5059 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5061 32. Implemented hosts_connection_nolog.
5063 33. Added an ACL for QUIT.
5065 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5068 35. Added mailbox_size and mailbox_filecount to appendfile.
5070 36. Added control = no_multiline_responses to ACLs.
5072 37. There was a bug in the logic of the code that waits for the clock to tick
5073 in the case where the clock went backwards by a substantial amount such
5074 that the microsecond fraction of "now" was more than the microsecond
5075 fraction of "then" (but the whole seconds number was less).
5077 38. Added support for the libradius Radius client library this is found on
5078 FreeBSD (previously only the radiusclient library was supported).
5084 1. When certain lookups returned multiple values in the form name=value, the
5085 quoting of the values was not always being done properly. Specifically:
5086 (a) If the value started with a double quote, but contained no whitespace,
5088 (b) If the value contained whitespace other than a space character (i.e.
5089 tabs or newlines or carriage returns) it was not quoted.
5090 This fix has been applied to the mysql and pgsql lookups by writing a
5091 separate quoting function and calling it from the lookup code. The fix
5092 should probably also be applied to nisplus, ibase and oracle lookups, but
5093 since I cannot test any of those, I have not disturbed their existing code.
5095 2. A hit in the callout cache for a specific address caused a log line with no
5096 reason for rejecting RCPT. Now it says "Previous (cached) callout
5097 verification failure".
5099 3. There was an off-by-one bug in the queryprogram router. An over-long
5100 return line was truncated at 256 instead of 255 characters, thereby
5101 overflowing its buffer with the terminating zero. As well as fixing this, I
5102 have increased the buffer size to 1024 (and made a note to document this).
5104 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5105 when Exim is waiting for an SMTP response from a remote server, Exim
5106 restarts its select() call on the socket, thereby resetting its timeout.
5107 This is not a problem when such interrupts are rare. Somebody set up a cron
5108 job to run exiwhat every 2 minutes, which is less than the normal select()
5109 timeout (5 or 10 minutes). This meant that the select() timeout never
5110 kicked in because it was always reset. I have fixed this by comparing the
5111 time when an interrupt arrives with the time at the start of the first call
5112 to select(). If more time than the timeout has elapsed, the interrupt is
5113 treated as a timeout.
5115 5. Some internal re-factoring in preparation for the addition of Sieve
5116 extensions (by MH). In particular, the "personal" test is moved to a
5117 separate function, and given an option for scanning Cc: and Bcc: (which is
5118 not set for Exim filters).
5120 6. When Exim created an email address using the login of the caller as the
5121 local part (e.g. when creating a From: or Sender: header line), it was not
5122 quoting the local part when it contained special characters such as @.
5124 7. Installed new OpenBSD configuration files.
5126 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5127 try to make them clearer.
5129 9. Callout options, other than the timeout value, were being ignored when
5130 verifying sender addresses in header lines. For example, when using
5132 verify = header_sender/callout=no_cache
5134 the cache was (incorrectly) being used.
5136 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5137 only the Cygwin environment.
5139 11. When return_path_on_delivery was set as a log selector, if different remote
5140 addresses in the same message used different return paths and parallel
5141 remote delivery occurred, the wrong values would sometimes be logged.
5142 (Whenever a remote delivery process finished, the return path value from
5143 the most recently started remote delivery process was logged.)
5145 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5146 lines when AUTH and/or TLS are in use. This is the "received protocol"
5147 field. Exim used to use "asmtp" for authenticated SMTP, without any
5148 indication (in the protocol name) for TLS use. Now it follows the RFC and
5149 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5150 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5151 names appear in log lines as well as in Received: header lines.
5153 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5154 extensions, and comparison tests, and to fix some bugs.
5156 14. Changes to the "personal" filter test:
5158 (1) The test was buggy in that it was just doing the equivalent of
5159 "contains" tests on header lines. For example, if a user's address was
5160 anne@some.where, the "personal" test would incorrectly be true for
5162 To: susanne@some.where
5164 This test is now done by extracting each address from the header in turn,
5165 and checking the entire address. Other tests that are part of "personal"
5166 are now done using regular expressions (for example, to check local parts
5167 of addresses in From: header lines).
5169 (2) The list of non-personal local parts in From: addresses has been
5170 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5171 taken from the Sieve specification recommendations.
5173 (3) If the message contains any header line starting with "List-" it is
5174 treated as non-personal.
5176 (4) The test for "circular" in the Subject: header line has been removed
5177 because it now seems ill-conceived.
5179 15. Minor typos in src/EDITME comments corrected.
5181 16. Installed latest exipick from John Jetmore.
5183 17. If headers_add on a router specified a text string that was too long for
5184 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5185 of string_sprintf() is now avoided.
5187 18. $message_body_size was not set (it was always zero) when running the DATA
5188 ACL and the local_scan() function.
5190 19. For the "mail" command in an Exim filter, no default was being set for
5191 the once_repeat time, causing a random time value to be used if "once" was
5192 specified. (If the value happened to be <= 0, no repeat happened.) The
5193 default is now 0s, meaning "never repeat". The "vacation" command was OK
5194 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5195 (I found it when inspecting the code).
5197 20. There is now an overall timeout for performing a callout verification. It
5198 defaults to 4 times the callout timeout, which applies to individual SMTP
5199 commands during the callout. The overall timeout applies when there is more
5200 than one host that can be tried. The timeout is checked before trying the
5201 next host. This prevents very long delays if there are a large number of
5202 hosts and all are timing out (e.g. when the network connections are timing
5203 out). The value of the overall timeout can be changed by specifying an
5204 additional sub-option for "callout", called "maxwait". For example:
5206 verify = sender/callout=5s,maxwait=20s
5208 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5209 to the end before writing, but this should make it even safer).
5211 22. Exim was forgetting that it had advertised PIPELINING for the second and
5212 subsequent messages on an SMTP connection. It was also not resetting its
5213 memory on STARTTLS and an internal HELO.
5215 23. When Exim logs an SMTP synchronization error within a session, it now
5216 records whether PIPELINING has been advertised or not.
5218 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5219 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5220 rather than long int.
5222 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5224 26. Added the never_mail option to autoreply.
5230 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5231 crash if the getsockname() call failed; this can happen if a connection is
5232 closed very soon after it is established. The problem was simply in the
5233 order in which certain operations were done, causing Exim to try to write
5234 to the SMTP stream before it had set up the file descriptor. The bug has
5235 been fixed by making things happen in the correct order.
5241 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5242 before the connection was closed, thus losing the rejection response.
5244 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5245 some early Solaris releases, but causes trouble in current releases where
5246 socklen_t is defined.
5248 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5251 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5254 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5255 that releasing the top of it at the end releases what was used for sub-
5256 expansions (unless the block got too big). However, discard this block if
5257 the first thing is a variable or header, so that we can use its block when
5258 it is dynamic (useful for very large $message_headers, for example).
5260 6. Lookups now cache *every* query, not just the most recent. A new, separate
5261 store pool is used for this. It can be recovered when all lookup caches are
5262 flushed. Lookups now release memory at the end of their result strings.
5263 This has involved some general refactoring of the lookup sources.
5265 7. Some code has been added to the store_xxx() functions to reduce the amount
5266 of flapping under certain conditions.
5268 8. log_incoming_interface used to affect only the <= reception log lines. Now
5269 it causes the local interface and port to be added to several more SMTP log
5270 lines, for example "SMTP connection from", and rejection lines.
5272 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5274 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5276 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5277 could be overwritten at the end of the current message (or the start of a
5278 new message if it was set in a HELO ACL). The value is now preserved for
5279 the duration of the SMTP connection.
5281 12. If a transport had a headers_rewrite setting, and a matching header line
5282 contained an unqualified address, that address was qualified, even if it
5283 did not match any rewriting rules. The underlying bug was that the values
5284 of the flags that permit the existence of unqualified sender and recipient
5285 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5286 non-local messages, and by -bnq for local messages) were not being
5287 preserved with the message after it was received.
5289 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5290 "next input=" as part of the text comprising the host identity instead of
5291 the correct text. The code was using the same buffer for two different
5292 strings. However, depending on which order the printing function evaluated
5293 its arguments, the bug did not always show up. Under Linux, for example, my
5294 test suite worked just fine.
5296 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5297 doesn't work with some older versions of Perl. It has been changed to "my",
5298 which in any case is probably the better facility to use.
5300 15. A really picky compiler found some instances of statements for creating
5301 error messages that either had too many or two few arguments for the format
5304 16. The size of the buffer for calls to the DNS resolver has been increased
5305 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5306 for addresses that have a lot of PTR records. This alleviates a problem; it
5307 does not fully solve it.
5309 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5310 buffer now truncates the list and logs the incident, which is the same
5311 action as happens when Exim is looking up a host name and its aliases.
5312 Previously in this situation something unpredictable would happen;
5313 sometimes it was "internal error: store_reset failed".
5315 18. If a server dropped the connection unexpectedly when an Exim client was
5316 using GnuTLS and trying to read a response, the client delivery process
5317 crashed while trying to generate an error log message.
5319 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5320 string, for example:
5322 warn message = H1: something\nH2: something
5324 the text was added as a single header line from Exim's point of view
5325 though it ended up OK in the delivered message. However, searching for the
5326 second and subsequent header lines using $h_h2: did not work. This has been
5327 fixed. Similarly, if a system filter added multiple headers in this way,
5328 the routers could not see them.
5330 20. Expanded the error message when iplsearch is called with an invalid key to
5331 suggest using net-iplsearch in a host list.
5333 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5334 ACLs are no longer actually imposed (and a message to that effect is
5337 22. If a "gecos" field in a passwd entry contained escaped characters, in
5338 particular, if it contained a \" sequence, Exim got it wrong when building
5339 a From: or a Sender: header from that name. A second bug also caused
5340 incorrect handling when an unquoted " was present following a character
5341 that needed quoting.
5343 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5344 was not being matched caselessly.
5346 24. Arranged for all hyphens in the exim.8 source to be escaped with
5349 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5350 properly. Recipient callout cache records were still being keyed to include
5351 the sender, even when use_sender was set false. This led to far more
5352 callouts that were necessary. The sender is no longer included in the key
5353 when use_sender is false.
5355 26. Added "control = submission" modifier to ACLs.
5357 27. Added the ${base62d: operator to decode base 62 numbers.
5359 28. dnsdb lookups can now access SRV records.
5361 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5362 the configuration file.
5364 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5365 (-v) mode. This makes the output for a verbose queue run more intelligible.
5367 31. Added a use_postmaster feature to recipient callouts.
5369 32. Added the $body_zerocount variable, containing the number of binary zero
5370 bytes in the message body.
5372 33. The time of last modification of the "new" subdirectory is now used as the
5373 "mailbox time last read" when there is a quota error for a maildir
5376 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5378 35. Added +ignore_unknown as a special item in host lists.
5380 36. Code for decoding IPv6 addresses in host lists is now included, even if
5381 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5382 address was recognized as an IP address, but was then not correctly decoded
5383 into binary, causing unexpected and incorrect effects when compared with
5390 1. Very minor rewording of debugging text in manualroute to say "list of
5391 hosts" instead of "hostlist".
5393 2. If verify=header_syntax was set, and a header line with an unqualified
5394 address (no domain) and a large number of spaces between the end of the
5395 name and the colon was received, the reception process suffered a buffer
5396 overflow, and (when I tested it) crashed. This was caused by some obsolete
5397 code that should have been removed. The fix is to remove it!
5399 3. When running in the test harness, delay a bit after writing a bounce
5400 message to get a bit more predictability in the log output.
5402 4. Added a call to search_tidyup() just before forking a reception process. In
5403 theory, someone could use a lookup in the expansion of smtp_accept_max_
5404 per_host which, without the tidyup, could leave open a database connection.
5406 5. Added the variables $recipient_data and $sender_data which get set from a
5407 lookup success in an ACL "recipients" or "senders" condition, or a router
5408 "senders" option, similar to $domain_data and $local_part_data.
5410 6. Moved the writing of debug_print from before to after the "senders" test
5413 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5414 problems for message scanning, either using a data ACL, or using
5415 local_scan() because the Received: header was not generated till after they
5416 were called (in order to set the time as the time of reception completion).
5417 I have revised the way this works. The header is now generated after the
5418 body is received, but before the ACL or local_scan() are called. After they
5419 are run, the timestamp in the header is updated.
5425 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5426 before starting a queue runner without re-exec. This happened only when
5427 deliver_drop_privilege was set or when the Exim user was set to root. The
5428 effect of the bug was that timeouts during subsequent deliveries caused
5429 crashes instead of being properly handled. The handler is now left at its
5430 default (and expected) setting.
5432 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5433 message, again when deliver_drop_privilege is set or Exim is run as root.
5434 The bug described in (1) was not present in this case, but the tidying up
5435 of the other signals was missing. I have made the two cases consistent.
5437 3. The ignore_target_hosts setting on a manualroute router was being ignored
5438 for hosts that were looked up using the /MX notation.
5440 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5443 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5444 operated on the sender address. After changing the $sender_address to <>
5445 for the sender address verify, Exim was re-instated it as the original
5446 (before rewriting) address, but remembering that it had rewritten it, so it
5447 wasn't rewriting it again. This bug also had the effect of breaking the
5448 sender address verification caching when the sender address was rewritten.
5450 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5451 This has been changed so that if the ip literal address matches
5452 ignore_target_hosts, the router declines.
5454 7. Added expansion conditions match_domain, match_address, and match_local_
5455 part (NOT match_host).
5457 8. The placeholder for the Received: header didn't have a length field set.
5459 9. Added code to Exim itself and to exim_lock to test for a specific race
5460 condition that could lead to file corruption when using MBX delivery. The
5461 issue is with the lockfile that is created in /tmp. If this file is removed
5462 after a process has opened it but before that process has acquired a lock,
5463 there is the potential for a second process to recreate the file and also
5464 acquire a lock. This could lead to two Exim processes writing to the file
5465 at the same time. The added code performs the same test as UW imapd; it
5466 checks after acquiring the lock that its file descriptor still refers to
5467 the same named file.
5469 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5470 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5473 11. Added the smtp_active_hostname option. If used, this will typically be made
5474 to depend on the incoming interface address. Because $interface_address is
5475 not set up until the daemon has forked a reception process, error responses
5476 that can happen earlier (such as "too many connections") no longer contain
5479 12. If an expansion in a condition on a "warn" statement fails because a lookup
5480 defers, the "warn" statement is abandoned, and the next ACL statement is
5481 processed. Previously this caused the whole ACL to be aborted.
5483 13. Added the iplsearch lookup type.
5485 14. Added ident_timeout as a log selector.
5487 15. Added tls_certificate_verified as a log selector.
5489 16. Added a global option tls_require_ciphers (compare the smtp transport
5490 option of the same name). This controls incoming TLS connections.
5492 17. I finally figured out how to make tls_require_ciphers do a similar thing
5493 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5494 before starting the TLS session.
5496 18. Tabs are now shown as \t in -bP output.
5498 19. If the log selector return_path_on_delivery was set, Exim crashed when
5499 bouncing a message because it had too many Received: header lines.
5501 20. If two routers both had headers_remove settings, and the first one included
5502 a superfluous trailing colon, the final name in the first list and the
5503 first name in the second list were incorrectly joined into one item (with a
5504 colon in the middle).
5510 1. Added -C and -D options to the exinext utility, mainly to make it easier
5511 to include in the automated testing, but these could be helpful when
5512 multiple configurations are in use.
5514 2. The exinext utility was not formatting the output nicely when there was
5515 an alternate port involved in the retry record key, nor when there was a
5516 message id as well (for retries that were specific to a specific message
5517 and a specific host). It was also confused by IPv6 addresses, because of
5518 the additional colons they contain. I have fixed the IPv4 problem, and
5519 patched it up to do a reasonable job for IPv6.
5521 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5522 delivery, the log line now contains "pipelined" if PIPELINING was used.
5524 4. An SMTP transport process used to panic and die if the bind() call to set
5525 an explicit outgoing interface failed. This has been changed; it is now
5526 treated in the same way as a connect() failure.
5528 5. A reference to $sender_host_name in the part of a conditional expansion
5529 that was being skipped was still causing a DNS lookup. This no longer
5532 6. The def: expansion condition was not recognizing references to header lines
5533 that used bh_ and bheader_.
5535 7. Added the _cache feature to named lists.
5537 8. The code for checking quota_filecount in the appendfile transport was
5538 allowing one more file than it should have been.
5540 9. For compatibility with Sendmail, the command line option
5548 and sets the incoming protocol and host name (for trusted callers). The
5549 host name and its colon can be omitted when only the protocol is to be set.
5550 Note the Exim already has two private options, -pd and -ps, that refer to
5551 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5552 "s", but I don't think that's a major issue.
5554 10. A number of refactoring changes to the code, none of which should affect
5557 (a) The number of logging options was getting close to filling up the
5558 32-bit word that was used as a bit map. I have split them into two classes:
5559 those that are passed in the argument to log_write(), and those that are
5560 only ever tested independently outside of that function. These are now in
5561 separate 32-bit words, so there is plenty of room for expansion again.
5562 There is no change in the user interface or the logging behaviour.
5564 (b) When building, for example, log lines, the code previously used a
5565 macro that called string_cat() twice, in order to add two strings. This is
5566 not really sufficiently general. Furthermore, there was one instance where
5567 it was actually wrong because one of the argument was used twice, and in
5568 one call a function was used. (As it happened, calling the function twice
5569 did not affect the overall behaviour.) The macro has been replaced by a
5570 function that can join an arbitrary number of extra strings onto a growing
5573 (c) The code for expansion conditions now uses a table and a binary chop
5574 instead of a serial search (which was left over from when there were very
5575 few conditions). Also, it now recognizes conditions like "pam" even when
5576 the relevant support is not compiled in: a suitably worded error message is
5577 given if an attempt is made to use such a condition.
5579 11. Added ${time_interval:xxxxx}.
5581 12. A bug was causing one of the ddress fields not to be passed back correctly
5582 from remote delivery subprocesses. The field in question was not being
5583 subsequently used, so this caused to problems in practice.
5585 13. Added new log selectors queue_time and deliver_time.
5587 14. Might have fixed a bug in maildirsizefile handling that threw up
5588 "unexpected character" debug warnings, and recalculated the data
5589 unnecessarily. In any case, I expanded the warning message to give more
5592 15. Added the message "Restricted characters in address" to the statements in
5593 the default ACL that block characters like @ and % in local parts.
5595 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5596 Three changes have been made:
5598 (a) There was a serious bug; a negative response to MAIL caused the whole
5599 recipient domain to be cached as invalid, thereby blocking all messages
5600 to all local parts at the same domain, from all senders. This bug has
5601 been fixed. The domain is no longer cached after a negative response to
5602 MAIL if the sender used is not empty.
5604 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5607 (c) A new callout option, "use_sender" has been added for people who want
5608 the modified behaviour.
5614 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5617 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5618 indeed breaks things for older releases.
5620 3. Added additional logging to the case where there is a problem reading data
5621 from a filter that is running in a subprocess using a pipe, in order to
5622 try to track down a specific problem.
5624 4. Testing facility fudge: when running in the test harness and attempting
5625 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5626 getting "No route to host". Convert this to a timeout.
5628 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5631 6. Some OS don't have socklen_t but use size_t instead. This affects the
5632 fifth argument of getsockopt() amongst other things. This is now
5633 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5634 can be set for individual OS. I have set it for SunOS5, OSF1, and
5635 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5636 some earlier ones do not.
5638 7. Change 4.30/15 was not doing the test caselessly.
5640 8. The standard form for an IPv6 address literal was being rejected by address
5641 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5642 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5643 this, as well as the form without the "IPv6" on the front (but only when
5644 address literals are enabled, of course).
5646 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5648 10. Exim crashed if a message with an empty sender address specified by -f
5649 encountered a router with an errors_to setting. This could be provoked only
5650 by a command such as
5654 where an empty string was supplied; "<>" did not hit this bug.
5656 11. Installed PCRE release 4.5.
5658 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5659 remained set. It is now erased.
5661 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5662 times from message ids (which are base 36 rather than the normal 62).
5664 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5665 were being counted as actual protocol errors, and logged if the log
5666 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5667 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5668 and DATA following a set of rejected RCPTs do not count as protocol errors.
5669 In other words, Exim assumes they were pipelined, though this may not
5670 actually be the case. Of course, in all cases the client gets an
5671 appropriate error code.
5673 15. If a lookup fails in an ACL condition, a message about the failure may
5674 be available; it is used if testing the ACL cannot continue, because most
5675 such messages specify what the cause of the deferral is. However, some
5676 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5677 that caused an old message to be retained and used if a later statement
5678 caused a defer, replacing the real cause of the deferral.
5680 16. If an IP address had so many PTR records that the DNS lookup buffer
5681 was not large enough to hold them, Exim could crash while trying to process
5682 the truncated data. It now detects and logs this case.
5684 17. Further to 4.21/58, another change has been made: if (and only if) the
5685 first line of a message (the first header line) ends with CRLF, a bare LF
5686 in a subsequent header line has a space inserted after it, so as not to
5687 terminate the header.
5689 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5690 unnecessarily, used atoi() instead of strtol(), and didn't check the
5691 termination when getting file sizes from file names by regex.
5693 19. Completely re-implemented the support for maildirsize files, in the light
5694 of a number of problems with the previous contributed implementation
5695 (4.30/29). In particular:
5697 . If the quota is zero, the maildirsize file is maintained, but no quota is
5700 . If the maildir directory does not exist, it is created before any attempt
5701 to write a maildirsize file.
5703 . The quota value in the file is just a cache; if the quota is changed in
5704 the transport, the new value overrides.
5706 . A regular expression is available for excluding directories from the
5709 20. The autoreply transport checks the characters in options that define the
5710 message's headers; it allows continued headers, but it was checking with
5711 isspace() after an embedded newline instead of explicitly looking for a
5714 21. If all the "regular" hosts to which an address was routed had passed their
5715 expiry times, and had not reached their retry times, the address was
5716 bounced, even if fallback hosts were defined. Now Exim should go on to try
5719 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5720 equivalent code in the SMTP transport. Some hosts send humungous responses
5721 to HELO/EHLO, more than 1024 it seems.
5723 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5724 gives compiler warnings in some environments. I've now done it "properly",
5727 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5728 (because of problems with the built-in one) was declared to return uschar *
5729 instead of char *, causing compiler failure.
5731 25. Fixed a file descriptor leak when processing alias/forward files.
5733 26. Fixed a minor format string issue in dbfn.c.
5735 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5737 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5738 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5741 29. When a sender address is verified, it is cached, to save repeating the test
5742 when there is more than one recipient in a message. However, when the
5743 verification involves a callout, it is possible for different callout
5744 options to be set for different recipients. It is too complicated to keep
5745 track of this in the cache, so now Exim always runs a verification when a
5746 callout is required, relying on the callout cache for the optimization.
5747 The overhead is duplication of the address routing, but this should not be
5750 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5751 to be verified with callout=postmaster, and the main callout worked but the
5752 postmaster check failed, the verification correctly failed. However, if a
5753 subsequent RCPT command asked for sender verification *without* the
5754 postmaster check, incorrect caching caused this verification also to fail,
5757 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5758 it was not caching the DNS options (qualify_single, search_parents) that
5759 were used when the lookup failed. A subsequent lookup with different
5760 options therefore always gave the same answer, though there were cases
5761 where it should not have. (Example: a "domains = !$mx_any" option on a
5762 dnslookup router: the "domains" option is always processed without any
5763 widening, but the router might have qualify_single set.) Now Exim uses the
5764 cached value only when the same options are set.
5766 32. Added John Jetmore's "exipick" utility to the distribution.
5768 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5769 than a timeout (e.g. a certificate is required, and is not provided), an
5770 Exim server now closes the connection immediately. Previously it waited for
5771 the client to close - but if the client is SSL, it seems that they each
5772 wait for each other, leading to a delay before one of them times out.
5774 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5775 maintained 0.8.x compatibility because I don't think many are using it, and
5776 it is clearly obsolete.
5778 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5781 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5782 client certificate was expired. A simple patch fixes this, though I don't
5783 understand the full logic of why the verify callback is called multiple
5786 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5787 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5788 which causes problems with some clients (such as the Certicom SSL Plus
5789 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5790 disables the coutermeasure allowing Eudora to connect."
5792 38. Exim was not checking that a write() to a log file succeeded. This could
5793 lead to Bad Things if a log got too big, in particular if it hit a file
5794 size limit. Exim now panics and dies if it cannot write to a log file, just
5795 as it does if it cannot open a log file.
5797 39. Modified OS/Makefile-Linux so that it now contains
5799 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5801 The two -D definitions ensure that Exim is compiled with large file
5802 support, which makes it possible to handle log files that are bigger than
5805 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5806 instance) a domain was checked against a named list that involved a lookup,
5807 causing $domain_data to be set, then another domain was checked against the
5808 same list, then the first domain was re-checked, the value of $domain_data
5809 after the final check could be wrong. In particular, if the second check
5810 failed, it could be set empty. This bug probably also applied to
5813 41. The strip_trailing_dot option was not being applied to the address given
5814 with the -f command-line option.
5816 42. The code for reading a message's header from the spool was incrementing
5817 $received_count, but never initializing it. This meant that the value was
5818 incorrect (doubled) while delivering a message in the same process in which
5819 it was received. In the most common configuration of Exim, this never
5820 happens - a fresh exec is done - but it can happen when
5821 deliver_drop_privilege is set.
5823 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5824 it now includes up to 150 characters of the unexpected data in the log
5827 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5828 and building data strings. The size of both of these buffers was 10 000
5829 bytes - far larger than anybody would *ever* want, thought I. Needless to
5830 say, somebody hit the limit. I have increased the maximum line length to
5831 20 000 and the maximum data length of concatenated lines to 100 000. I have
5832 also fixed two bugs, because there was no checking on these buffers. Tsk,
5833 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5834 buffer is too small.
5836 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5837 lsearch lookups. Now it does.
5839 46. When parsing a route_list item in a manualroute router, a fixed-length
5840 buffer was used for the list of hosts. I made this 1024 bytes long,
5841 thinking that nobody would ever have a list of hosts that long. Wrong.
5842 Somebody had a whole pile of complicated expansion conditions, and the
5843 string was silently truncated, leading to an expansion error. It turns out
5844 that it is easier to change to an unlimited length (owing to other changes
5845 that have happened since this code was originally written) than to build
5846 structure for giving a limitation error. The length of the item that
5847 expands into the list of hosts is now unlimited.
5849 47. The lsearch lookup could not handle data where the length of text line was
5850 more than 4095 characters. Such lines were truncated, leading to shortened
5851 data being returned. It should now handle lines of any length.
5853 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5854 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5857 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5858 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5859 no longer happens in comment lines. A list of replacements is now placed
5860 at the head of all of the source files, except those whose only change is
5861 to replace PERL_COMMAND in the very first #! line.
5863 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5864 messages on the queue, with a bottom-up merge sort, using code contributed
5865 by Michael Haardt. This should make operations like -bp somewhat faster on
5866 large queues. It won't affect queue runners, except when queue_run_in_order
5869 51. Installed eximstats 1.31 in the distribution.
5871 52. Added support for SRV lookups to the dnslookup router.
5873 53. If an ACL referred to $message_body or $message_body_end, the value was not
5874 reset for any messages that followed in the same SMTP session.
5876 54. The store-handling optimization for building very long strings was not
5877 differentiating between the different store pools. I don't think this
5878 actually made any difference in practice, but I've tidied it.
5880 55. While running the routers to verify a sender address, $sender_address
5881 was still set to the sender address. This is wrong, because when routing to
5882 send a bounce to the sender, it would be empty. Therefore, I have changed
5883 it so that, while verifying a sender address, $sender_address is set to <>.
5884 (There is no change to what happens when verifying a recipient address.)
5886 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5887 target A or AAAA records (if not already returned) without resetting the
5888 qualify_single or search_parents options of the DNS resolver. These are
5889 inappropriate in this case because the targets of MX and SRV records must
5890 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5891 target that, when qualified, matched something in the local domain. These
5892 two options are now turned off when doing these lookups.
5894 57. It seems that at least some releases of Reiserfs (which does not have the
5895 concept of a fixed number of inodes) returns zero and not -1 for the
5896 number of available inodes. This interacted badly with check_spool_inodes,
5897 which assumed that -1 was the "no such thing" setting. What I have done is
5898 to check that the total number of inodes is greater than zero before doing
5899 the test of how many are available.
5901 58. When a "warn" ACL statement has a log_message modifier, the message is
5902 remembered, and not repeated. This is to avoid a lot of repetition when a
5903 message has many recipients that cause the same warning to be written.
5904 However, Exim was preserving the list of already written lines for an
5905 entire SMTP session, which doesn't seem right. The memory is now reset if a
5906 new message is started.
5908 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5909 debugging output unless log_rewrite was also set.
5911 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5912 of (int)(handle) when we know that handle contains (void *)(-1).
5914 61. The Exim daemon panic-logs an error return when it closes the incoming
5915 connection. However "connection reset by peer" seems to be common, and
5916 isn't really an error worthy of noting specially, so that particular error
5919 62. When Exim is trying to find all the local interfaces, it used to panic and
5920 die if the ioctl to get the interface flags failed. However, it seems that
5921 on at least one OS (Solaris 9) it is possible to have an interface that is
5922 included in the list of interfaces, but for which you get a failure error
5923 for this call. This happens when the interface is not "plumbed" into a
5924 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5925 failure of the "get flags" call assumes that the interface is down.
5927 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5928 makes life easier for people who are doing arithmetic on fields extracted
5929 from dates, where you often get leading zeros that should not be
5930 interpreted as octal.
5932 64. Added qualify_domain to the redirect router, to override the global
5935 65. If a pathologically long header line contained very many addresses (the
5936 report of this problem mentioned 10 000) and each of them was rewritten,
5937 Exim could use up a very large amount of memory. (It kept on making new
5938 copies of the header line as it rewrote, and never released the old ones.)
5939 At the expense of a bit more processing, the header rewriting function has
5940 been changed so that it no longer eats memory in this way.
5942 66. The generation of the Received: header has been moved from the time that a
5943 message starts to be received, to the time that it finishes. The timestamp
5944 in the Received: header should now be very close to that of the <= log
5945 line. There are two side-effects of this change:
5947 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5948 logged header lines no longer include the local Received: line, because
5949 it has not yet been created. The same applies to a copy of the message
5950 that is returned to a non-SMTP sender when a message is rejected.
5952 (b) When a filter file is tested using -bf, no additional Received: header
5953 is added to the test message. After some thought, I decided that this
5956 This change does not affect the value of $received_for. It is still set
5957 after address rewriting, but before local_scan() is called.
5959 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5961 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5962 gave an unhelpful panic error message, and a defer error. I have managed to
5963 change this behaviour so that it now rejects any supplied certificate,
5964 which seems right, as the list of acceptable certificates is empty.
5966 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5967 gave an unhelpful defer error. I have not managed to make this reject any
5968 supplied certificates, but the error message it gives is "no certificate
5969 supplied", which is not helpful.
5971 70. exigrep's output now also includes lines that are not associated with any
5972 message, but which match the given pattern. Implemented by a patch from
5973 Martin Sluka, which also tidied up the Perl a bit.
5975 71. Recipient callout verification, like sender verification, was using <> in
5976 the MAIL FROM command. This isn't really the right thing, since the actual
5977 sender may affect whether the remote host accepts the recipient or not. I
5978 have changed it to use the actual sender in the callout; this means that
5979 the cache record is now keyed on a recipient/sender pair, not just the
5980 recipient address. There doesn't seem to be a real danger of callout loops,
5981 since a callout by the remote host to check the sender would use <>.
5982 [SEE ABOVE: changed after hitting problems.]
5984 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5985 temporary errors. However, in the case of such a code being given after
5986 the end of a data transmission (i.e. after ".") Exim was failing to write
5987 a retry record for the message. (Yes, there was some broken host that was
5988 actually sending 8xx at this point.)
5990 73. An unknown lookup type in a host list could cause Exim to panic-die when
5991 the list was checked. (An example that provoked this was putting <; in the
5992 middle of a list instead of at the start.) If this happened during a DATA
5993 ACL check, a -D file could be left lying around. This kind of configuration
5994 error no longer causes Exim to die; instead it causes a defer error. The
5995 incident is still logged to the main and panic logs.
5997 74. Buglet left over from Exim 3 conversion. The message "too many messages
5998 in one connection" was written to the rejectlog but not the mainlog, except
5999 when address rewriting (yes!) was being logged.
6001 75. Added write_rejectlog option.
6003 76. When a system filter was run not as root (that is, when system_filter_user
6004 was set), the values of the $n variables were not being returned to the
6005 main process; thus, they were not subsequently available in the $sn
6008 77. Added +return_path_on_delivery log selector.
6010 78. A connection timeout was being treated differently from recipients deferred
6011 when testing hosts_max_try with a message that was older than the host's
6012 retry timeout. (The host should not be counted, thus allowing all hosts to
6013 be tried at least once before bouncing.) This may have been the cause of an
6014 occasionally reported bug whereby a message would remain on the queue
6015 longer than the retry timeout, but would be bounced if a delivery was
6016 forced. I say "may" because I never totally pinned down the problem;
6017 setting up timeout/retry tests is difficult. See also the next item.
6019 79. The ultimate address timeout was not being applied to errors that involved
6020 a combination of host plus message (for example, a timeout on a MAIL
6021 command). When an address resolved to a number of possible hosts, and they
6022 were not all tried for each delivery (e.g. because of hosts_max_try), a
6023 message could remain on the queue longer than the retry timeout.
6025 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6028 81. Fixed an obscure SMTP outgoing bug which required at least the following
6029 conditions: (a) there was another message waiting for the same server;
6030 (b) the server returned 5xx to all RCPT commands in the first message so
6031 that the message was not completed; (c) the server dropped the connection
6032 or gave a negative response to the RSET that Exim sends to abort the
6033 transaction. The observed case was a dropped connection after DATA that had
6034 been sent in pipelining mode. That is, the server had advertised PIPELINING
6035 but was not implementing it correctly. The effect of the bug was incorrect
6036 behaviour, such as trying another host, and this could lead to a crash.
6042 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6043 and daemon.c were passed as pointers to ints; they should have been
6044 pointers to socklen_t variables (which are typically unsigned ints).
6046 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6049 3. Fixed a really odd bug that affected only the testing scheme; patching a
6050 certain fixed string in the binary changed the value of another string that
6051 happened to be identical to the end of the original first string.
6053 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6054 name", it returns that address as the IP address. On some operating
6055 systems (e.g. Solaris), it also passes back the IP address string as the
6056 "host name". However, on others (e.g. Linux), it passes back an empty
6057 string. Exim wasn't checking for this, and was changing the host name to an
6058 empty string, assuming it had been canonicalized.
6060 5. Although rare, it is permitted to have more than one PTR record for a given
6061 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6062 all the names associated with an address, because they do in Solaris.
6063 However, it seems that they do not in Linux for data that comes from the
6064 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6065 I found this out when I moved to a new Linux workstation and tried to run
6066 the Exim test suite.
6068 To get round this problem I have changed the code so that it now does its
6069 own call to the DNS to look up PTR records when searching for a host name.
6070 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6071 addresses that are only in /etc/hosts are still found.
6073 This behaviour is, however, controlled by an option called host_lookup_
6074 order, which defaults to "bydns:byaddr". If people want to use the other
6075 order, or indeed, just use one or the other means of lookup, they can
6076 specify it in this variable.
6078 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6079 some operating systems, this comes back from gethostbyaddr() as an empty
6080 string, and this is what Exim used to test for. However, it seems that in
6081 other systems, "." is yielded. Exim now tests for this case too.
6083 7. The values of check_spool_space and check_log_space are now held internally
6084 as a number of kilobytes instead of an absolute number of bytes. If a
6085 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6086 kilobyte. This means that much larger values can be stored.
6088 8. Exim monitor: an attempt to get the action menu when not actually pointing
6089 at a message produces an empty menu entitled "No message selected". This
6090 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6091 no entries in it ("Shell widget menu has zero width and/or height"). So I
6092 have added a single, blank menu entry in this case.
6094 9. Added ${quote_local_part.
6096 10. MIME decoding is now applied to the contents of Subject: header lines when
6099 11. Now that a reference to $sender_host_address automatically causes a reverse
6100 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6101 host lookup before query-style lookups in lists that might use this
6102 variable. This has therefore been abolished, and the "net-" prefix is no
6103 longer necessary for query-style lookups.
6105 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6106 have been a typo for LFLAGS, so it has been changed.
6108 13. The install script calls Exim with "-C /dev/null" in order to find the
6109 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6110 to be output. However, since Exim outputs its version number before the
6111 error, it didn't break the script. It just looked ugly. I fixed this by
6112 always allowing "-C /dev/null" if the caller is root.
6114 14. Ignore overlarge ACL variable number when reading spool file - insurance
6115 against a later release with more variables having written the file.
6117 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6118 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6119 this, as well as the form without the "IPv6" on the front.
6121 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6122 OS/Makefile-Darwin file.
6124 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6125 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6127 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6128 variable. However, some OS use ints for inodes. I've added cast to long int
6129 to get rid of the compiler warning.
6131 19. I had forgotten to lock out "/../" in configuration file names when
6132 ALT_CONFIG_PREFIX was set.
6134 20. Routers used for verification do not need to specify transports. However,
6135 if such a router generated a host list, and callout was configured, Exim
6136 crashed, because it could not find a port number from the (non-existent)
6137 transport. It now assumes port 25 in this circumstance.
6139 21. Added the -t option to exigrep.
6141 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6142 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6143 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6144 them is set, LOOKUP_LSEARCH is forced.
6146 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6148 24. Added sender and host information to the "rejected by local_scan()" log
6149 line; previously there was no indication of these.
6151 25. Added .include_if_exists.
6153 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6154 Linux. It turns out that not all file systems support this. Apparently some
6155 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6156 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6157 not supported on the file descriptor, is now ignored when Exim is trying to
6158 sync a directory. This applies only to Linux.
6160 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6162 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6163 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6166 29. Added support for maildirsize files from supplied patch (modified a bit).
6168 30. The use of :fail: followed by an empty string could lead Exim to respond to
6169 sender verification failures with (e.g.):
6171 550 Verification failed for <xxx>
6172 550 Sender verify failed
6174 where the first response line was missing the '-' that indicates it is not
6175 the final line of the response.
6177 31. The loop for finding the name of the user that called Exim had a hardwired
6178 limit of 10; it now uses the value of finduser_retries, which is used for
6179 all other user lookups.
6181 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6184 33. Exim was neglecting to zero errno before one call of strtol() when
6185 expanding a string and expecting an integer value. On some systems this
6186 resulted in spurious "integer overflow" errors. Also, it was casting the
6187 result into an int without checking.
6189 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6190 did not work. The code looks as if it has *never* worked, though it appears
6191 to have been documented since at least release 1.62. I have made it work.
6193 35. The "timeout_DNS" error in retry rules, also documented since at least
6194 1.62, also never worked. As it isn't clear exactly what this means, and
6195 clearly it isn't a major issue, I have abolished the feature by treating it
6196 as "timeout", and writing a warning to the main and panic logs.
6198 36. The display of retry rules for -brt wasn't always showing the error code
6201 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6202 timeout_connect_A, timeout_connect_MX.
6204 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6205 to the empty sender.
6207 39. The daemon was not analysing the content of -oX till after it had closed
6208 stderr and disconnected from the controlling terminal. This meant that any
6209 syntax errors were only noted on the panic log, and the return code from
6210 the command was 0. By re-arranging the code a little, I've made the
6211 decoding happen first, so such errors now appear on stderr, and the return
6212 code is 1. However, the actual setting up of the sockets still happens in
6213 the disconnected process, so errors there are still only recorded on the
6216 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6217 connections (as happens on some IP stacks) was logged at start up time as
6218 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6219 it from "IPv6 and IPv4", which means that two separate sockets are being
6222 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6223 says whether AF_INET or AF_INET6 was passed as an argument.
6225 42. Exiwhat output was messed up when time zones were included in log
6228 43. Exiwhat now gives more information about the daemon's listening ports,
6229 and whether -tls-on-connect was used.
6231 44. The "port" option of the smtp transport is now expanded.
6233 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6234 silently ignored. Now an error message is written to the main and panic
6237 46. There's a new ACL modifier called "logwrite" which writes to a log file
6238 as soon as it is encountered.
6240 47. Added $local_user_uid and $local_user_gid at routing time.
6242 48. Exim crashed when trying to verify a sender address that was being
6245 49. Exim was recognizing only a space character after ".include". It now also
6246 recognizes a tab character.
6248 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6249 extracting the relevant information from the specification. The man page no
6250 longer contains scrambled data for the -d option, and I've added a section
6251 at the front about calling Exim under different names.
6253 51. Added "extra_headers" argument to the "mail" command in filter files.
6255 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6258 53. Installed eximstats 1.29.
6260 54. Added transport_filter_timeout as a generic transport option.
6262 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6263 Cc: header lines. This was required by RFC 822, but it not required by RFC
6266 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6267 incoming messages that did not have them. Now it does so only if the
6268 message originates locally, that is, if there is no associated remote host
6269 address. When Resent- header lines are present, this applies to the Resent-
6270 lines rather than the non-Resent- lines.
6272 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6273 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6275 58. Messages for configuration errors now include the name of the main
6276 configuration file - useful now that there may be more than one file in a
6277 list (.included file names were always shown).
6279 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6280 for those rare installations that do not start the daemon as root or run it
6281 setuid root. I've cut out the call to initgroups() if the daemon is not
6284 60. The Exim user and group can now be bound into the binary as text strings
6285 that are looked up at the start of Exim's processing.
6287 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6289 62. Added $mailstore_basename variable.
6291 63. Installed patch to sieve.c from Michael Haardt.
6293 64. When Exim failed to open the panic log after failing to open the main log,
6294 the original message it was trying to log was written to stderr and debug
6295 output, but if they were not available (the usual case in production), it
6296 was lost. Now it is written to syslog before the two lines that record the
6297 failures to open the logs.
6299 65. Users' Exim filters run in subprocesses under the user's uid. It is
6300 possible for a "deliver" command or an alias in a "personal" command to
6301 provoke an address rewrite. If logging of address rewriting is configured,
6302 this fails because the process is not running as root or exim. There may be
6303 a better way of dealing with this, but for the moment (because 4.30 needs
6304 to be released), I have disabled address rewrite logging when running a
6305 filter in a non-root, non-exim process.
6311 1. The buildconfig auxiliary program wasn't quoting the value set for
6312 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6313 not defined. This bug was masked in 4.22 by the effect that was fixed in
6316 2. Some messages that were rejected after a message id was allocated were
6317 shown as "incomplete" by exigrep. It no longer does this for messages that
6318 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6320 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6321 have allow_domain_literals set, the ID did not get logged in the <= line.
6322 Domain literals are now always recognized in Message-ID: header lines.
6324 4. The first argument for a ${extract expansion item is the key name or field
6325 number. Leading and trailing spaces in this item were not being ignored,
6326 causing some misleading effects.
6328 5. When deliver_drop_privilege was set, single queue runner processes started
6329 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6330 same command in the process it spins off) were not dropping privilege.
6332 6. When the daemon running as "exim" started a queue runner, it always
6333 re-executed Exim in the spun-off process. This is a waste of effort when
6334 deliver_drop_privilege is set. The new process now just calls the
6335 queue-runner function directly.
6341 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6344 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6345 ignored. Though the use of -oP was forcing the writing of a pid file, it
6346 was always written to the default place.
6348 3. If the message "no IP address found for host xxxx" is generated during
6349 incoming verification, it is now followed by identification of the incoming
6350 connection (so you can more easily find what provoked it).
6352 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6354 5. Added some features to "harden" Exim a bit more against certain attacks:
6356 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6357 be put in Local/Makefile. This is like the never_users runtime option,
6358 but it cannot be overridden. The default setting is "root".
6360 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6361 prefix string with which any file named in a -C command line option
6364 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6365 is retained for -C and -D only if the caller of Exim is root. Without
6366 it, the exim user may also use -C and -D and retain privilege.
6368 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6369 command line option is disabled.
6371 6. Macro names set by the -D option must start with an upper case letter, just
6372 like macro names defined in the configuration file.
6374 7. Added "dereference=" facility to LDAP.
6376 8. Two instances of the typo "uknown" in the source files are fixed.
6378 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6379 the Configure-Makefile script screwed up while processing it.
6381 10. Incorporated PCRE 4.4.
6383 11. The SMTP synchronization check was not operating right at the start of an
6384 SMTP session. For example, it could not catch a HELO sent before the client
6385 waited for the greeting. There is now a check for outstanding input at the
6386 point when the greeting is written. Because of the duplex, asynchronous
6387 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6388 way, but not yet received, when the check is performed.
6390 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6391 on TCP/IP sockets, because this apparently causes some broken clients to
6394 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6395 unchanged) from the Cygwin maintainer.
6397 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6398 was supported instead of testing for mbx. Effectively a typo.
6400 15. The spa authenticator server code was not checking that the input it
6401 received was valid base64.
6403 16. The debug output line for the "set" modifier in ACLs was not showing the
6404 name of the variable that was being set.
6406 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6408 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6409 lookup on its own. Something else was needed to trigger the lookup. For
6410 example, a match in host_lookup or the need for a host name in a host list.
6411 Now, if $sender_host_name is referenced and the host name has not yet been
6412 looked up, a lookup is performed. If the lookup fails, the variable remains
6413 empty, and $host_lookup_failed is set to "1".
6415 19. Added "eqi" as a case-independent comparison operator.
6417 20. The saslauthd authentication condition could segfault if neither service
6418 nor realm was specified.
6420 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6421 error message that was logged was misleading, and incoming SMTP
6422 connections were dropped. The message is now more accurate, and temporary
6423 errors are given to SMTP connections.
6425 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6426 (except RSET) with a 421 error, until QUIT is received. However, it was
6427 failing to send a response to QUIT.
6429 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6430 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6431 ACL is now run after the helo_try_verify_hosts code.
6433 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6434 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6435 case-independent, so other case variants are also recognized). Apparently
6436 some systems use these upper case variants.
6438 25. If more than two messages were waiting for the same host, and a transport
6439 filter was specified for the transport, Exim sent two messages over the
6440 same TCP/IP connection, and then failed with "socket operation on non-
6441 socket" when it tried to send the third.
6443 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6445 27. The extern definition of crypt16() in expand.c was not being excluded when
6446 the OS had its own crypt16() function.
6448 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6449 as a preferred synonym for return_size_limit, both as an option and as an
6452 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6454 30. Changed the default configuration ACL to relax the local part checking rule
6455 for addresses that are not in any local domains. For these addresses,
6456 slashes and pipe symbols are allowed within local parts, but the sequence
6457 /../ is explicitly forbidden.
6459 31. SPA server authentication was not clearing the challenge buffer before
6462 32. log_message in a "warn" ACL statement was writing to the reject log as
6463 well as to the main log, which contradicts the documentation and doesn't
6464 seem right (because no rejection is happening). So I have stopped it.
6466 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6467 However, I am unable to do any testing of this.
6469 34. Fixed an infelicity in the appendfile transport. When checking directories
6470 for a mailbox, to see if any needed to be created, it was accidentally
6471 using path names with one or more superfluous leading slashes; tracing
6472 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6474 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6475 contents are added to the log line that is written for every discarded
6476 recipient. (Previously a log_message setting was ignored.)
6478 36. The ${quote: operator now quotes the string if it is empty.
6480 37. The install script runs exim in order to find its version number. If for
6481 some reason other than non-existence or emptiness, which it checks, it
6482 could not run './exim', it was installing it with an empty version number,
6483 i.e. as "exim-". This error state is now caught, and the installation is
6486 38. An argument was missing from the function that creates an error message
6487 when Exim fails to connect to the socket for saslauthd authentication.
6488 This could cause Exim to crash, or give a corrupted message.
6490 39. Added isip, isip4, and isip6 to ${if conditions.
6492 40. The ACL variables $acl_xx are now saved with the message, and can be
6493 accessed later in routers, transports, and filters.
6495 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6496 strings in the file are not string-expanded.
6498 42. If a MAIL command specified a SIZE value that was too large to fit into an
6499 int variable, the check against message_size_limit failed. Such values are
6500 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6501 day this will have to be increased, but I don't think I want to be around
6502 when emails are that large.
6509 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6510 iconv() is not standard in FreeBSD.
6512 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6513 IPv6 enabled. The observed symptom was a segmentation fault on return from
6514 the function os_common_find_running_interfaces() in src/os.c.
6516 3. In the check_special_case() function in daemon.c I had used "errno" as an
6517 argument name, which causes warnings on some systems. This was basically a
6518 typo, since it was named "eno" in the comments!
6520 4. The code that waits for the clock to tick (at a resolution of some fraction
6521 of a second) so as to ensure message-id uniqueness was always waiting for
6522 at least one whole tick, when it could have waited for less. [This is
6523 almost certainly not relevant at current processor speeds, where it is
6524 unlikely to ever wait at all. But we try to future-proof.]
6526 5. The function that sleeps for a time interval that includes fractions of a
6527 second contained a race. It did not block SIGALRM between setting the
6528 timer, and suspending (a couple of lines later). If the interval was short
6529 and the sigsuspend() was delayed until after it had expired, the suspension
6530 never ended. On busy systems this could lead to processes getting stuck for
6533 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6534 process, before it forks any delivery processes. The open lookup caching
6535 mechanism meant that the open file or database connection was passed into
6536 the delivery process. The problem was that delivery processes always tidy
6537 up cached lookup data. This could cause a problem for the next delivery
6538 process started by the queue runner, because the external queue runner
6539 process does not know about the closure. So the next delivery process
6540 still has data in the lookup cache. In the case of a file lookup, there was
6541 no problem because closing a file descriptor in a subprocess doesn't affect
6542 the parent. However, if the lookup was caching a connection to a database,
6543 the connection was closed, and the second delivery process was likely to
6544 see errors such as "PGSQL: query failed: server closed the connection
6545 unexpectedly". The problem has been fixed by closing all cached lookups
6546 in a queue runner before running a delivery process.
6548 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6549 seem to have the "const" qualifier which it has on other OS. I've
6552 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6553 *of the same type* (client or server) with the same public name that an
6554 error should be diagnosed.
6556 9. When Exim looked up a host name for an IP address, but failed to find the
6557 original IP address when looking up the host name (a safety check), it
6558 output the message "<ip address> does not match any IP for NULL", which was
6559 confusing, to say the least. The bug was that the host name should have
6560 appeared instead of "NULL".
6562 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6563 user that is built into the binary, and the -C or -D options is used, root
6564 privilege is dropped before the configuration file is read. In addition,
6565 logging is switched to stderr instead of the normal log files. If the
6566 configuration then re-defines the Exim user, the unprivileged environment
6567 is probably not what is expected, so Exim logs a panic warning message (but
6570 However, if deliver_drop_privilege is set, the unprivileged state may well
6571 be exactly what is intended, so the warning has been cut out in that case,
6572 and Exim is allowed to try to write to its normal log files.
6578 1. smtp_return_error_details was not giving details for temporary sender
6579 or receiver verification errors.
6581 2. Diagnose a configuration error if two authenticators have the same public
6584 3. Exim used not to create the message log file for a message until the first
6585 delivery attempt. This could be confusing when incoming messages were held
6586 for policy or load reasons. The message log file is now created at the time
6587 the message is received, and an initial "Received" line is written to it.
6589 4. The automatically generated man page for command line options had a minor
6590 bug that caused no ill effects; however, a more serious problem was that
6591 the procedure for building the man page automatically didn't always
6592 operate. Consequently, release 4.20 contains an out-of-date version. This
6593 shouldn't happen again.
6595 5. When building Exim with embedded Perl support, the script that builds the
6596 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6597 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6599 6. The freeze_tell option was not being used for messages that were frozen on
6600 arrival, either by an ACL or by local_scan().
6602 7. Added the smtp_incomplete_transaction log selector.
6604 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6605 was accepting AUTH without a new EHLO.
6607 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6608 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6611 10. Logging of TCP/IP connections (when configured) now happens in the main
6612 daemon process instead of the child process, so that the TCP/IP connection
6613 count is more accurate (but it can never be perfect).
6615 11. The use of "drop" in a nested ACL was not being handled correctly in the
6616 outer ACL. Now, if condition failure induced by the nested "drop" causes
6617 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6618 or "require"), the connection is dropped.
6620 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6621 that yield "discard" can now be used with an "accept" or a "discard" verb,
6622 but an error is generated for any others (because I can't see a useful way
6623 to define what should happen).
6625 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6626 are now processed in the same way as lines in the Exim configuration file.
6627 In particular, continuation lines are supported.
6629 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6631 15. Added -ti meaning -t -i.
6633 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6634 domains, and warn by logging if others are found.
6636 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6637 returned from GIFCONF when Exim is trying to find the list of interfaces on
6638 a host. The code in os.c has been modified to copy each ifreq to an aligned
6639 structure in all cases.
6641 Also, in some cases, the returned ifreq's were being copied to a 'struct
6642 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6643 means the last couple of bytes of an IPv6 address could be chopped if the
6644 ifreq contained only a normal sockaddr (14 bytes storage).
6646 18. Named domain lists were not supported in the hosts_treat_as_local option.
6647 An entry such as +xxxx was not recognized, and was treated as a literal
6650 19. Ensure that header lines added by a DATA ACL are included in the reject log
6651 if the ACL subsequently rejects the message.
6653 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6654 MD5 (which is deprecated).
6656 21. When testing a filter file using -bf, Exim was writing a message when it
6657 took the sender from a "From " line in the message, but it was not doing so
6658 when it took $return_path from a Return-Path: header line. It now does.
6660 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6661 with a valid header line field name (a series of printing characters
6662 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6664 23. Changed "disc" in the source to "disk" to conform to the documentation and
6665 the book and for uniformity.
6667 24. Ignore Sendmail's -Ooption=value command line item.
6669 25. When execve() failed while trying to run a command in a pipe transport,
6670 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6671 could be confused with a return value of 69 from the command itself. This
6672 has been changed to 127, the value the shell returns if it is asked to run
6673 a non-existent command. The wording for the related log line suggests a
6674 non-existent command as the problem.
6676 26. If received_header_text expands to an empty string, do not add a Received:
6677 header line to the message. (Well, it adds a token one on the spool, but
6678 marks it "old" so that it doesn't get used or transmitted.)
6680 27. Installed eximstats 1.28 (addition of -nt option).
6682 28. There was no check for failure on the call to getsockname() in the daemon
6683 code. This can fail if there is a shortage of resources on the system, with
6684 ENOMEM, for example. A temporary error is now given on failure.
6686 29. Contrary to the C standard, it seems that in some environments, the
6687 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6688 program. Exim now does this explicitly; it affects the formatting of
6689 timestamps using strftime().
6691 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6692 complaints. I've now initialized all the variables, to avoid this.
6694 32. Header lines added by a system filter were not being "seen" during
6695 transport-time rewrites.
6697 33. The info_callback() function passed to OpenSSL is set up with type void
6698 (*)(SSL *, int, int), as described somewhere. However, when calling the
6699 function (actually a macro) that sets it up, the type void(*)() is
6700 expected. I've put in a cast to prevent warnings from picky compilers.
6702 34. If a DNS black list lookup found a CNAME record, but there were no A
6703 records associated with the domain it pointed at, Exim crashed.
6705 35. If a DNS black list lookup returned more than one A record, Exim ignored
6706 all but the first. It now scans all returned addresses if a particular IP
6707 value is being sought. In this situation, the contents of the
6708 $dnslist_value variable are a list of all the addresses, separated by a
6711 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6712 to accept a host name and all its aliases if the forward lookup for any of
6713 them yielded the IP address of the incoming connection. Now it accepts only
6714 those names whose forward lookup yields the correct IP address. Any other
6715 names are discarded. This closes a loophole whereby a rogue DNS
6716 administrator could create reverse DNS records to break through a
6717 wildcarded host restriction in an ACL.
6719 37. If a user filter or a system filter that ran in a subprocess used any of
6720 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6721 the wrong values were passed to the pipe command ($thisaddress had the
6722 value of $0, $0 had the value of $1, etc). This bug was introduced by
6723 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6725 38. Improved the line breaking for long SMTP error messages from ACLs.
6726 Previously, if there was no break point between 40 and 75 characters, Exim
6727 left the rest of the message alone. Two changes have been made: (a) I've
6728 reduced the minimum length to 35 characters; (b) if it can't find a break
6729 point between 35 and 75 characters, it looks ahead and uses the first one
6730 that it finds. This may give the occasional overlong line, but at least the
6731 remaining text gets split now.
6733 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6734 file descriptors might be low, and that setting 1000 would always raise it.
6735 It turns out that in some environments, the limit is already over 1000 and
6736 that lowering it causes trouble. So now Exim takes care not to decrease it.
6738 40. When delivering a message, the value of $return_path is set to $sender_
6739 address at the start of routing (routers may change the value). By an
6740 oversight, this default was not being set up when an address was tested by
6741 -bt or -bv, which affected the outcome if any router or filter referred to
6744 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6745 log only when "message" or "log_message" are set. However, if one of the
6746 conditions was an address verification, or a call to a nested ACL, the
6747 messages generated by the underlying test were being passed through. This
6748 no longer happens. The underlying message is available in $acl_verify_
6749 message for both "message" and "log_message" expansions, so it can be
6750 passed through if needed.
6752 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6753 new expansion $bh_ to give the encoded byte string without charset
6754 translation. Translation happens only if iconv() is available; HAVE_ICONV
6755 indicates this at build time. HEADERS_CHARSET gives the charset to
6756 translate to; headers_charset can change it in the configuration, and
6757 "headers charset" can change it in an individual filter file.
6759 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6760 that creates RFC 2047 encoded "words" labels them as that charset instead
6761 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6762 expansion operator; (ii) when Exim creates a From: line for a local
6763 message; (iii) when a header line is rewritten to include a "phrase" part.
6765 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6766 buggy, causing it to skip the first lines of messages whose message ID
6767 ended in 'D'. This would not have bitten before Exim release 4.14, because
6768 message IDs were unlikely to end in 'D' before then. The effect was to have
6769 incorrect size information for certain domains.
6771 45. #include "config.h" was missing at the start of the crypt16.c module. This
6772 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6775 46. If there was a timeout during a "random" callout check, Exim treated it as
6776 a failure of the random address, and carried on sending RSET and the real
6777 address. If the delay was just some slowness somewhere, the response to the
6778 original RCPT would be taken as a response to RSET and so on, causing
6779 mayhem of various kinds.
6781 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6782 when I implemented it. It didn't allow for the fact that some option values
6783 may legitimately be negative (e.g. size_addition), and it didn't even do
6784 the right test for positive values.
6786 48. Domain names in DNS records are case-independent. Exim always looks them up
6787 in lower case. Some resolvers return domain names in exactly the case they
6788 appear in the zone file, that is, they may contain uppercase letters. Not
6789 all resolvers do this - some return always lower case. Exim was treating a
6790 change of case by a resolver as a change of domain, similar to a widening
6791 of a domain abbreviation. This triggered its re-routing code and so it was
6792 trying to route what was effectively the same domain again. This normally
6793 caused routing to fail (because the router wouldn't handle the domain
6794 twice). Now Exim checks for this case specially, and just changes the
6795 casing of the domain that it ultimately uses when it transmits the message
6798 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6801 50. If a filter generated a file delivery with a non-absolute name (possible if
6802 no home directory exists for the router), the forbid_file option was not
6805 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6806 the existing equality matching.
6808 52. Exim was using ints instead of ino_t variables in some places where it was
6809 dealing with inode numbers.
6811 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6812 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6813 called TMPDIR, and if it finds it is different, it changes its value.
6815 54. The smtp_printf() function is now made available to local_scan() so
6816 additional output lines can be written before returning. There is also an
6817 smtp_fflush() function to enable the detection of a dropped connection.
6818 The variables smtp_input and smtp_batched_input are exported to
6821 55. Changed the default runtime configuration: the message "Unknown user"
6822 has been removed from the ACL, and instead placed on the localuser router,
6823 using the cannot_route_message feature. This means that any verification
6824 failures that generate their own messages won't get overridden. Similarly,
6825 the "Unrouteable address" message that was in the ACL for unverifiable
6826 relay addresses has also been removed.
6828 56. Added hosts_avoid_esmtp to the smtp transport.
6830 57. The exicyclog script was not checking for the esoteric option
6831 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6832 will work only if exicyclog is run under the appropriate euid.
6834 58. Following a discussion on the list, the rules by which Exim recognises line
6835 endings on incoming messages have been changed. The -dropcr and drop_cr
6836 options are now no-ops, retained only for backwards compatibility. The
6837 following line terminators are recognized: LF CRLF CR. However, special
6838 processing applies to CR:
6840 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6841 nor a local message in the state where . is a terminator.
6843 (ii) If a bare CR is encountered in a header line, an extra space is added
6844 after the line terminator so as not to end the header. The reasoning
6845 behind this is that bare CRs in header lines are most likely either
6846 to be mistakes, or people trying to play silly games.
6848 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6849 was being incorrectly given as 18 bytes larger than it should have been.
6850 This is a VOB (very old bug).
6852 60. This may never have affected anything current, but just in case it has:
6853 When the local host is found other than at the start of a list of hosts,
6854 the local host, those with the same MX, and any that follow, are discarded.
6855 When the list in question was part of a longer list of hosts, the following
6856 hosts (not currently being processed) were also being discarded. This no
6857 longer happens. I'm not sure if this situation could ever has previously
6860 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6863 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6864 header. This is something that is recommended in a new Internet Draft, and
6865 is something that is documented as being done by Sendmail. There are two
6866 possible values. For messages generated by the autoreply transport, Exim
6869 Auto-Submitted: auto-replied
6871 whereas for all other generated messages (e.g. bounces) it adds
6873 Auto-Submitted: auto-generated
6875 63. The "personal" condition in filters now includes a test for the
6876 Auto-Submitted: header. If it contains the string "auto-" the message it
6877 not considered personal.
6879 64. Added rcpt_include_affixes as a generic transport option.
6881 65. Added queue_only_override (default true).
6883 66. Added the syslog_duplication option.
6885 67. If what should have been the first header line of a message consisted of
6886 a space followed by a colon, Exim was mis-interpreting it as a header line.
6887 It isn't of course - it is syntactically invalid and should therefore be
6888 treated as the start of the message body. The misbehaviour could have
6889 caused a number of strange effects, including loss of data in subsequent
6890 header lines, and spool format errors.
6892 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6893 client host had authenticated. This control can now be exercised by an ACL
6894 for more flexibility.
6896 69. By default, callouts do not happen when testing with -bh. There is now a
6897 variant, -bhc, which does actually run the callout code, including
6898 consulting and updating the callout cache.
6900 70. Added support for saslauthd authentication, courtesy of Alexander
6903 71. If statvfs() failed on the spool or log directories while checking their
6904 size for availability, Exim confusingly gave the error "space shortage".
6905 Furthermore, in debugging mode it crashed with a floating point exception.
6906 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6907 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6908 serious problem, Exim now writes to the main and panic logs when this
6909 happens, with details of the failure. It then refuses to accept the
6910 incoming message, giving the message "spool directory problem" or "log
6911 directory problem" with a 421 code for SMTP messages.
6913 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6914 0, 1, and 2 exist, because some OS complain for execs without them (see
6915 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6916 descriptors. However, the code omitted to check that the open succeeded,
6917 causing mysterious errors if for some reason the permissions on /dev/null
6918 got screwed. Now Exim writes a message to the main and panic logs, and
6919 bombs out if it can't open /dev/null.
6921 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6922 interact so that it is all more flexible. It is supposed to remain
6923 backwards compatible. Also added extra_local_interfaces.
6925 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6926 to bomb out with an assertion failure - to the client this appears as a
6927 connection drop. This problem occurs in the part of the code that was taken
6928 from the Samba project. Fortunately, the assertion is in a very simple
6929 function, so I have fixed this by reproducing the function inline in the
6930 one place where it is called, and arranging for authentication to fail
6931 instead of killing the process with assert().
6933 75. The SPA client code was not working when the server requested OEM rather
6934 than Unicode encoding.
6936 76. Added code to make require_files with a specific uid setting more usable in
6937 the case where statting the file as root fails - usually a non-root-mounted
6938 NFS file system. When this happens and the failure is EACCES, Exim now
6939 forks a subprocess and does the per-uid checking as the relevant uid.
6941 77. Added process_log_path.
6943 78. If log_file_path was not explicitly set, a setting of check_log_space or
6944 check_log_inodes was ignored.
6946 79. If a space check for the spool or log partitions fails, the incident is now
6947 logged. Of course, in the latter case the data may get lost...
6949 80. Added the %p formatting code to string_format() so that it can be used to
6950 print addresses in debug_print(). Adjusted all the address printing in the
6951 debugging in store.c to use %p rather than %d.
6953 81. There was a concern that a line of code in smtp_in.c could overflow a
6954 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6955 initially expressed, the concern was not well-founded, because trailing
6956 spaces are removed early. However, if the trailing spaces were followed by
6957 a NULL, they did not get removed, so the overflow was possible. Two fixes
6960 (a) I re-wrote the offending code in a cleaner fashion.
6961 (b) If an incoming SMTP command contains a NULL character, it is rejected
6964 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6965 runs initgroups(), so that if the Exim user is in any additional groups,
6966 they will be used during message reception.
6972 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob